Univerzita Pardubice · 2019. 10. 1. · SCIENTIFIC PAPERS OF THE UNIVERSITY OF PARDUBICE Series D Faculty of Economics and Administration No. 46 (2/2019) Vol. XXVII Registration

SCIENTIFIC PAPERS OF THE UNIVERSITY OF PARDUBICE

Series D

Faculty of Economics and Administration

No. 46 (2/2019)

Vol. XXVII

SCIENTIFIC PAPERS OF THE UNIVERSITY OF PARDUBICE

Series D

Faculty of Economics and Administration

No. 46 (2/2019)

Vol. XXVII

Registration MK ČR E 19548

ISSN 1211-555X (Print)

ISSN 1804-8048 (Online)

Contribution in the journal have been reviewed and approved by the editorial board. Contributions are not edited.

© University of Pardubice, 2019

ABOUT JOURNAL

Scientific Papers of the University of Pardubice, Series D journal aims to be an open platform for publication of innovative results of theoretical, applied and empirical research across a broad range of disciplines such as economics, management, finance, social sciences, law, computer sciences and system engineering with the intention of publishing high quality research results, primarily academics and researchers.

The journal is published every year since 1996 and papers are submitted to review. The paper is included in the List of reviewed non-impacted periodicals published in the Czech Republic, it is also indexed in Scopus, EBSCO Publishing, ProQuest and CNKI Scholar. The journal is published 3x per year.

CONTENTS

ON THE CONSEPTUAL MODEL OF OIL AND GAS BUSINESS TRANSFORMATION IN THE TRANSITIONAL CONDITIONS TO THE INDUSTRY 4.0 ALEXANDROVA TATYANA V., PRUDSKY VLADIMIR G. ...................................................................................... 5

BLACK FRIDAY SHOPPING IN HUNGARY ANDRÁS KOVÁCS, TAMÁS SIKOS T. ........................................................................................................................ 18

DESIGNING AN ADVANCED PPE MODEL TO MEASURE PERFORMANCE AND EFFICIENCY OF SLOVAK SPA ENTERPRISES ČABINOVÁ VERONIKA, ONUFEROVÁ ERIKA ........................................................................................................ 30

MODEL PRO URČENÍ PŘIROZENÉ MÍRY NEZAMĚSTNANOSTI Z MEZERY PRODUKTU DOBRYLOVSKÝ JIŘÍ .................................................................................................................................................... 44

WORKING CAPITAL AND ITS IMPACT ON BUSINESS PERFORMANCE HONKOVÁ IRENA ......................................................................................................................................................... 55

SELECTED DETERMINANTS OF TOURISM DEMAND – PANEL COINTEGRATION ANALYSIS JEŘÁBEK TOMÁŠ .......................................................................................................................................................... 67

SOLUTION OF SOCIO-ECONOMIC EFFICIENCY OF PUBLIC PROJECTS UNDER THE CONDITIONS OF VARIABLE AMOUNT OF SUBSIDY JIŘÍČEK PETR, DVOŘÁKOVÁ STANISLAVA ........................................................................................................... 79

LIFE SATISFACTION OF CZECH SENIORS: WHAT ACTUALLY MATTERS? KUBÍČKOVÁ LEA, HAJKO VLADIMÍR, HAZUCHOVÁ NAĎA, RAŠTICOVÁ MARTINA .................................. 91

ELEKTRONIZACE VEŘEJNÉ SPRÁVY JAKO NÁSTROJ OMEZOVÁNÍ KORUPCE LINHARTOVÁ VERONIKA, TVRDÍKOVÁ VERONIKA ......................................................................................... 102

CLASSICAL MANAGEMENT BIASES AND BEHAVIORAL APPORACH COMPREHENSION MANDYSOVÁ IVANA ................................................................................................................................................ 114

THE PROBLEM OF DIFFERENT PAIRWISE COMPARISON SCALES IN THE MULTIPLICATIVE AHP FRAMEWORK MAZUREK JIŘÍ ............................................................................................................................................................. 124

DETERMINANTS OF INNOVATION ACTIVITIES AND SME ABSORPTION – CASE STUDY OF GERMANY PROKOP VIKTOR, STEJSKAL JAN ........................................................................................................................... 134

STAKEHOLDER POWER, SUSTAINABILITY REPORTING, AND CORPORATE GOVERNANCE: A CASE STUDY OF MANUFACTURING INDUSTRY AT INDONESIA’S STOCK EXCHANGE SETIAWANTA YULITA, PURWANTO AGUS .......................................................................................................... 147

FACTORS AFFECTING CHOICE OF EMPLOYMENT BY UNIVERSITY GRADUATES ŠNÝDROVÁ MARKÉTA, VNOUČKOVÁ LUCIE, ŠNÝDROVÁ IVANA ............................................................... 159

DOPADY SYSTÉMU OBCHODOVÁNÍ S EMISNÍMI POVOLENKAMI V ZEMÍCH EU28 ZIMMERMANNOVÁ JARMILA, PÁSZTO VÍT, VÍCHA ONDŘEJ .......................................................................... 171

ON THE CONSEPTUAL MODEL OF OIL AND GAS BUSINESS TRANSFORMATION IN THE TRANSITIONAL

CONDITIONS TO THE INDUSTRY 4.0

Tatyana V. Alexandrova, Vladimir G. Prudsky

Abstract: Digital transformation of business centers in the transition period to the Industry 4.0 is becoming a key competitiveness growth factor for Russian oil and gas factories. It is characterized by a wide range of potential effects, which often do not show themselves fully in reality due to dominant fragmentary approach to the introduction of the Industry 4.0 innovation technologies; as well as the fact that companies’ managers do not have an integral view on the business digital transformation. The aim of the research consists in developing a conceptual model of oil and gas business digital transformation, allowing to make more reasonable managerial decisions on the implementation of digital technologies at enterprises. The research used the method of information statistical analysis about the state of the oil and gas industry and the methodology of conceptual approach to ensuring the transition of these companies to the Industry 4.0. The authors made the conclusion that the developed conceptual model improves the coordination and planning of works implementing digital transformation of oil and gas business, defines strategic priorities of the company’s digital development, promotes the development of corporate knowledge about business processes digitalization and formation of positive social and economic effect.

Keywords: Oil and Gas Business, Digital Transformation, Conceptual Model, Industry 4.0 Technologies, Potential Effects.

JEL Classification: М15, L71, O32

Introduction

The new digital technologies dissemination in the conditions of implementing the Industry 4.0 principles leads to radical changes in business process in all economic sectors. Nowadays the development of technological infrastructure and big data use have led not only to the provision of mass access to the Internet, but also integrated a wide range of digital services, products and systems into one cyber physical system (Cagnin et al., 2013; Jeschke et al, 2017).

Digital business transformation in traditional sectors, including the oil and gas one, is the key direction of Russian economy digitalization, which successful implementation can significantly increase the country’s economic role in global digital processes. To speed up the transition of traditional economic sectors to modern digital technologies and through digital processes one needs to develop a systematic policy in this sphere, taking into account specific features of separate branches digitalization. A united conceptual approach is also required to implement digital transformation in any given sector. Digital transformation needs a profound understanding of its priorities, opportunities and risk for a certain business; the problems it solves and the ways of achieving that (Rogers, 2017).

5

1 Statement of a problem

Today Russia does not have an official document reflecting the business digital transformation features in key economic sectors. The governmental program «The Russian Federation digital economy» (Legislation, 2017), accepted in 2017 and functioning to 2024, has a general character. It does not provide detailed plans on digitalizing oil and gas sector. «Russian energy strategy to 2030» (Ministry, 2015) gives a long term forecast of oil and gas industry development. But it does not shoow how digital technologies influence on achieving the forecast indicators. The measures, providing the energy strategy implementation, include only general directions of an increase in oil and gas business efficiency, such as «energy sector modernization» and «innovative technologies acqisition».

The absence of ISO international standards regulating the implementation of innovative digital technologies makes the situation, connected to the lack of a systematic conceptual approach to macroeconomic digital transformation of oil and gas industry, more complicated. So far the project of ISO standard «Information technology. Compatibility requirements and model for devices within IIoT systems» has been developed. Three more ISO standards «Artificial Intelligence» are also being worked out (Comnews, 2019). The new standards are suggested to come into force in 2020. Now companies mostly focus their digital solutions on current ISO standards of 9000-9004 series «Quality management system», 14000 series «Ecological management», and 21500 series «Project management». The digital business transformation is seen in this case not as an independent development strategy, but as a special process, focused on achieving the aims of the sphere regulated by a current ISO standard.

In the scientific aspect, the research prevail, which deal with studying separate areas of digital transformation. Especially, there are many works, describing the issues of digital fields use (Berezina, 2015; Dmitriev et al, 2016; Tcharo 2018), and summarizing the best practices of enterprises’ digitalization (Tcharo et al, 2018). There are few publications where authors approach to digital processes as complex ones, connected to green economy demands (Perelet, 2018), current ISO standards (Levchenko, 2018), and company’s strategic development (Bauer et al, 2018 At the moment there are no publications, containing a developed model of business digital transformation in relation to specific economic conditions of oil and gas business in Russia.

The literature review showed that currently there is no unified notion of «Digital business transformation». Different authors interpret it differently trying to link it to the Industry 4.0 principles (Berman, 2012; Garifullin, 2018; Gray & Rumpe, 2017; Mat et al, 2015: Nylén & Holmström, 2015). Besides, international organizations, focused on doing research in the digital economy sphere, greatly contribute to studying this issue (World Bank, 2016; UNO, 2018; OEСD, 2017; EEC, 2017). In our opinion, the definition, given by OEСD, suits oil and gas business most of all: «Digital business transformation is a process integrating digital technologies in all aspects of business activity, and requiring the transformation of business operation models, managerial paradigms, economic relations and social practices» (OEСD, 2017).

As far as Russian digital business transformation practice is concerned, in most cases enterprises introduce only separate digital projects, without developing the

6

digitalization strategy for all business lines. At the same time, the leading oil companies have generated some experience in the sphere of complex digital business transformation. For example, the oil company «Transneft» actively introduces digital solutions on pipelines. The oil company «Gazprom Neft» started the first Russian open digital platform EvOil and began the construction of the first digital refinery. The «LUKOIL» group of companies is implementing the technologies of integrated production modeling and planning. The companies «Gazprom Neft», «Tatneft», «Rosneft», and «LUKOIL» are the leaders in the number of digital oil fields. But even the leaders need to bolster their efforts to achieve a better integration of digital solutions with a general strategy of the company’s development.

The given problems are extremely actual for Russian oil and gas companies, whose share in the country’s GDP reaches about 70%, in the export structure about 50%, and the in federal budget formation up to 40% (VYGON consulting, 2018). Their scientifically based solution could speed up the digital transformation process of the oil and gas industry, as well as Russian national economy in general. So far Russia has taken the 42nd place out of 63 countries in the IMD digital competitivness rating (VYGON consulting, 2018).

The above listed issues determine the aim and tasks of the research. The research purpose is to develop a conceptual model of Russian oil and gas sector transformation in the transition to the Industry 4.0. To achieve the aim, the following tasks had to be solved:

To formulate strategic challengers for Russian oil and gas industry, which predetermine the necessity of oil and gas business digital transformation;

To structure the conceptual model of oil and gas business digital transformation in the transitional conditions to the Industry 4.0.

To determine the conditions of the developed model’s successful implementation into managerial practice as well as its influence on oil and gas company’s activity indicators.

2 Methods

The research used the methodology of conceptual approach to stating and solving a scientific problem, as well as the statistics analysis method of the information on oil and gas companies’ activity.

Table 1 (Tab. 1) presents the statistic indicators, characterizing the efficiency of Russian oil and gaz companies development. The sample includes data from 2013, when the first information collection started to make an annual public rating of Russian oil and gas companies’ ecological responsibility. Before 2013, Russian official statistics did not single out oil and gas industry as a separate one in extracting industries of the economy; and these companies did not provide nonfinancial accounts on their ecological activity. The data demonstrate a number of negative trends in their activity, such as low growth rates of labour efficiency indicators; the increase in fixed assets depretiation; the decrease in returns on assets accompanied by the rise in capital-labour ratio; a low share of highly technological working places.

7

Tab. 1: Russian oil and gas industry efficiency development indicators Indicators 2013 2014 2015 2016 2017

1. Oil conversion ratio, % 71,6 72,4 74,3 79,0 81,0 2. Labour productivity index 100,8 102,8 98,3 100,3 100,4 3. Capital-labour ratio change index 103,7 106,6 106,4 104,7 104,5 4. Returns on assets change index 91,1 96,6 94,9 94,5 93,45 5. Fixed assets depreciation, % 53,2 55,8 55,4 57,5 57,7 6. A share of highly technological

working places, % 5,0 4,7 5,1 5,3 5,8

7. Specific emissions of pollutants into the atmosphere kg/ tfoe

3,82 3,16 2,09 1,92 2,48

8. Specific gross greenhouse emissions kg/ tfoe

No data No data 41,18 71,81 87,68

9. The ratio of polluted land area at the end of the year to the beginning of the year ha/ ha

0,39 0,18 0,17 0,19 0,31

10. The level of oil associated gas utilization, %

78,22 84,88 85,90 88,21 86,7

* Note:accountability on greenhouse gases emission was introduced only in 2015

Source: made based on data (FSSS, 2018; WWF.2018)

Changes that are even more negative can be seen in the dynamics of oil and gas companies’ impact on the environment. For example, for the period from 2015 to 2017 specific emissions of pollutants into the atmosphere increased by 18%; specific gross greenhouse emissions rose more than twice; the land pollution index grew by 1.8 times. This situation can be partly explained by the introduction of sectoral sanctions in 2014 – 2015, as the result, oil and gas industry forfeited the benefits of buying foreign innovative technologies in the reqiured quantity. The decrease in environmental pollution level should be considered one of the priorities of Russian oil and gas sector digital transformation.

Tab. 2: The profitability indicators of the oil and gas industry leaders for 2013-2017 Indicators 2013 2014 2015 2016 2017

Sales profitability, % LUKOIL Rosneft Gazprom

82,987 5,880 24,459

83,540 3,626 23,061

82,165 3,049 18,733

85,548 9,782 8,455

70,605 7,357 8,706

Assets profitability, % LUKOIL Rosneft Gazprom

16,828 5,628 6,017

24,371 7,855 1,636

16,000 2,778 3,199

9,197 1,023 3,066

9,197 1,321 0,710

The average, % Sales profitability Assets profitability

37,77 9,491

36,742 11,287

34,649 7,326

34,595 4,429

28,889 3,743

Sourse: made based on data (LUKOIL, 2018; Rosneft, 2018; Gazprom, 2018)

The problem of decreasing operating costs has become actual for oil and gas companies in the conditions of world oil price decrease. The data in the table 2 (Tab.2) allow to conclude that the profitability of sales and assets of Russian oil and gas companies is constantly falling. One more problem of Russian oil and gas

8

industry development consists in the growth of costs for current land wellsites repair and reconstruction, as well as high capital costs for offshore oil field development, mainly in Arctic.

Fig. 1: Strategic challenges for Russian oil and gas sector

Source: made by the authors

The cost of offshore works can exceed hundreds millions dollars. The question inevitably arises about the purposefulness of investments in such expensive projects, because such «expensive oil» cannot compete with traditional cheap oil (VYGON consulting, 2018). The use of digital technologies can reduce operating costs; extend the brown fields’ lifecycle; reduce risks and cost of offshore drilling considerably; increase the depth of oil extraction of the volume of oil production.

The strategic challenges for Russian oil and gas sector are presented in the figure 1 (Fig. 1).

The given data prove the necessety of systematic modernization for Russian oil and gas sector based on new Industry 4.0 digital technologies, which allow to solve business development problems quicker, more economical, with less risks and better results. It is forecasted that due to the implementation of the Industry 4.0 technologies the potential oil deposit extraction will reach 6.8 bln tonns to 2035; the oil production volume will reach 607 mln tonns; the depth of oil extraction will increase to 90%; the operating costs will decrease by 2.2 tln. roubles; the production cost of oil and gas extraction and refining will decrease by 15% (VYGON consulting, 2018). This will allow Russian oil and gas companies compensate for fall in income due to the constant decrease in world oil prices.

Strategic challenges for

Russian oil and gas sector in

2019

High level of fixed assets depreciation Low speed of labour

productivity growth

Decrease in the activity

profitability

Late stage of wellsite

development

World oil price decrease, sectoral

sanctions

High cost of offshore oil field

development

High level of environmental

pollution

Low level of innovations

implementation

9

3 Problem solving

The development process of Russian oil and gas industry is regulated by different legislative documents, which lack the concept of the sector’s digital transformation. An objective reason for that constitutes in the complexity and variety of business processes in oil and gas industry, which leads to a more labour intensive process to develop a strategic vision of digital modernization of oil and gas companies’ activity.

The authors believe, that the specific features determining the necessity of conceptual approach to oil and gas business digital transformation include the following ones:

The technological process is complex and resource intensive, which functions in a close connection with the development of energy saving programs, the development of energy efficiency, the use of low waste and wasteless as well as resource efficient technologies.

The oil and gas business is based on fragmented combination of different spheres. That is why different digital directions of the 4.0 Industry are often united here in one technological solution. For example, a digital wellsite includes the elements of the commercial Internet of things, big data, robotics and others.

Ecological requirements to the oil and gas business activities in the implementation of green economy concept (CSD, 2018) predetermine the priority focus of digital transformation process in oil and gas business on implementing tasks that are connected to providing industrial security and environmental protection.

The leading oil and gas companies aim at achieving the leadership position in the sphere of oil and gas extraction and refining. They include such oil business giants as Petroleum, Conoco Philips, Exxon Mobil, Royal Dutch Shell, Total,Gazpromneft, LUKOIL and other big corporations (Perfiljev & Solomatov, 2016). Moreover, exporting companies get involved into the processes of forming intercountries’ digital space (EU, EAEU, BRICS and others), and participate in global digital processes.

The biggest oil and gas companies are characterized by the trend to diversification of their activities. One should also note a quite complex product supply chain and a large number of business partners from other industries. In these conditions, the intensive development of digital processes in oil and gas sector contributes to a more efficient cross-sector digital transformation.

The figure 2 (Fig. 2) demonstrates the digital environment of Russian oil and gas sector, modeled taking into account all given above specific functioning features of oil and gas companies.

10

Fig. 2. The model of digital environment in oil and gas business


The methodological approach, developed by the research group of the World Bank and Eurasian Economic Commission (EEC, 2017) constituted the basis to form a conceptual model of oil and gas business digital transformation. This methodology recommends a complex approach to digitalization process, taking into account possibilities and risks of the Industry 4.0 technologies, qualitative and quantitative aspects of digitalization, measures to provide business digital transformation. In the research process it was specified relating to peculiarities of oil and gas companies development.

The authors’ model is presented in figure 3 (Fig. 3). It includes 6 blocks, each having its own structure and performing a certain function:

Block 1 «Priority purposes». It performs the goal setting functions.

Block 2 «Target indicators». It gives details of priority purposes in the form of quantity indicators, which can be measured and estimated.

Block 3 «Digital transformation directions». It reflects the key trends of using digital technologies in oil and gas business.

Block 4 «Possible effects». It reflects potential benefits from implementing digital transformation in oil and gas business.

Block 5 «Possible risks». It informs about potential risks, connected to ignoring digital transformation of business.

Block 6 «Implementation mechanisms». It describes the necessary measures to regulate the digital transformation process in oil and gas business.

Key digital technologies Big data Commercial Internet of things Robots Digital twins Smart materials 3D-printing Block chain

Global digital environment

Cross-sector digital environment

Digital environment of international state unions

Digital environment of perspective pilot projects

Digital refineries

Digital infrastructure

Intellectual oil fields

Digital pipelines

11

Fig.3 The conceptual model of Russian oil and gas business digital transformation


Priority purposes

A sustainable development of oil and gas business in Industry 4.0 Digital environment formation in oil and gas sector Environmental protection and restoration

Increase in oil extraction coefficient Decrease in operating and capital costs Decrease in pollutants emission Increase in highly technological work places Increase in the energy efficiency of production

Target indicators

Directions Digital twins creation

Labour automation

Providing the staff with

digital devices

Creation of digital ecosystem

Increase in efficiency of oil and gas

extraction and refining

Optimization of the number

of staff, development

of digital labour

Improvement of working

conditions, the increase in production safety rate

Increase in transactions

speed, improvement of

colaboration with third parties

Increase in quality and speed of decision making Increase in labour productivity and quality Optimization and cutting of costs

Possible risks Missed profit risk Risk of digital technologies late implementation Risk of decrease in competitive ability Risk of increase in digital space control from leading market players The risk of resistance to digital innovations

Implementation mechanisms

Support of digital initiatives and pilot projects

Working out of new generation standards

Stimulating investments in digital technologies

Creating sectorial centers of competence and techno parks

Promoting best practices of digital business transformation

Possible effects

12

The main research results can be used by oil and gas companies to organize the digital business transformation process; by scientific-research and expert organizations to work out digital programs and projects; by sectorial ministries to work out strategies of industry development; and by governmental bodies to develop state innovation policy for both country and regions.

4 Discussion

In 2018 the conceptual model of digital oil and gas business transformation, developed by the authors, was tested in the management practice of OOO «LUKOIL Perm» in the part concerning the measures of planning digital technologies implementation in the spheres of industrial security and environmental protection. The company conducts these activities based on ISO standard 14001. The model approbation results are presented in Table 3 (Tab. 3).

The presence of general concept of oil and gas business transformation in the conditions of 14001 ISO standard allowed of OOO «LUKOIL Perm» to implement a complex approach to developing and introducing digital solutions for providing industrial security and environmental protection. The actions taken have led to positive effects, for example, in 2018 OOO «LUKOIL Perm» decreased the water intake from surface watercourses by 7%, atmosphere pollution dropped by 14%, the number of new wellsites rose by 30%, operating costs fell by 10%, and energy consumption by 15%.

Tab. 3: The approbation results of oil and gas business digital transformation model at OOO «LUKOIL Perm»

Conditions necessary for model implementation

Measures taken by OOO «LUKOIL Perm»

Measures efficiency prof*

Orientation on digital standards demand regulating the development of «smart» production, when they are absent, the choice of business priorities, which aims will make a basis for model functioning.

The model is focused on maintaining a high level of production security and environmental protection in compliance with 14001 ISO standard.

Economic efficiency: Decrease of operating

costs by 10% Drop of energy

consumption by 15% Increase in the

number of new wellsites by 30%

Ecological efficiency: Decrease of fresh

water intake from watercourses by 7%

Fall in atmosphere pollution by 14%

Social efficiency: Development of the

workers’ digital thinking

Increasing staff’s adaptability to digital technologies introduction

Top priority introduction of Industry 4.0 innovation technologies into production business processes, which provide a decisive contribution into forming positive effects of digitalization.

Introduction and gradual development of three intellectual oil fields. A preparatory work of mastering digital technologies of off-shore oil drilling.

Use of automated management methods in making strategic and tactical decisions.

Using methods of preventive diagnosis of wellsites states; monitoring of productivity loses; oil extraction indicators analysis in real time.

The use of model in the company’s strategic planning.

The model is used at the initial stage of strategic planning as a set of key guidelines for developing programs, plans and projects.

13

Creating managerial departments to coordinate different business processes.

Three Centers of integrated operations are created within the company’s management.

Public efficiency: The visit of «Ernst & Young» experts to study the digitalization experience at OOO «LUKOIL Perm».

The development of staff’s digital competencies, necessary for comprehension and implementation of the model; providing the implementation of developed measures.

Working out further training programs in the digital sphere; introducing personal responsibility of workers for fulfillment of the company’s digitalization plan.

* Note: the indicators are calculated as the ratio of 2018 data to 2017 ones. Source: created by the authors

However, some difficulties were observed alongside with the possitive effects from implementing this model. They were connected to the fact, that the digital business transformation procedure has become more complex, varied and labour intensive. Managerial and production staff needed a higher qualification level and rediness to work in new conditions. A specialized training in digital technologies was organized at OOO «LUKOIL Perm» to solve the problem of knowledge deficiency and unstable performance discipline. The company also introduced personal responsibility of workers for compliance with the requirments of 14001 ISO standard and performance of plan actions connected to development and implementation of digital solutions in the sphere of industrial security and environmental protection.

Conclusions

The conducted research allows to make the following conclusions:

The worked out conceptual model contributes to the development of corporate knowledge about managing digital processes and technologies. It forms a holistic vision of oil and gas business digital transformation perspectives in the transition to Industry 4.0, increases the level of coordination between departments and services of managerial apparatus in digitalization processes;

Methodological approach, used at model working out, develops the methodology of making the digitalization concept of the World Bank and Eurasian Economic Commission as it takes into account industrial peculiarities of business digital transformation using the example of Russian oil and gas companies;

The developed conceptual model of oil and gas business digital transformation satisfies managerial demands in justification the choice of companies’ strategic development options. Compliance with parameters, embedded in separate blocks of the model, can serve as a datum mark for including corporate measures and projects in strategic programs and development plans of oil and gas companies;

To implement the developed model into managerial practice it is necessary to comply with conditions, which guarantee a successful integration of an innovation into oil and gas companies‘ activities. Positive effects are noticed as the result of model approbation at OOO «LUKOIL Perm». They prove a positive influence of this approach on the oil and gas business digital transformation process and an increase in the company’s efficiency;

14

The use of conceptual approach to business digital transformation does not only contributes to improving productive and managerial activity of an oil and gas company, but also creates problems, connected to using new knowledge and skills. Companies’ managers must quickly identify these problems and provide measures to develop staff’s competence and performance discipline.

The conducted research contributes to the theory of modern industry-based management. If this model is applied, it will allow to formalize the process of making managerial decisions in the sphere of business digital transfornmation; it will help to determine the features of digital technologies implementation for separate oil and gas companies more reasonably. The use of this model will also contribute to the growth of positive influence of oil and gas sector on acjieving target indicators of the governmental program «The digital economy of the Russian Federation», which will provide a facter and a more successful transition of the country’s economy to the Industry 4.0.

The authors see further research direction in working out the adoptation strategy of Russian oil and gas companies to the principles of the Industry 4.0.

Acknowledgement

The research is conducted with the financial support of the Russian Foundation for Basic Research (RFBR) in the frames of the scientific project №18-010-00955 «The development of models to manage cyberphysical organizations».

References

Bauer, V. P., Podvoisky, G. L. & Kotova N. E. (2018). Adaptation strategies of the U.S. companies to the digitalization of production. World of the new economy, 12(2), pp. 78-89. http:// DOI: 10.26794/2220-6469-2018-12-2-78-89

Berezina A.A. (2015). Economic and managerial criteria for choosing projects for implementing intellectual field technology. Internet-journal «Naukovedenie», 7(1), рр. 1-8. http://DOI: 10.15862/20EVN115

Berman, S.J. (2012). Digital transformation: Opportunities to create new business models. Strategy and Lea dership, 40(2), pp. 16–24. https://doi.org/10.1108/10878571211209314

Bukht R. & Heeks R. (2018). Defining, Conceptualising and Measuring the Digital Economy. International Organizations Research Journal, 13( 20), pp. 143–172. http://DOI: 10.17323/1996-7845-2018-02-07.

Commission of Sustainable Development (CSD) UNO. (2018). The 2030 Agenda for Sustainable Development [online]. Geneva: Commission of Sustainable Development. Available at: <https://sustainabledevelopment.un.org/post2015/transformingourworld> [Accessed 21.01.2019].

Comnews (2019). Russia standardizes the world IIoT. [online]. Comnews. Available at: https://www.comnews.ru/content/115580/2018-10-30/rossiya-standartiziruet-mirovoy-iiot. [Accessed 21.02.2019].

Dmitrievsky A.N., Martynov V.G., Abukova L.A., Eremin N.A. Digitalization and intellectualization of oil and gas fields. Automation and IT in the oil and gas field. 2(24), рр. 13-19. [online]. Available at: http://www.ipng.ru/files/_72af087e-fd28-4c55-8e7c-9190a9d4abdc-DmitrievskiyAN_Neftegaz_20 16_2.pdf. [Accessed 11.02.2019].

Eurasian Economic Commission (EEC), 2017. The joint research of the World bank and Eurasian Economic Commission: Digital Agenda of EAEU 2015: perspectives and recommendations. [online].

15

Minsk: EEC. Аvailable at: http://documents.vsemirnyjbank.org/curated/ru/413921522436739705/pdf/EAEU-Overview-Full-RUS-Final.pdf. [Accessed 20. 02. 2019].

Federal State Statistics Service (FSSS). (2018). Official statistics. Russian economy efficiency. Moscow: FSSS. [online]. Аvailable at: http://www.gks.ru/wps/wcm/connect/rosstat_main/rosstat/ru/statistics/efficiency/. [Accessed 25. 02. 2019].

Cagnin, C., Havas, A., & Saritas O. (2013). Future-oriented technology analysis: Its potential to address disruptive transformations. Technological Forecasting and Social Change, Vol. 80, pp. 379–385. Available at: http://publications.jrc.ec.europa.eu/repository/handle/JRC 77930. [accessed 18.01.2019].

Garifullin B.M. & Zyabrikov V.V. (2018). Digital transformation of business: models and algorithms. Creative economy, 12(9), рр. 1345-1358. http:// doi: 10.18334/ce.12.9.39332

Gazprom. (2018). Information disclosure: accounting. [online]. Moscow: Gazprom. Аvailable at: http://www.gazprom.ru/investors/disclosure/. [Accessed 10. 02. 2019].

Gray, J. & Rumpe, B. (2017). Models for the Digital Transformation. Soft ware & Systems Modelin, 16(2), pp.. 307–308. https://doi.org/10.1007/s10270-017-0596-7 Rogers D. L. (2017). Digital transformation (translated from English). Moscow: Publishing Group «Tochka».

Jeschke, S., Brecher, C., Song, H. & Rawat, D.B. (2017). Industrial Internet of Things: Cybermanufacturing Systems. Switzerland: Springer International Publishing

Legislation of the Russian Federation (2017). The program «Digital economy of the Russian Federation». [online]. Legislation of the Russian Federation. Available at: http://rulaws.ru/goverment/Rasporyazhenie-Pravitelstva-RF-ot-28.07.2017-N-1632-r//. [Accessed 24.01.2019].

Levchenko, E.V. (2018). The impact of digitalization on the development of quality management system. Vestnik of Saratov State Socio-Economic University” journalа, 4(73), рр. 9-14. [online] https://elibrary.ru/download/elibrary_36413118_87280251.pdf

LUKOIL. (2018). Financial results. [online]. Moscow: LUKOIL. Аvailable at: http://www.lukoil.ru/InvestorAndShareholderCenter/ReportsAndPresentations/FinancialReports. [Accessed 10. 02. 2019].

Mat, t C., Hess, T. & Benlian A. (2015). Digital Transformation Strategies. Business and Information Systems Engineering, 57(5), рр. 339–343. http:// DOI: 10.1007 / s12599-015-0401-5

Ministry of energy of the Russian Federation (2015). Russian energy strategy for the period to 2030. [online]. Ministry of energy of the Russian Federation. Available at: https://minenergo.gov.ru/node/1026. [Accessed 24.01.2019].

Nylén, D & Holmström, J. (2015). Digital innovation strategy: A framework for diagnosing and improving digital product and service innovation. Business Horizons, 58(10), pp. 57-67. https://doi.org/10.1016/j.bushor.2014.09.001

OECD (2017). OECD Digital Economy Outlook 2017. OECD Publishing: Paris. http://dx.doi.org/10.1787/9789264276284-en

Perelet, R. A. (2018). Ecological aspects of digital economy. The world of new economy, 12(4), рр. 39-45. http:// DOI: 10.26794/2220-6469- 2018-12-4-39-45

Perfiljev, S. V. & Solomatov, A. P., Key aspects of consolidation in the world oil refining industry at the current stage, 2016. Internet-journal «Naukovedenie», 8(4). Available at: https://cyberleninka.ru/article/v/klyuchevye-aspekty-konsolidatsii-v-mirovoy-neftepererabatyvayuschey-promyshlennosti-na-sovremennom-etape [Accessed 15 января 2019]

Rosneft. (2018). Financial statment. [online]. Moscow: Rosneft. Аvailable at: https://www.rosneft.ru/Investors/statements_and_presentations/Statements/. [Accessed 10. 02. 2019].

16

Tcharo, H,, Vorobev A.E. & Vorobyev K.A. (2018). Digitalization of the oil industry: basic approaches and rationale for "intelligent" technologies. The Eurasian Scientific Journal, 2(10), рр. 1-17. [online] Available at: https://esj.today/PDF/88NZVN218.pdf. [Accessed 15.02.2019].

UNO, 2018. The initiative in the sphere of data for development purposes «Global pulse». [online]. Geneva: UNO. Available at: http://www.un.org/en/sections/issues-depth/big-data-sustainable-development/index.html [Accessed 20. 02. 2019].

Vorobev A.E. & Tcharo, H, (2018). Digitalization of the oil industry of Kazakhstan. The problems of natural resources management, 1, рр. 66-75. http:// DOI: 10.25635/2313-1586.2018.01.066

VYGON consulting (2018). Digital oil extraction: tuning for the industry. [online]. VYGON consulting. Available at: http://vygon.consulting/products/issue-1322/. [Accessed 21.02.2019].

World Bank (2016). World Development Report 2016: Digital Dividends. Washington, DC: World Bank. http://doi:10.1596/978-1-4648-0671-1

World Wildlife Foundation (WWF) in Russia (2018). Openness ratings of Russian oil and gas companies in the ecological responsibility sphere for the period 2014 – 2018. [online]. Moscow:

World Wildlife Foundation, Booklets, materials and documents. Available at: https://wwf.ru/resources/publications/booklets/ [Accessed 22. 01. 2019].

Contact Address

Tatyana V. Alexandrova, Candidate of Economics, associate Professor Perm State National Research University, Faculty of Economics, Department of management Bukireva Street, 15, 614990, Perm, Russia Email: [email protected] Phone number: +79058607708

Vladimir G. Prudsky, Doctor of Economics, Professor Perm State National Research University, Faculty of Economics, Department of management Bukireva Street, 15, 614990, Perm, Russia Email: [email protected] Phone number: +79127824530

Received: 18. 03. 2019, reviewed: 17. 06. 2019 Approved for publication: 30. 08. 2019

17

BLACK FRIDAY SHOPPING IN HUNGARY

Kovács András, Sikos T. Tamás

Abstract: The central question of our study is to examine what features the Black Friday phenomenon has in Hungary, how it resembles to or differs from the international Black Friday trends. To answer this question we provide an overview of the main ideas on the topic discussed in international literature. Then with the help of answers of a questionnaire research from 2016 and one from 2017 we will highlight the most important similarities and differences. As we will see, in Hungary the buyers’ intent to participate is similar to other countries, but retailer behavior shows considerable differences compared to international examples. Descriptive methods, cross-tabulation, cluster analysis applied in our research.

Keywords: Black Friday, Retail, Customer Behavior

JEL Classification: M31

Introduction

By now, Black Friday has not only become the day of biggest turnover for retailers in the United States, but it is also the real start to the "Christmas shopping fever", a serious social event, a hedonistic feast of consumption for the U.S. consumer society as Bell and her colleagues point out (Bell et al., 2014).

Since due to its importance and size Black Friday is now far from having simply business or marketing significance, it also raises serious social, psychological, decision-making and security issues. In the international (mainly US and UK) literature, math-based decision theory study (Wu, Zitikis, 2017) and an analysis of security issues related to Black Friday can also be found (Smith, Raymen, 2015).

Obviously, most researches related to Black Friday focus mainly on the aspects of consumer behavior. Consumer/shopper behavior researches primarily focus on the question of purchase/non-purchase and purchasing motivations (Kwon, Brinthaupt, 2015) (Delcea, Ioanas, Paun, 2017), examine the relationship between Black Friday and its dynamically growing online counterpart, Cyber Monday (Swilley, Goldsmith 2013), and reveal the buyers’ shop-preferences and shopping behavior (Simpson et al., 2011). In addition to customer and shopper behavioral researches, commercial logistic studies (McLeod et al., 2016) and social responsibility analyzes (Lennon et al., 2014) can be found in international literature.

We approach the well-known “Black Friday phenomenon” from a marketing point-of-view. After providing an international literature overview, we will highlight the special features and characteristics of Black Friday in Hungary. Relying on the results of online questionnaires, shopping behaviour (product and shop preferences, prices) and shopper-groups will be set up and described.

Our aim is to contribute to a better understanding of this important retail and marketing phenomenon by unveiling shopper behaviour, and describing and grouping different customers.

18


Hereinafter, we attempt to summarize all the relevant aspects of Black Friday, as it would be a rough simplification to examine Black Friday sales from a single specific marketing sub-perspective (Törőcsik, 2017) (Rekettye, 2003) (e.g. pricing, shopping behavior, or marketing communication aspect) (Tab. 1).

Tab. 1. Black Friday’s marketing relevancies Marketing sub-area

concerned Connection point to Black

Friday Significance

Shopper, consumer behavior

revealing and understanding shopper motivation

It can effectively contribute to reaching/retaining target groups

Marketing strategy -planning

Inserting Black Friday sales into company marketing strategy

Might influence the implementation of marketing strategy goals

Product portfolio management

inventory management, “swap outs”

contributes to creating optimal product portfolio

Pricing, pricing strategy

Discounts considering company, competitor, buyer points of view

It contributes to maximizing profit through its price tactics and price-strategic role

Marketing communications

Communicating Black Friday and company involvement

It supports the growth of Black Friday's reputation and the achievement of corporate sales goals

CSR – corporate social responsibility

hedonistic vs. conscious shopping

Responsible buyer-incentives to increase long-term loyalty

Distribution appropriate handling of extremely large number of orders

Prerequisite for customer satisfaction (current stock information, on-time delivery)

Source: own edition

The most important lessons learned from Tab. 1 are that Black Friday sales (in optimal cases) are not just a communication campaign with intensive price promotions aimed at boosting short-term pre-Christmas turnover, but are also affecting the marketing strategies and activities of the company for long-term business success in a rather complex way (Józsa, 2014). In order for the Black Friday sales to achieve the desired favorable effect, it is very important to consider logistical, CSR, product management and strategic aspects besides communication and pricing tasks.

Our above review highlights that even a sales-promotion activity may have many additional marketing implications, so it is particularly important to explore and analyze these connecting points.

2 Methods

Since previous research in Hungary did not deal with the "Black Friday phenomenon", we consider it important to reveal the most important features of consumer and shopper behavior related to it.

19

In the next part of our study we analyze the results of the questionnaire research (carried out between November 25-27, 2016 and November 24-27, 2017) on the Black Friday phenomenon, with the aim of exploring the most important features of the Hungarian consumer behavior.

As outlined in Tab. 1, we have already highlighted that several marketing dimensions of the Black Friday phenomenon could be investigated. Naturally, all aspects of Black Friday cannot be investigated within a framework of a single research because the different marketing areas, the company (supply) and the buyer (demand) side can only be explored with different methodologies and tools.

In our research, from the areas enumerated in Tab. 1 we focus only on the most important features of shopper behavior related to Black Friday in Hungary, and we do not deal with other marketing areas, such as Black Friday's marketing strategy and its specific sub-areas.

Our aim is to reveal awareness connected to Black Friday, purchase/non-purchase relations and the key determinants of purchases, including spending, shop and product preferences, as well as opinions related to Black Friday.

We have chosen the online questionnaire method for the research, although several other methods are available today, as pointed out by Kemény, Simon 2015. The questionnaire seemed to be the most suitable method for conducting the research for two reasons: on the one hand, as Babbie points out, this method is capable of executing a large number of direct observations (Babbie, 2008). On the other hand, the aim of our research is to describe and analyze the opinions and behavioral characteristics of consumers on Black Friday, and to this the questionnaire method provides a good basis (Ghauri, Grønhaug, 2011). The questionnaire method was supported by the fact that in the international literature a similar approach was used to explore the Black Friday shopping habits (Tsiotsou, 2017). We preferred the method of online questionnaire (Google Forms) because this made it possible to reach a large number of potential respondents in a short time, since due to the nature of Black Friday (it takes place once a year), it is of utmost importance to reach customers and get the most accurate information. In the period 27-30 November 2016 a total of 272 respondents filled in our questionnaire and in November 2017 a total of 96 respondents provided answers. The results (after data cleansing and re-encoding) were analyzed using IBM SPSS Statistics 20 and Microsoft Excel 2017 with descriptive statistical methods. Descriptive statistical methods, cross-tabulation and cluster analysis (Ward-method) were applied.

In our research we aimed to answer the following research questions:

• What is the extent of awareness of Black Friday in international and in Hungarian dimension?

• How can Black Friday customers be characterized by the terms of spending, product and shop preference?

• What is Black Friday's satisfaction like among buyers and among non-buyers?

• How did the respondents' opinions change between 2016 and 2017, given the fact that the special features of the Hungarian Black Friday campaign have also changed considerably.

20

• What are the “Black Friday shopper groups” formed with statistical methods (cluster analysis)?

In the course of the research, single- and two-pole differential scales were used to ensure that online response is fast and smooth, whatever platform respondent (computer, tablet, smartphone) used. Hereinafter, after the presentation of the sample, we will deal with each research question in separate chapters.

In our research, from a methodological point of view, the biggest challenge was how to measure the awareness of the Black Friday phenomenon and the shoppers’ behavior in the relatively short duration (a few days) of the event. The advantage of online questionnaires is that they are very quickly "applicable" and a large number of potential respondents can be reached, but the problem is that the resulting sample is not representative. Generally, as it is in our case, among the respondents young urban residents with higher education, and above the average digital literacy are overrepresented. (Tab. 2).

Tab. 2: Descriptives of the samples 2016 and 2017 2016 2017

Sex, percent male 35,3 42,7

female 64,7 57,3

Age, percent

under 20 years 9,9 0

21-30 years 41,5 33,3

31-40 years 21,7 30,2

41-50 years 14,7 13,5

51-60 years 7,7 7,4

over 60 years 4,4 15,6

Highest school degree, percent

Elementary school 4,4 1,1

Vocational school 1,5 5,2

Secondary school 20,2 12,5

BSc, BA degree 33,1 30,2

MSc, MA degree 40,8 51,0

Place of residence, percent

capital (Budapest) 33,2 38,5

county seats 30,6 22,9

other towns 23,8 24,0

villages 12,5 14,6

Source: own research

The demographic structure of the two examined samples represents the social groups that are important target groups for the Black Friday campaign: urban, highly educated, groups, relevant for advertising.

Although our sample does not represent all layers and groups of the Hungarian society, we aimed to reach people living in different parts of Hungary. Altogether we surveyed our responders in 83 different settlements and in all districts of the capital, Budapest.

21

3 Problem solving

In connection with the Black Friday phenomenon, we first investigated what sources the respondents use to obtain information. As Fig. 1 shows, Hungarian customers primarily use online sources to get informed. One possible reason for this is that – opposed to the international trends – Cyber Monday, specialized in online shopping, is not wide-spread in Hungary, so online retailers and companies using multi-channel distribution systems all focus on Black Friday. While in the US, Cyber Monday generates more revenue for e-tailers than Black Friday (Richter, 2015) (McGee, 2017), its role in Hungary – for the time being – is negligible.

Fig. 1: Information sources of Black Friday, 2016


It is also a special Hungarian phenomenon that the "official" day of Black Friday is not considered "binding" for commercial companies. This phenomenon was particularly strengthened in 2017 when, in addition to the "official" Black Friday Day of November 24, several major domestic retailers announced an “alternative Black Friday" day, or even a week. (The trend can be observed in other countries (Telegraph Reporters, 2017), but it has become rather "extreme" in Hungary.)

In our research we asked the respondents about the dates of Black Friday sales for the four largest domestic retailers dealing in home appliances.

A smaller proportion (20-40 percent) of respondents were able to the Black Friday day (Tab. 3) of the particular retailer select from the predetermined list. The best result was achieved by eMAG online retailer, who held its own sales on the "official" Black Friday Day on the 24th of November.

Tab 3: Dates of Black Friday deals by the largest electronic retailers and their reputation among responders, 2017

Company Date(s) of Black Friday deals Ratio of right answers, percent

Media Markt 17th and 24th of November 19,8

Euronics Between 20th-26th of November 30,2

E-Digital 17th of November 18,8

eMAG 24th of November 44,8

Source: own data collection from the websites of the selected retailers

It is also a Hungarian phenomenon that the expression "Black Friday" is merely a synonym for a significant price discount, and has no strong historical, socio-cultural

22

background like it does in the USA (Bell et al., 2014). As a result, in 2017 and 2018, "Winter Black Friday" and "Spring Black Friday" are advertised by major touristic and retail companies (Auchan, szallasvadasz.hu, etc.) If this trend continues, we are afraid, Black Friday will become only a synonym for a (significant?) price reduction in Hungary and will lose (or fail to get) the prestige value it has in the US and other Western European countries.

3.1 Product preferences, spending, distribution channels

According to our research, 37 percent of the respondents did some shopping on Black Friday in 2016. This rate rose to 44 percent in 2017, which is roughly equivalent to the rate of participation observed in countries of similar development. Tsiotsou also found a 1/3 turnout in Greece (Tsiotsou, 2017), but in developed countries this rate is much higher. In the US Kwon and Brinthaupt identified 63 percent customer involvement (Kwon, Brinthaupt, 2015), but the research conducted by Pricewaterhose Coopers in 2017 showed “only” 35 percent participation rate. At the same the significance of Cyber Monday increased considerably (PricewaterhouseCoopers, 2017).

The product preference of Black Friday buyers is in line with international trends. Most of the surveyed respondents purchased technical items (50 percent) and clothing products (30 percent) during the period of sales. However, our research reveals that none of the retailers wants to miss out on Black Friday, so more food retailers, furniture shops, and book and toy stores have announced a Black Friday sale.

The amount spent and type of product purchased during the Black Friday campaign is well illustrated by spendings (Fig. 2). In Hungary willingness to spend is naturally limited by the amount of the available disposable income (Medgyesi, 2016), so spending per capita is far behind the values of the more developed countries. While the modal value of Hungarian spending allows for the purchase of a small household appliance, an IT accessory or a few fast-fashion garments (between EUR 30 and EUR 80), in the US the per capita Black Friday spending is 100-500 USD (Kwon, Brinthaupt, 2015).

Fig. 2: Spending on Black Friday, 2016, 2017


As mentioned earlier, the fact that Cyber Monday has not (yet) been established in Hungary has a significant role in the preference of retail outlet, so online retailers concentrate almost exclusively on Black Friday. This is due to the fact that the popularity of webstores is the greatest (Fig. 3), which is in line with international trends (Swilley, Goldsmith, 2013) (PricewaterhouseCoopers, 2017).

23

Fig. 3: Preferred shop types on Black Friday, 2016


3.2 Shopping satisfaction

In the last part of our research, we investigated customer satisfaction connected to Black Friday sales on a four-stage, two-pole differential scale where we offered the possibility to answer "I do not know" because we asked the question from all respondents, not only from those who actually did buy something. (We asked all respondents because we have also investigated factors (prices, marketing communication, etc.) which are not directly related to the purchase.).

In the following, we examined satisfaction with product selection, prices and advertising (marketing communication) based on the respondents' answers. Generally speaking, both in 2016 and 2017, 30-40 percent of the respondents (typically those who did not buy anything on Black Friday) were unable to judge prices, supply, and company communications. Among the respondents – those who expressed a firm opinion about prices and the supply of products – those who were satisfied and those who were not were almost in the same proportion (Fig. 4). From 2016 to 2017, the proportion of respondents who were satisfied somewhat reduced, which can be explained by the fact that retailers lured buyers with smaller (generally 20-40 percent) discounts, 60-70 percent reductions were low. The rate of price reduction is also a significant difference between the Hungarian and international Black Friday events.

Fig. 4: Satisfaction with prices, 2016, 2017


Satisfaction with product availability and price levels is considerably lower than satisfaction with marketing messages (Fig. 5). This indicates that most Black Friday retailers are investing significant resources in informing potential buyers about the range of discounted products and the extent of their sales.

24

Fig. 5: Satisfaction with marketing communication, 2016, 2017


3.3 Classification of shoppers and non-shoppers

After not finding space-based differences in shopping patterns, we tried to form groups from surveyed people, based on their opinions related to the Black Friday product supply, prices, communication, availability, services and delivery. These six categories (Tab. 4) were measured in 1-4, bipolar Likert scales with an additional “I cannot judge” (0) value. (value 1: totally dissatisfied; 2: rather dissatisfied; 3: rather satisfied; 4: totally satisfied)

In order to group the sample, we first used hierarchical clustering with Ward’s method. The test and its visualisation (dendrogram) indicated 6 different clusters, then we launched a K-mean clustering with 6 clusters. In Tab. 4 the final cluster center values can be seen which show the differences among the clusters. Using these values and cross-tabs with cluster membership and shopping activity we named each cluster, as Tab. 4 indicates. In Tab. 5 a crosstab of the shopping activity in each cluster (shopper/non-shopper numbers) can be found.

Tab. 4: Final cluster center data, 2016

Clusters 1 2 3 4 5 6

Satis-fied

shop-pers

Advertis-ment

sensitive non-

shoppers

Dissa-tisfied shop-pers

Informed shoppers

who hardly shop

Semi-satisfied shoppers and non-shoppers

Non-informed people –

non shopper

Satisfaction with prices 3,15 ,74 1,52 2,55 2,63 ,05

Satisfaction with product supply

2,94 ,81 1,38 2,45 2,86 ,06

Satisfaction with advertisment and communication

2,89 2,90 1,24 3,27 3,34 ,04

Satisfaction with product availability

2,86 ,29 1,76 1,77 2,26 ,02

Satisfaction with servise level

3,39 ,23 1,19 ,00 2,31 ,02

Satisfaction with delivery

3,39 ,00 1,29 ,14 1,06 ,01


25

Tab. 5: Crosstabs on shopping and cluster membership, 2016

Cluster Number of Case 1 2 3 4 5 6

Satis-fied

shop-pers

Advert-isment

sensitive non-

shoppers

Dissatisfied

shoppers

Informed shoppers

who hardly shop

Semi-satisfied shoppers and non-shoppers

Non-informed people-

non shoppers

Count Count Count Count Count Count Have you shopped on Black Friday week?

Yes 60 0 9 7 17 8

No 6 31 12 15 18 89

Number of Cases in each Cluster

272 66 31 21 22 35 97


The above mentioned groups of customers can be characterised as follows:

1. Satisfied shoppers

They are the most satisfied group in each measured dimension (Tab. 4). In this group most responders shopped during the Black Friday week (90 percent), and – according to the data in Tab. 4 – they were rather or totally satisfied as the average values show in Tab. 4. To this group belong 22 percent of all responders, and they supposed to be the “Black Friday fans in Hungary”.

2. Advertisement sensitive – non-shoppers

This group of customers was focused on Black Friday communication. They had information from different ad-sources, but none of them shopped during the Black Friday week (Tab. 5). The group average values are close to zero in case of prices, product availability, etc. (the responders could not judge them – because they did not shop), but they were rather satisfied with the marketing communication of the Black Friday, it means that the messages of companies reached them. We can conclude that in case of these shoppers the marketing communication was partly successful because retailers reached the awareness of customers, but they failed to convince them to shop.

3. Dissatisfied shoppers

In this group customers were rather or totally dissatisfied with the Black Friday event. The average values of all the measured dimensions are near 1 (Tab. 4). Maybe this is the reason why only 43 percent of this group shopped during the Black Friday week. Supposedly they originally planned to shop in the Black Friday week, but they were dissatisfied with the offer of retailers in general.

4. Informed shoppers who hardly shop

Only 32 percent of this group reported that he or she shopped during the Black Friday week. On average basis they were rather satisfied the product supply and prices (average values over 2), and they were almost totally satisfied with the advertisement activity of retailers (value over 3), but they could not judge the product availability, retail services and delivery. It means that they were reached by the retailers’ messages, they reviewed the market offer but eventually they did not shop. The main differences between the 3rd group (“Dissatisfied shoppers”) and this 4th group (“Informed

26

shoppers who hardly shop”) can be depicted by the judgement of product supply, prices and advertisement. While the 3rd group members were totally dissatisfied with them, the members of 4th group were generally satisfied with these dimensions. Further researches are needed to unveil the reasons behind. The 4th group composes nearly 10 percent of the entire sample, it means their shopping behavior may be relevant and important enough for different retailers.

5. Semi-satisfied shoppers and non-shoppers

This group of surveyed people forms about 15 percent of the total sample. This group of responders are the “unconfirmed ones”. About 50-50 percent are the ratio of shoppers and non-shoppers in the group. They average satisfaction with the researched dimensions (prices, product availability, delivery, etc.) are between the “rather unsatisfied” and “rather satisfied” levels. Considering these data, we can conclude that there is still room to improve the retail companies’ marketing and retailing activities in order to extend the ratio of Black Friday shoppers.

6. Non-informed people – non shoppers

More than 36 percent of the sample reported that they had no information on Black Friday at all. The mass majority of this group (over 90 percent) has not shopped during the Black Friday week. Although the Black Friday event started in 2014 in Hungary, 1/3rd of the surveyed people was still totally uninformed after 3 years. These people mean the most important target group for retailers even nowadays if searching for new customers.

4 Discussion

In our paper the Hungarian Black Friday is described from a customer behavioural point-of-view. Because of the non-representative sampling method our findings may be not valid for the entire Hungarian society, although our results are in line with other international examples especially in those countries that have similar socio-economic conditions (e. g. Greece, Romania) (Tsiotsou, 2017) (Delcea, Ioanas, Paun, 2017). The shoppers-non-shoppers ratio shows similar values (1/3-2/3 of the shoppers), the frequencies also show similarities in international comparison, but spending values may vary depending on the economic development level of the selected country.

Further Black Friday researches are needed to unveil the shopping motivations and the behaviour of different retailers in Black Friday sales. Our findings may contribute to a more effective and efficient marketing activities of retail companies that are or will be engaged to Black Friday sales in Hungary and in third countries too.

Conclusions

During our research we highlighted some important features of shopper behavior observed during the 2016 and 2017 Hungarian Black Friday campaigns, in comparison with strong international trends.

Our main findings are as follows:

• In the Hungarian Black Friday sales more than 1/3 of the respondents participate as a buyer, and this trend is in line with international examples, especially in respect of countries where the per capita income is similar to the Hungarian income level.

27

• Hungarian spending significantly lags behind the international spending level, primarily smaller household appliances, fast fashion clothing and smaller IT devices are bought during the sales period.

• During the Hungarian Black Friday sales period online retailers are the most successful, presumably due to the demand side, because domestic buyers prefer shopping opportunities with practical, fast, and lower-priced products to shopping experience. On the supply side, it is crucial that Cyber Monday is missing from the Hungarian trade palette

• The decisive part of the respondents (about half of the strong opinion formers) is not satisfied with either the range of discounted products or with the available discounts. The reason for this might be that a wide range of customers in the online and offline media are aware of the product range and price discounts of the international Black Friday phenomenon, compared to which the discounts and the available product range is more modest. Thus, it is understandable that a large number of customers are dissatisfied with the characteristics of domestic supply

• There are no geography-based differences in the shopping behavior of the surveyed people, we could not identify statistically significant differences in several space-divisions.

• We defined six different groups with cluster analysis. There are solid differences among the groups referring to the advertisement sensitivity, the evaluation of product supply and price levels, and different retail services (availability, delivery)

We found that Hungarian Black Friday is in line with international trends with some special national characteristics. Among them the low level of spending, the online dominancy (lack of Cyber Monday) are the most important. Our customer grouping highlights the differences in shopping/non-shopping, satisfaction/dissatisfaction and information supply. These findings may be useful for international comparison in future Black Friday researches.

References Babbie, E. (2008). A társadalomtudományi kutatás gyakorlata [The practice of social research] Budapest: Balassi Kiadó.

Bell G. C., Weathers M. R., Hastings S. O., Peterson E. B. (2014). Investigating the Celebration of Black Friday as a Communication Ritual. Journal of Creative Communications, 9(3) pp. 235–251.

Delcea C., Ioanas E., Paun R. (2017). Are You Really Influencing Your Customers?: A Black Friday Analysis. In: BILGIN M H. et al (eds.) Empirical Studies on Economics of Innovation, Public Economics and Management, Springer, pp. 225-240

Ghauri P., Grønhaug K. (2011). Kutatásmódszertan az üzleti tanulmányokban, [Research methods in business studies] Budapest: Akadémiai Kiadó.

Józsa L. (2014). Marketingstratégia – A tervezés gyakorlata és elmélete [Marketing strategy – The application and theory of planning]. Budapest: Akadémiai Kiadó.

Kemény I., Simon J. (2015). Az online vásárláshoz köthető minőségészlelés alakulása egy konkrét hazai online könyvesbolt esetében [E-purchase quality perception in the case of a Hungarian online bookstore] Marketing & Menedzsment, 2015/1. pp. 30-42

Kwon H. J., Brinthaupt T. M (2015). The motives, characteristics and experiences of US Black Friday shoppers. Journal of Global Fashion Marketing, Vol. 6, No. 4, pp. 292–302

Lennon S. J., Lee J., Kim M., Johnson K. K. P. (2014). Antecedents of consumer misbehavior on Black Friday: A social responsibility view. Fashion Style and Popular Culture, Vol 1 Number 2.

28

McGee T. (2017). Millennial Habits on Black Friday 2015 vs. 2017. Target Marketing. Available at: http://www.targetmarketingmag.com/post/millennial-habits-on-black-friday-2015-vs-2017/ [17.07.2019].

McLeod F., Cherrett T., Bailey G., Dickinson J. (2016). ‘SHOP AND WE’LL DROP’ ‐ Understanding the impacts of student e-shopping on deliveries to university halls of residence during Black Friday week. Conference Paper. https://www.researchgate.net/publication/312627294

Medgyesi M. (2016). A háztartások megtakarításai és eladósodottsága Magyarországon [Dept status and savings of Hungarian households], 2008-2015 In: Kolosi T., Tóth I. GY. (2016), Társadalmi Riport [Social Report] 2016. Budapest: TÁRKI.

PricewaterhouseCoopers (2017). Holiday Outlook.

https://www.pwc.com/us/en/consumermarkets/assets /pwc-consumer-markets-2017-holiday-report.pdf [03.04.2018]

Rekettye G. (2003). Az ár a marketingben [Price in marketing]. Budapest: KJK-Kerszöv.

Richer, F. (2015). The Billion-Dollar Holiday. Statista.com. Available at: https://www.statista.com/chart/4063/thanksgiving-weekend-e-commerce-sales/ [17.07.20019].

Simpson L., Taylor L., O’Rourke K., Shaw K. (2011). An Analysis of Consumer Behavior on Black Friday. American International Journal of Contemporary Research, Vol. 1 No.1; July http://www.aijcrnet.com/journals/Vol._1_No.1_July_2011/1.pdf

Smith O., Raymen T. (2015). Shopping with violence: Black Friday sales in the British context. Journal of Consumer Culture, 17(3) pp. 677-694.

Swilley, E., Goldsmith, R. E. (2013). Black Friday and Cyber Monday: Understanding consumer intentions on two major shopping days. Journal of Retailing and Consumer Services, 20(1), pp. 43-52

Telegraph Reporters (2017). Which shops aren't taking part in Black Friday 2017 and why?. The Telegraph. Available at: https://www.telegraph.co.uk/black-friday/0/shops-arent-taking-part-black-friday-2017/ [17.07.2019].

Törőcsik M. (2017). Fogyasztói magatartás - Insight, trendek, vásárlók. [Customer behavior –insight, trends, shoppers] Budapest: Akadémiai Kiadó.

Tsiotsou H. R. (2017). „Black Friday”: Attitudes Behavior and Intentions of Greek Consumers In: Martínez-López F. J. et al (eds.) Advances in National Brand and Private Label Marketing. Springer. pp. 11-17

Wu J.–Zitikis R. (2017). Should we opt for the Black Friday discounted price or wait until the Boxing Day?. The Mathematical Scientist 2017/1. https://arxiv.org/pdf/1612.05855.pdf

Contact Address

Dr. habil. Kovács András Ph.D., Budapest Business School, Faculty of Commerce, Catering and tourism Alkotmány street 9-11., H-1054 Budapest, Hungary Email: [email protected] Phone number: +36705039796

Prof. Dr. Sikos T. Tamás, DSc., National University of Public Services, Faculty of Science of Public Governance and Administration, Institute of Public Management and Administration

Ludovika square 2., H-1083, Budapest, Hungary Email: [email protected] Phone number: +36309212273


29

DESIGNING AN ADVANCED PPE MODEL TO MEASURE PERFORMANCE AND EFFICIENCY OF SLOVAK SPA

ENTERPRISES

Veronika Čabinová, Erika Onuferová

Abstract: Each model for measuring performance and prediction is different as it uses different mathematical approaches and works with different indicators. In the era of rapidly changing economic environment, the standard methods for measuring financial performance and assessing financial health are less adequate. Most authors focus on enhancing the predictive ability of original models by responding appropriately to the existing changed economic environment as the identification, increasing and managing enterprise performance and efficiency represent a key tool of today's competitive struggle. The objective of the paper is to create and then apply in practice a new innovative 3D model (PPE model) evaluating the current financial position of Slovak spa companies (P – positon), their future development prognosis (P – prognosis) as well as their efficiency (E – efficiency). The aim of the paper is to identify and implement traditional key indicators, predictive models and efficiency indicators within each of the model dimensions while respecting sectoral characteristics and financial particularities of Slovak spa enterprises. Creating a PPE model will help to better identify the current financial position of the sampled enterprises and, in this way, it will be able to reveal the causes hindering the development of their financial performance to a more accurate extent.

Keywords: PPE Model, Financial Health, Bankruptcy Models, Enterprise Efficiency, Spa Health Sector.

JEL Classification: C0, C53, M31, M21.

Introduction

Nowadays, the analysis and management of enterprise performance and efficiency is getting more and more attention than in the past due to the constantly changing global business environment bringing new, modern approaches to solve the issue in question. Kita, Šimberová (2018) claim that business activities have a large influence on the economy, environment, and society. Therefore, choosing the right key performance metrics is very important to ensure the performance evaluation with the high information value and the ability to subsequently influence and manage performance of enterprises as well as economies. As reported by Gallo, Mihalčová (2016), the main factor to company success is monitoring the actual market situation, therefore, a competitive struggle is won only by enterprises that are adequately dedicated to measuring and evaluating performance as well as efficiency and use the right approaches and measuring tools. Neither the Slovak spa enterprises are no exception, as thanks to them Slovakia belongs to one of the major and the most interesting spa countries in Europe. Although Slovakia is not well known on the international tourism market yet, regional specificity and variability predetermine the Slovak area for the development of tourism, which has become increasingly popular in recent years (Štefko et al., 2018).

30


Performance and efficiency assessment helps managers guide their organization toward achieving excellence leadership and impressive results (Hajiagha et al. 2013). As reported by Stříteská, Zapletal, Jelínková (2016), there is a growing effort to continually develop new financial performance assessment tools that can access the financial situation as accurately as possible, identify future risks, forecast business performance with the intention of determining the starting position of the business. In this regard, Yadav, Sushil, Sagar (2013) add that effective performance management systems and models should be based on system dynamics, sustainability and simulated performance. Their goal is according to Almajali, Alamro, Al-Soub (2012) to gain useful information and background on the efficiency and effectiveness, while encouraging managers to make the best decisions in favour of the enterprise. However, managers solve problems how to measure the performance to prevent the improvement of one part of the business at the expense of another as increasing the implementation of enterprise performance systems is linked to many problems in need of answering.

1.1 The importance of assessing the financial health of enterprises

As reported by Robinson et al. (2015), Narkunienė, Ulbinaitė (2018), despite the significant glorification of modern concepts, financial analysis indicators are most often used in current practice to assess the company's financial health. Palepu, Healy (2013) state that financial analysis can be defined as a comprehensive and systematic analysis of historical and planned financial statements using the financial indicators systems in order to evaluate the current and expected financial situation and to support the quality of its strategic management. According to Kiseľáková, Šoltés (2017), the main objective of financial analysis is to assess not only the state of the company's finances, but rather to analyse the financial health of the company, that is, the ability of the company to obtain returns from the capital invested.

According to Knápková, Homolka, Pavelková (2014), the financial and economic analysis allows revealing whether the company is profitable enough, whether it has the appropriate capital structure, whether it effectively exploits its assets and a whole range of other significant financial facts. Kraftová, Kašparová (2017) add that its aim is not only analysing the accounting data, but also identify the internal and external conditions in which the company carries out its activities. Mihalčová, Gallo, Pružinský (2017) emphasize that measuring company performance by means of generally accepted financial indicators is a source to key information on company efficiency and its future prospects. Vavrek, Adamišin, Kotulič (2017) add that these indicators and methods are easy to implement. However, they are the most distorted and were created for the purposes of the private sector. As reported by Goel (2016), the analysis of the set of financial ratios (indicators of liquidity, activity, capital structure, profitability and market value) is crucial in understanding the company's financial statements, identifying future trends and measuring the overall financial health of an enterprise.

1.2 Bankruptcy and creditworthy prediction models

The knowledge of the financial position of an enterprise ought not to be associated only with its past; however, financial management ought to be predominantly oriented towards the estimation of the company's future development through ex-ante's predictive financial analysis forecasting future development based on the use of

31

creditworthy and bankruptcy models. Therefore, methods based on ex-ante financial-economic analysis are currently getting more and more into the forefront. Kubíčková, Jindřichovská (2015) claim that there are several types of classification available in the literature, however, the most commonly used one is the following:

Bankruptcy models – Altman's Z-score, Taffler's Model, Springate Model, Fulmer's Model, Beerman's Model, Bilderbeek's Model, Index IN95, IN99, IN01, IN05, Virág-Hajdu model, Poznański's model.

Creditworthy models – Quick test, Doucha's Balance Analysis, Creditworthiness index, Aspect Global Rating.

Creditworthy models are based on a company's financial health diagnostics following the score evaluation of the individual areas of economic development. Consequently, the company is assigned according to the gained points to a certain performance category, on the basis of which the financial development is estimated. Bankruptcy models tackle the questions of whether or not the enterprise is going bankrupt, or more precisely its bankruptcy possibility (Karas, Řezňáková, 2017). The models of one-dimensional analysis (classifying companies as being prosperous or not based upon only one criterion) and multidimensional analysis (taking into account several indicators) are applied in this area. The main difference between bankruptcy models and creditworthy ones is that the bankruptcy models are based on real empirical data, while creditworthy models are partly based on theoretical knowledge.

1.3 Enterprise efficiency and its measurement

According to Cyrek (2017), the efficiency is understood as the relationship between outputs and inputs and it is often analysed in terms of goals. Carstina et al. (2015) emphasise that efficiency is closely interdependent to effectiveness, meaning that an efficiency undertaking without being effective will not have a very long period of existence, and an effectiveness of enterprise without obtaining efficiency automatically lead to unfavourable economic results. Enterprise performance evaluation is also based on the analysis of the manner of fulfilment of indicators specific to the different activities performed within processes. In many research studies, the concepts of performance, efficiency and effectiveness are perceived in the same meaning. Wagner (2009) states that efficiency can be perceived as one of the enterprise performance dimensions (see Fig. 1).

Fig. 1: Basic dimensions of performance

Source: (Wagner, 2009)

The efficiency is an important prerequisite for business performance as it represents one aggregate value comprised of multiple areas of financial and business performance assessment. Hedija, Fiala, Kuncová (2017) investigated this link between efficiency

„Do the right things“ – performance in terms of choosing the activity we perform.

„Doing things right“ – performance in terms of how we perform the activity.

Effectiveness Efficiency

Performance dimensions

32

and performance (measured by ROE, ROA and ROS). They found very weak or in many cases not statistically significant relationship between efficiency and profitability indicators. Based on the above-mentioned statements we can define performance measurement as a process of quantifying the efficiency and effectiveness of enterprise activities.

2 Methods

Spa tourism is one of the economy sectors with high growth potential and the main product line of tourism in Slovakia. For this reason, we focused on creating an innovative PPE model evaluating the current financial position of Slovak spa companies, their efficiency and forecasting their future development while respecting sectoral characteristics and financial particularities. We analysed links between the dimensions of the business performance and efficiency as we consider them to be some of the most important areas of business activity leading to the achievement of the strategic goals set. When modelling and analysing business processes and models, the main emphasis is usually put on model validity and accuracy, i.e., the model meets the formal specification and also models the relevant system (Ibl, Čapek, 2016). The new PPE model's structure is illustrated in the following Fig. 2.

Fig. 2: The basic scheme of the new PPE model

Source: (own processing)

2.1 Research sample and data

The research sample consisted of the 28 Slovak health spa enterprises with the official permission from the Ministry of Health of the Slovak Republic to operate the natural health spas and spa medical institutions in Slovakia. According to the statistical classification of economic activities (SK NACE Rev. 2), Slovak spa enterprises belong to section Q Health and social assistance, Division 89 Health and specific subcategory 86.909 Other health care activities. During the years 2013 – 2017, the total number of employees operating in the section Q – Health and social assistance was 6,305 on average and registered number of employees fluctuated on the average level of 81,792 (Statdat, 2019).The financial statements of the analysed spa enterprises' sample were drawn from a publicly available internet portal managed by the company DataSpot, Ltd. Data related to upper and lower quartile values within the subcategory SK NACE 86.909 were provided by CRIF - Slovak Credit Bureau, Ltd. All data obtained were processed in Microsoft Excel and statistical program STATISTICA, 5th edition. The identified outliers of financial indicators and prediction models have been removed due to the distortion of the final results and model created.

Efficiency Position Prognosis

PPE model(performance efficiency)

Simplex Linear Programming Method

Financial-economic analysis ex-post

Financial-economic analysis ex-ante

33

2.2 Quantification of individual dimensions of the PPE model

The assessment of the current financial position of the mentioned group of companies for the period of 2013-2017 was carried out on the basis of an ex-post financial-economic analysis. From each of the basic groups of ratio indicators (liquidity, activity, rentability and indicators of capital structure) we selected and quantified 10 ratio indicators (see Tab. 1) generally widespread and most commonly used in business practice. The calculation of the initial values of the financial indicators entering the financial health dimension was performed on the basis of Transformation Tab. 1. It was compiled according to the average upper and lower quartile values of the selected 10 indicators within the sector SK NACE 86 909. These values were used to determine the range of intervals (a total of 6) that were assigned points between 0 and 10. On the basis of the average value reached over the years 2013-2017, a particular spa company was rated by a maximum of 100 points. For a better understanding the calculation methodology and converting values of financial indicators to points, the transformation table is set out in Attachment 1.

So as to evaluate the financial prediction dimension, 10 selected ex-ante financial prediction models were used, focusing on the selection of creditworthy and bankruptcy models designed and applied under European conditions (see Tab. 2). The calculation of the original values of the predictive models entering the financial prediction dimension was performed on the basis of a Transformation table compiled in the same way as in the previous dimension. The only difference was that ranges of intervals were not based on quartile values, but on the generally recommended ratings and multilevel scales according to the individual prediction models. In the case of Bilderbeek's and Poznański's Model, the limit value between "safe" and "distress" zone of bankruptcy is 0. It would not be possible to set up a range of intervals, so we have determined the values of 5 and -5 as limit values. For a better understanding the calculation methodology and converting values of prediction models to points, the transformation table is set out in Attachment 2.

The calculation of the efficiency of spa companies was addressed using the Simplex Linear Programming Method (SLPM). According to Grell, Hyránek (2012), in the practical solution it is necessary to start from its simplification, while minimizing the deviations between the indicators of efficiency and effectiveness. The ui vectors (in relation to efficiency indicators) and tr vectors (in relation to effectiveness indicators) were obtained as a solution of:

the basic equation:

min Σj wj = 0 (1)

under the conditions:

Σi uiSJM,ij Σr tr crj wj = 0; Σ tr = 1; uitrwj ≥ 0 (2)

so finally, the order of transformation process efficiency was calculated by:

Ej = Σr tr crj / Σi uiSJM,ij, (3)

where: wj – deviations in individual years, ui – value of weights for inputs, SJ

M,ij – inputs needed for linear programming,

tr – value of weights for outputs, crj – outputs needed for linear programming, Ej efficiency.

34

The conversion of the original values of financial ratio indicators, bankruptcy models and achieved level of enterprise efficiency was realized on the basis of modified scoring method according to Rejnuš (2014). Each indicator from a given dimension can get a maximum of 10 points. If the maximum is not reached, points are assigned as follows: in the case of indicators whose development is going to be growing, we calculate the point rating by substituting the highest value of the indicator into the denominator; in the case of indicators whose development is going to be decreasing, we calculate the point rating by substituting the lowest value of the indicator into the numerator. A summary of points within the individual PPE dimensions was then plotted along the x-axis (the dimension of financial position), the y-axis (the dimension of financial prognosis) and the z-axis (the dimension of enterprise efficiency). Based on the cross point of the values achieved, we can determine the position of particular spa enterprise within the five performance fields:

< 0 – 20 > → unacceptable result, < 20 – 40 > → unsatisfactory result, < 40 – 60 > → acceptable result, < 60 – 80 > → satisfactory result, < 80 – 100 > → superior result.

3 Problem solving

3.1 Evaluation of financial position of enterprises (1st dimension)

The following partial analysis is aimed at evaluating and interpreting the development of selected financial ratio indicators of all spa enterprises forming the first dimension of the PPE model (see Tab. 1).

Tab. 1: The development of average values of selected ratio indicators applied to spa companies in Slovakia over the years 2013 2017

Financial indicator (ex-post)

Unit 2013 2014 2015 2016 2017 Std. Dev.

Current Liquidity coeff. 0.99 0.94 1.06 1.02 0.98 0.0402Total Liquidity coeff. 1.04 0.99 1.10 1.08 1.13 0.0487Days Short-term Receivable days 37 47 84 77 43 19.096Days Short-term Payable days 103 108 148 147 118 19.156Total Indebtedness % 28.40 28.32 33.31 33.57 31.90 2.3084Interest Coverage Ratio coeff. 2.33 5.51 8.00 8.91 0.56 3.2054Total Credit Indebtedness % 42.96 18.15 12.50 11.22 27.05 11.708Return On Assets % 1.29 2.61 2.67 2.79 0.19 1.0183Return On Equity % 0.33 2.13 2.36 2.61 -1.03 1.4069Return On Sales % 0.55 3.46 3.67 4.09 -1.53 2.1836


As for ratio indicators, we focused on Current liquidity, which belongs to the key performance indicators. In the course of the monitored period, the indicator indicating the illiquidity of the spa enterprises was below the value of 1, except for years of 2015 and 2016. Identical below-average values were also recorded for the Total Liquidity indicator. The analysis of Days Short-term Receivable Outstanding pointed out the poor payment discipline for business entities in the spa tourism. The spa enterprises

35

had been receiving a commercial loan throughout the period, so businesses lived to the detriment of their suppliers and acted relatively unethically. This matter is an issue of the bad business and credit policy. Based on the average Total Indebtedness of the spa enterprises at 31.10%, the level of Creditor Risk can be considered as optimal. The enterprises were financed mainly by means of their own capital, which demonstrates their financial independence. In 2013, they were unable to pay the price of foreign capital without problems. However, in the last year, the pre-tax profit was down by 93.45%, causing a significant negative downward trend in the indicator and directly reducing the business performance. Return on Assets, as one of the key performance indicators, averaged only at 1.91%, indicating the weak manufacturing power of the industry. A more profound analysis of Return on Equity and Return on Sales showed that during the years of 2013 and 2016 all indicators demonstrated a slow and insignificant, yet positively growing trend.

3.2 Evaluation of financial prediction of enterprises (2nd dimension)

The following partial analysis was aimed at evaluating the prediction model results of all spa enterprises forming the second dimension of the PPE model (see Tab. 2).

Tab. 2: The development of average values of the selected prediction models applied to spa companies in Slovakia over the years 2013 2017

Prediction model (ex-ante)

Unit 2013 2014 2015 2016 2017 Std. Dev.

Quick Test coeff. 12.96 10.93 11.46 11.25 11.14 0.7032Doucha's Balance Analysis coeff. 2.85 1.07 0.82 1.19 4.24 1.3151Aspect Global Rating Model coeff. 9.04 4.60 4.76 4.04 3.27 2.0176Altman's Model (SR) coeff. 0.48 0.52 0.49 0.48 0.50 0.0150Taffler's Model coeff. -0.04 0.33 0.42 0.43 0.40 0.1775Creditworthiness Index coeff. 0.34 1.24 1.29 1.24 0.62 0.3911Beerman's Model coeff. -0.23 -0.08 -0.11 -0.08 -0.05 0.0629Index IN05 coeff. 4.54 0.66 2.71 1.45 -1.51 2.0223Bilderbeek's Model coeff. -9.32 -4.96 -6.37 -8.52 -5.28 1.4706Poznański's Model coeff. 13.27 5.51 5.05 4.64 3.79 3.4280

The legend: the financial situation and prospects are very good, grey zone, the financial situation is uncertain,

the financial situation and the prospects are very bad. Source: (own processing)

On the basis of the resulting values of the selected credit models, it is possible to state that the spa enterprises were in the grey zone as regards the tested results as part of the Quick Test scoring and their financial situation for the future is, therefore, uncertain. Excellent score evaluation, except in 2015, were achieved by spa companies when the Doucha's Balance Analysis was applied, based on which one can expect a favourable positive financial performance of businesses. The last (and at the same time the only) creditworthy model, according to which the prospects of spa companies are very poor, based on the results from 2017, is the Aspect Global Rating Model (spa enterprises were included into the CCC rating group). According to Bilderbeek's Model, Beerman's Model and Poznański's Model, the financial situation of the companies was assessed as very good, with a very low probability of bankruptcy, free of any problems regarding solvency and expected future. According to the results of

36

the Taffler's Model and Index IN 05, the group of the analysed enterprises was alternately assigned to successful and unsuccessful zones regarding their future development.

3.3 Evaluation of enterprises efficiency (3rd dimension)

To conclude, we evaluated the effectiveness of Slovak spa enterprises by means of SLPM. As input variables, we chose total costs, personnel costs and material costs. Output variables were represented by total revenues, net profit and value added. Based on the variables we considered to be the most determinant of the performance within the analysed sample of enterprises, we quantified the cost of returns, wage efficiency and material efficiency by applying the modified matrix system. The outputs were quantified in relation to the total revenues (index). In the following Tab. 3 are stated initial values of the selected inputs and outputs that represent the main basis for the efficiency quantification. Individual values (in €) represent the average values within the all health spa enterprises and indexes are quantified to overall revenues.

Tab. 3: Input data needed for application of SLPM

Year

Inputs Outputs Total

costs (n1) Personnel costs (n2)

Material costs (n3)

Total revenues (v1)

Net profit (v2)

Added value (v3)

Cost of returns

Wage efficiency

Material efficiency

In relation to total revenues



2013 € 5,272,344 € 1,793,050 € 1,191,684 € 5,446,131 € 173,787 € 3,050,390

0.9681 0.3292 0.2188 1.0000 0.0319 0.5601

2014 € 5,342,795 € 1,861,158 € 1,187,809 € 5,535,894 € 193,099 € 3,249,967

0.9651 0.3362 0.2146 1.0000 0.0349 0.5871

2015 € 5,322,864 € 1,899,399 € 1,176,291 € 5,538,486 € 215,622 € 3,224,747

0.9611 03429 0.2124 1.0000 0.0389 0.5822

2016 € 6,013,584 € 2,060,184 € 1,211,600 € 5,927,758 € 85,825 € 3,408,810

1.0145 0.3475 0.2044 1.0000 -0.0145 0.5751

2017 € 5,979,006 € 2,235,240 € 1,217,483 € 6,182,136 € 203,130 € 3,195,759

0.9671 0.3616 0.1969 1.0000 0.0329 0.5169Source: (own processing)

The Simplex Linear Programming Method analysis of efficiency consisted of weights (ui, tr) of selected indicators and deviations (wj), the sum of which had to be minimized. Based on the results, we can state that the highest weight of the selected inputs was proved in the case of material efficiency indicator (u3 = 2.03604) and wage efficiency indicator (u2 = 1.35644). No significant weight was identified for the other selected input and output items. On the basis of calculated weights, we subsequently quantified the efficiency of the selected sample of enterprises. Maximum efficiency (1.000) was reached in 2015, vice-versa, the lowest in 2016 (0.7889). In the other years of the analysed period, the efficiency ranged from 0.8514 (2013) to 0.9912 (2017). The overall average efficiency of the Slovak health spa enterprises reached the level of 0.9254, so it can be considered relatively high.

3.4 Designing a final PPE model

On the basis of created methodology and results of the individual dimensions, we compiled the final PPE model evaluating the efficiency and effectiveness of Slovak spa companies over the years 2013 2017 (see Fig. 3).

37

Fig. 3: The position of Slovak spa companies after the application of PPE model


Based on the overall average score achieved after the transformation of values across all PPE dimensions, SE03 (Spa Bojnice, Inc.) obtained the first place. When carrying out a more detailed analysis of its position, we came to conclusion that the selected company achieved the highest overall values in the 1st dimension "Position" (72 points); for the 2nd dimension "Prognosis" achieved the 2nd place (86 points), while by obtaining 100 points in the area of efficiency, obtained the 1st place together with SE21 (Spa Pieniny Resort, Ltd.). The second highest overall score for the PPE was achieved by SE23 (Specialised spa Marina, s.p.), but it obtained leading score only in financial rating (68 points) and financial prediction (88 points). The resulting business efficiency reached 81 points, which was the main negative determinant worsening its overall score. The third most successful company was SE26 (Specialised spa SR Bystrá). However, the enterprise obtained significant score (84 points) only in the 2nd dimension of "Prognosis". Within the overall assessment of spa companies based on PPE model application as well as within its individual dimensions, it was SE04 (Spa Brusno, Inc.) that achieved the worst results. It was only in the dimension of financial prediction that it overcame another company, namely the SE10 (Spa Sliač, Inc.) by 4 points. Another cause of the unfavourable values of this business was its markedly low efficiency level achieved (16 points). It was SE15 (Spa Číž, Inc.) that ended up being the last but two, whose overall score was negatively affected by the results of the 1st dimension "Position" (26 points) and the third "Efficiency" (33 points).

In the end of our research, we categorized the average scores of individual spa companies into the performance and efficiency fields defined in the Methodology section. The results are graphically processed in Fig. 4. Based on the values, we may ascertain that only one of the enterprises achieved an unacceptable result (SE04 – having obtained the last place in the PPE model) and only one reached values highly above the average (SE03 – having obtained the first place in the PPE model). Nearly half of the analysed spa companies (46.43%) achieved on average acceptable results, which we consider quite positive. Most of the remaining research sample (28.57%) showed satisfactory results, with the remaining 17.86% pointing to problems with achieved levels of financial performance and efficiency.

38

Fig. 4: Structure of performance and efficiency fields of Slovak spa companies


4 Discussion

The authors Harumová, Janisová (2014), Šofranková, Kiseľáková, Horváthová (2017), Hyránek et al. (2018) have also focused on creating new models for evaluating the performance of Slovak enterprises using various methods and approaches. However, the motivation to create PPE model arose also from the fact that in the current Slovak conditions there is a lack of a 3D models analysing the performance and efficiency within the tourism sector. As we consider spa enterprises to be the main product line of tourism in Slovakia with a high growth potential, we have focused on creating an innovative PPE model that will help to analyse the current and future financial health of spa enterprises. Moreover, it will identify more accurately causes that hinder the development of financial performance and efficiency of the selected sample of enterprises.

Conclusion

Sustainable development is now found as a central theme (Lešáková, Baťa, Provazníková, 2017). Therefore, performance and efficiency measurement is considered to be a topical issue. Choosing the right key metrics is very important to ensure the performance and efficiency evaluation and the adaptation of the models created to the sectoral characteristics and financial specifics of individual sectors of the economy is increasingly desirable. As reported by Pawliczek, Kozel, Vilamová (2018), strategic planning as well as continuous improvement are very important factors close to business performance. Based on the above-mentioned facts, we focused on creating a 3D model design (PPE model) in the conditions of the Slovak spa enterprises, which would be able to evaluate the level of three important dimensions of each business activity – financial position (P – position), prognosis of future development (P prognosis) as well as efficiency (E efficiency).

After having applied the created PPE model in Slovak spa companies, we have come to the following conclusions. Based on the overall average score achieved after the transformation of values across all PPE dimensions, spa companies SE03, SE23, SE26 with an average score of 78 points reached the first three places. On the other hand, the worst rated companies were SE04, SE10, SE15 whose score in the PPE model was only around 23 points. It can be, however, generally stated that almost half of Slovakia's spa companies (46.43%) achieved on average acceptable results of

39

32.14%, while 32.14% of the results were satisfactory and in one case, they were much higher than the average. Only one of the enterprises showed an unsatisfactory level in all the dimensions of the PPE model. The remaining 17.86% pointed to certain problems in the analysed area of financial performance and efficiency, the more detailed research of which will be the subject of our further scientific studies.

Acknowledgement This contribution was supported by project GAMA/19/2, “Designing an Advanced

PPE Model to Measure Performance and Efficiency of Slovak Spa Enterprises”.

References

Almajali, A. Y., Alamro, S. A., Al-Soub, Y. Z.. (2012). Factors Affecting the Financial Performance of Jordanian Insurance Companies Listed at Amman Stock Exchange. Journal of Management Research, 4(2), pp. 266-289.

Cyrek, M.. (2017). Social Efficiency of Employment in Three Sectors – A Comparison of Polish Regions. Equilibrium. Quarterly Journal of Economics and Economic Policy, 12(3), pp. 417-432. DOI https://doi.org/10.24136/eq.v12i3.22

Carstina, S., Siminica, M., Circiumaru, D., Tanasie, A.. (2015). Correlation Analysis of the Indicators of Asset Management and Profitability. International Journal of Economics and Business Administration, 3(3), pp. 3-21.

Gallo, P., Mihalčová, B.. (2016). Models of Evaluation of Managing People in Companies. Quality – Access to Success, 17(155), pp. 116-119.

Goel, S.. (2016). Financial Ratios. New York: Business Expert Press, LLC.

Grell, M., Hyránek, E.. (2012). Maticové modely na meranie výkonnosti produkčných systémov. E+M Ekonomie a Management, 15(1), pp. 73-87.

Hajiagha, S. H. R., Akrami, H., Zavadskas, E. K., Hashemi, S. S.. (2013). An Intuitionistic Fuzzy Data Envelopment Analysis for Efficiency Evaluation Under Ucertainty: Case of a Finance and Credit Institution. E+M Ekonomie a Management, 16(1), pp. 128-137.

Harumová, A., Janisová, M.. (2014). Hodnotenie slovenských podnikov pomocou skóringovej funkcie. Ekonomický časopis, 62(5), pp. 522-539.

Hedija, V., Fiala, R., Kuncová, M.. (2017). Is Profitability a Good Proxy for Efficiency? Evidence from the Subsector of Tour operators. Review of Economic Perspectives, 17(4), pp. 425-440. DOI https://doi.org/10.1515/revecp-2017-0022

Hyránek, E., Grell, M., Nagy, L., Ďurinová, I.. (2018). The Economic-mathematical Nature of the HGN Model Concept as a Tool for Measuring Performance of Enterprises. In: Ekonomický časopis, 66(3), pp. 309-325.

Ibl, M., Čapek, J.. (2016). Measure of Uncertainty in Process Models Using Stochastic Petri Nets and Shannon Entropy. Entropy, 18(1), pp. 1-14. DOI https://doi.org/10.3390/e18010033

Karas, M., Řezňáková, M.. (2017). The Stability of Bankruptcy Predictors in the Construction and Manufacturing Industries at Various Times before Bankruptcy. E+M Ekonomie a Management, 20(2), pp. 116-133.

Kiseľáková, D., Šoltés, M.. (2017). Modely řízení finanční výkonnosti v teorii a praxi malých a středních podniků. 1st edition. Prague: Grada Publishing.

Kita, P., Šimberová, I.. (2018). Business Model Research Proposal: Novel Business Model Concepts Based on Sustainable Multiple Customer Value Creation in a Selected Industry. Scientific Papers of the University of Pardubice: Series D, 26(44(3/2018)), pp. 114-126.

Knápková, A., Homolka, L., Pavelková, D.. (2014). Využití Balanced Scorecard a vliv jeho využívání na finanční výkonnost podniků v ČR. E+M Ekonomie a Management, 17(2), pp. 146-160.

40

Kraftová, I., Kašparová, L.. (2017). Assessment of the Financial Health of Regional Emergency Medical Services in the Czech Republic Using the BAMF Model. Journal of Business Economics and Management, 18(2), pp. 340-353. DOI https://doi.org/10.3846/16111699.2016.1272067

Kubíčková, D., Jindřichovská, I.. (2015). Finanční analýza a hodnocení výkonnosti firem. 1st edition. Prague: C. H. Beck.

Lešáková, P., Baťa, R., Provazníková, R.. (2017). Implementation of the SDGs under Czech Conditions and Measuring Selected Indicators. In: 29th International Business Information Management Association Conference. Vienna: INT Business Information Management Assoc-IBIMA. pp. 1076-1084.

Mihalčová, B., Gallo P., Pružinský M.. (2017). Measurement of Company Performance in a Health Resort in Slovakia taken as an Example. Quality – Access to Success, 18(158), pp. 79-84.

Narkunienė, J., Ulbinaitė, A.. (2018). Comparative Analysis of Company Performance Evaluation Methods. In: Entrepreneurship and Sustainability Issues, 6(1), pp. 125-138.

Palepu, K. G., Healy, P. M.. (2013). Business Analysis Valuation: Using Financial Statements. 5th ed. Mason, OH: South-Western Cengage Learning.

Pawliczek, A., Kozel, R., Vilamová, Š.. (2018). Strategic Management and Performance of Enterprises Evaluated Using Chaid Decision Tree Analysis. Scientific Papers of the University of Pardubice: Series D, 26(44(3/2018)), pp. 196-208.

Rejnuš, O.. (2014). Finanční trhy. 4th edition. Prague: Grada Publishing.

Robinson, T. R., Henry, E., Pirie, W. L., Broihahn, M. A.. (2015). International Financial Statement Analysis. 3rd ed. Hoboken, NJ: John Wiley & Sons.

Statdat, (2019). Statistics. online. Statdat. Available at: http://statdat.statistics.sk 17.05.2019.

Stříteská, M., Zapletal, D., Jelínková L.. (2016). Performance Management Systems in Czech Companies: Findings from a Questionnaire Survey. E+M Ekonomie a Management, 19(4), pp. 44-55.

Šofranková, B., Kiseľáková, D., Horváthová, J.. (2017). Actual Questions of Risk Management in Models Affecting Enterprise Performace. Ekonomický časopis, 65(7), pp. 644-667.

Štefko, R., Vašaničová, P., Litavcová, E., Jenčová, S.. (2018). Tourism Intensity in the NUTS III Regions of Slovakia. Journal of Tourism and Services, 9(16), pp. 1804-5650. DOI https://doi.org/10.29036/jots.v9i16.43

Vavrek, R., Adamišin, P., Kotulič, R.. (2017). Multi-criteria Evaluation of Municipalities in Slovakia - Case Study in Selected Districts. Polish Journal of Management Studies, 16(2), pp. 290-301. DOI https://doi.org/10.17512/pjms.2017.16.2.25

Wagner, J.. (2009). Měření výkonnosti: Jak měřit, vyhodnocovat a využívat informace o podnikové výkonnosti. 1st edition. Prague: Grada Publishing.

Yadav, N., Sushil, Sagar, M.. (2013). Performance Measurement and Management Frameworks: Research Trends of the Last Two Decades. Business Process Management Journal, 19(6), pp. 947-970.

Contact Address

Mgr. Veronika Čabinová University of Prešov in Prešov, Faculty of Management Konštantínova 16, 080 01, Prešov, Slovakia E-mail: [email protected]

Mgr. Erika Onuferová University of Prešov in Prešov, Faculty of Management Konštantínova 16, 080 01, Prešov, Slovakia E-mail: [email protected]

41

Attachment 1: Transformation Tab. 1 – financial ratio indicators

Ratio indicator

Range of intervals PointsRatio

indicator Range of intervals Points

Current Liquidity

> 4.6980* 10

Interest Coverage

Ratio

< 0.0000** 10

4.6979 – 3.6765 8 0.0001 – 5.0050 8

3.6764 – 2.6550 6 5.0051 – 10.0100 6

2.6549 – 1.6335 4 10.0101 – 15.0150 4

1.6334 – 0.6120 2 15.0151 – 20.0200 2

< 0.6120** 0 > 20.0200* 0

Total Liquidity

> 4.8800* 10

Total Credit Indebtedness

> 41.0340* 10

4.8799 – 3.8230 8 41.0339 – 8.3615 8

3.8229 – 2.7660 6 8.3614 – -24.3110 6

2.7659 – 1.7090 4 -24.3111 – -56.9835 4

1.7089 – 0.6520 2 -56.9836 – -89.6560 2

< 0.6520** 0 < -89.6560** 0

Days Short-term

Receivable Outstanding

< 31.9080** 10

Return On Assets

> 20.0940* 10

31.9081 – 46.4790 8 20.0939 – 14.7135 8

46.4791 – 61.0500 6 14.7134 – 9.3330 6

61.0501 – 75.6210 4 9.3329 – 3.9525 4

75.6211 – 90.1920 2 3.9524 – -1.4280 2

> 90.1920* 0 < -1.4280** 0

Days Short-term Payable Outstanding

< 46.0960** 10

Return On Equity

> 39.1820* 10

46.0961 – 104.5730 8 39.1819 – 26.5985 8

104.5731 – 163.0500 6 26.5984 – 14.0150 6

163.0501 – 221.5270 4 14.0149 – 1.4315 4

221.5271 – 280.0040 2 1.4314 – -11.1520 2

> 280.0040* 0 < -11.1520** 0

Total Indebtedness

< 19.2180** 10

Return On Sales

> 16.6400* 10

19.2190 – 36.0720 8 16.6399 – 11.8030 8

36.0721 – 52.9260 6 11.8029 – 6.9660 6

52.9261 69.7800 4 6.9659 – 2.1290 4

69.7801 – 86.6340 2 2.1289 – -2.7080 2

> 86.6340* 0 < -2.7080** 0

Note: * average upper quartile values over the years 2013 – 2017 within the sector SK NACE 86 909 ** average lower quartile values over the years 2013 – 2017 within the sector SK NACE 86 909


42

Attachment 2: Transformation Tab. 2 – prediction models

Prediction model

Range of intervals PointsPrediction

model Range of intervals Points

Quick Test

< 4.0000* 10

Credit-worthiness

Index

> 3.0000* 10

4.0001 – 8.0000 8 2.9999 – 1.7500 8

7.9999 – 12.0000 6 1.7499 – 0.5000 6

11.9999 – 16.0000 4 0.4999 – -0.7500 4

16.0001 – 20.0000 2 -0.7501 – -2.0000 2

> 20.0000** 0 < -2.0000** 0

Doucha's Balance Analysis

> 1.0000* 10

Beerman's Model

< 0.2000* 10

0.9999 – 0.8750 8 0.2001 – 0.2375 8

0.8749 – 0.7500 6 0.2376 – 0.2750 6

0.7499 – 0.6250 4 0.2751 – 0.3125 4

0.6249 – 0.5000 2 0.3126 – 0.3500 2

< 0.5000** 0 > 0.3500** 0

Aspect Global Rating Model

> 8.5000* 10

Index IN05

> 1.6000* 10

8.4999 – 6.7500 8 1.5999 – 1.4250 8

6.7499 – 5.000 6 1.4249 – 1.2500 6

4.9999 – 3.2500 4 1.2499 – 1.0750 4

3.2499 – 1.5000 2 1.0749 – 0.9000 2

< 1.5000** 0 < 0.9000** 0

Altman's Model (SR)

> 5.0000* 10

Bilderbeek's Model

< -5.000* 10

4.9999 – 3.7500 8 -4.9999 – -2.5000 8

3.7499 – 2.5000 6 -2.4999 – 0.0000 6

2.4999 – 1.2500 4 0.0001 – 2.5000 4

1.2499 – 0.0000 2 2.5001 – 5.0000 2

< 0.0000** 0 > 5.0000** 0

Taffler's Model

> 0.3000* 10

Poznański's Model

> 5.000* 10

0.2999 – 0.2750 8 4.9999 – 2.5000 8

0.2749 – 0.2500 6 2.4999 – 0.0000 6

0.2499 – 0.2250 4 -0.0001 – -2.5000 4

0.2249 – 0.2000 2 -2.5001 – -5.0000 2

< 0.2000** 0 < -5.0000** 0

Note: * limit value to "safe zone" (negligible probability of filing bankruptcy) ** limit value to "distress zone" (high probability of reaching the stage of bankruptcy)



43

MODEL PRO URČENÍ PŘIROZENÉ MÍRY NEZAMĚSTNANOSTI Z MEZERY PRODUKTU

MODEL FOR DETERMINATION OF NATURAL RATE OF UNEMPLOYMENT FROM GDP GAP

Jiří Dobrylovský

Abstract: Since the moment when Milton Friedman introduced a natural rate of unemployment into economic theory, this indicator has been one of the most important quantities in macroeconomics. However, in contrast to other key macroeconomic indicators, the natural rate of unemployment is very difficult to be accurately quantified. The following paper suggests a simple mathematic model, on whose base the natural rate of unemployment can be determined. This model was used to set the natural rate of unemployment in the Czech Republic and its development over time period of the last 23 years based on the Macroeconomic Forecast by the Ministry of Finance of the Czech Republic and compared with Okun´s Law. When calculating the natural rate of unemployment the suggested model eliminates the differences resulting from different utilization of production capacities and the changing rate of economic activity of population in the course of time. As all other models, this model also works with certain simplified assumptions; however, its advantage is the possibility of gradual implementation of more and more difficult parameters of real economy as new variables. Nevertheless, even in its basic form, the model provides good results, which coincide with theoretic basis and monitored facts.

Key words: Natural Unemployment Rate, Participation Rate, Potential Product, Okun's Law, Labor Market.

JEL Classification: C29, E32, J64

Úvod

Přirozená míra nezaměstnanosti, kterou do ekonomické teorie uvedl Milton Friedman (Friedman, 1968), je jedním z klíčových pojmů ekonomie hlavního proudu. Je definována jako stav rovnováhy na agregátním trhu práce a z hlediska vymezení druhů nezaměstnanosti je souhrnem nezaměstnanosti frikční a strukturální (viz např. Mankiw, s. 158).

Namísto pojmu „přirozená míra nezaměstnanosti“ se zejména v posledních desetiletích často používá „míra nezaměstnanosti neakcelerující inflaci“ (non-accelerating rate o unemployment, zkratka NAIRU), odvozovaná z Phillipsovy křivky, kterou zavedli F. Modigliani a L. Papademos (1975), tehdy ještě jako NIRU, tj. neinflační míra nezaměstnanosti. Přestože se nejedná o definitoricky identické pojmy (na dlouhodobé Phillipsově křivce může tatáž přirozená míra nezaměstnanosti odpovídat různým mírám inflace, které již nebude dále akcelerovat), mnozí ekonomové je od sebe explicitně neodlišují. Pro účely této stati, a to zejména kvůli porovnání výsledků empirických analýz, budou oba pojmy považovány za totožné.

Z definičního hlediska nepředstavuje přirozená míra nezaměstnanosti ani NAIRU žádný problém, avšak o to větší komplikací je její kvantitativní určení. Výše zmíněná rovnováha na pracovním trhu, tedy rovnost nabízeného a poptávaného množství,

44

koresponduje hypotetickým dokonale konkurenčním podmínkám, tedy takovému trhu práce, na němž dochází k pružnému přizpůsobování ceny práce, neboli mzdové sazby, vzájemnému poměru poptávky a nabídky. Jenže trh práce z celé řady důvodů dokonale konkurenční není a ze stejných důvodů nedochází ani k pružným změnám mzdových sazeb. Z tohoto hlediska je tedy stanovení přirozené míry nezaměstnanosti spíše jen obtížně odhadovatelnou úrovní.

Podle MCAdama a McMorrowa (1999) se dokonce v případě přirozené míry nezaměstnanosti jedná o teoretickou konstrukci, pro kterou neexistují nástroje přímého měření.

Shodně s tímto názorem tvrdí i Jašová (2010), že není možné NAIRU přímo pozorovat, dodává ale, že může být odvozována z analýzy pozorovatelných proměnných, spojených s definicí této veličiny. Na základě toho pak autorka za použití empirických metod dochází ke stanovení intervalových hodnot NAIRU v několikaletých obdobích v letech 1999-2010.

Empirickou cestou se pokoušejí určit NAIRU Kadeřábková a Jašová (2011), a to dokonce nejen pro české národní hospodářství jako celek, ale i pro jednotlivé sektory primární, sekundární a terciární.

Je ale nicméně ještě jeden prostředek, pomocí něhož lze přirozenou míru nezaměstnanosti relativně přesně stanovit. Tímto prostředkem je tzv. potenciální produkt.

Potenciální produkt je vymezen (Okun, 1962) jako výstup ekonomiky (tj. reálný produkt), který je vytvářen při tzv. plné zaměstnanosti, jež koresponduje v té době ještě nedefinované přirozené míře nezaměstnanosti. Takto řečeno by se mohlo zdát, že se jedná o tautologii či důkaz v kruhu, ale skutečnost je naštěstí jiná. Jde o to, že velikost reálného produktu je možné pomocí statistických metod naprosto spolehlivě vyčíslit.

Lze provést i kvalifikovaný odhad mezery produktu, neboli rozdílu mezi produktem reálným a potenciálním. Ministerstvo financí ČR každé čtvrtletí vydává Makroekonomickou predikci, shrnující aktuální stav vývoje základních makroekonomických indikátorů České republiky a obsahující kvalifikované odhady vývoje v oblasti ekonomického výkonu (včetně mezery produktu za jednotlivé kvartály), cen a situace na českém trhu práce. Tato skutečnost umožňuje provádět návazné analýzy, protože za situace, kdy známe výši reálného produktu a velikost mezery produktu, můžeme pomocí vhodných nástrojů, jako je např. Okunův zákon, stanovit i produkt potenciální, resp. výši přirozené míry nezaměstnanosti, jež potenciálnímu produktu odpovídá.

Výpočet přirozené míry nezaměstnanosti na základě určení mezery produktu a dalších parametrů, uvedených v následujících kapitolách, si tato stať klade jako svůj stěžejní cíl.

1 Formulace problematiky

Graf na obr. 1 s připojenou tabulkou ukazuje hodnoty produktové mezery v ČR za jednotlivá čtvrtletí v letech 1995-2018. Z grafu je jasně patrný cyklický vývoj české ekonomiky, zároveň se na něm však ukazuje ještě jedna důležitá okolnost, a sice že česká ekonomika se v současné době sice nachází nad úrovní potenciálního produktu, avšak tzv. inflační mezera jako kladný rozdíl produktu reálného a potenciálního se pohybuje pod dvěma procentními body. Časté označování současného stavu

45

ekonomiky ČR jako „přehřáté konjunktury“ je tudíž za těchto okolností poněkud přehnané, což potvrzuje i poměrně nízká meziroční míra inflace, pohybující se v letech 2017-18 na úrovni 2 - 2,5 %.

Obr. 1: Mezera produktu v ČR 1995-2018 (v % potenciál. produktu, po čtvrtletích)

Zdroj: Makroekonomická predikce České republiky - listopad 2018

Tab. 1: Mezera produktu v ČR 1995-2018 (v % potenciál. produktu, po čtvrtletích)

Kvartál 95/I 95/II 95/III 95/IV 96/I 96/II 96/III 96/IV 97/I 97/II 97/III Mez. p. 1,1 1,6 2,3 2,7 3,4 3,3 3,0 2,5 1,7 0,4 -0,4 Kvartál 97/IV 98/I 98/II 98/III 98/IV 99/I 99/II 99/III 99/IV 00/I 00/II Mez. p. -1,1 -1,2 -1,3 -1,7 -1,9 -2,4 -2,2 -1,7 -0,9 0,0 0,9 Kvartál 00/III 00/IV 01/I 01/II 01/III 01/IV 02/I 02/II 02/III 02/IV 03/I Mez. p. 1,6 1,4 1,3 1,0 0,5 0,0 -0,5 -1,1 -1,3 -1,6 -2,1 Kvartál 03/II 03/III 03/IV 04/I 04/II 04/III 04/IV 05/I 05/II 05/III 05/IV Mez. p. -2,3 -2,4 -2,5 -2,2 -2,5 -2,1 -1,6 -1,4 -0,7 -0,3 0,4 Kvartál 06/I 06/II 06/III 06/IV 07/I 07/II 07/III 07/IV 08/I 08/II 08/III Mez. p. 1,1 2,1 2,9 2,9 3,0 3,2 3,4 4,3 4,4 4,2 3,7 Kvartál 08/IV 09/I 09/II 09/III 09/IV 10/I 10/II 10/III 10/IV 11/I 11/II Mez. p. 2,2 -3,0 -4,2 -4,1 -3,8 -3,2 -2,2 -1,7 -1,1 -1,0 -0,6 Kvartál 11/III 11/IV 12/I 12/II 12/III 12/IV 13/I 13/II 13/III 13/IV 14/I Mez. p. -0,6 -0,9 -1,3 -2,1 -2,9 -3,4 -4,1 -4,1 -4,2 -3,8 -3,0 Kvartál 14/II 14/III 14/IV 15/I 15/II 15/III 15/IV 16/I 16/II 16/III 16/IV Mez. p. -2,5 -2,0 -1,3 -0,5 0,2 0,6 0,6 0,3 -0,2 -0,6 -0,2 Kvartál 17/I 17/II 17/III 17/IV 18/I 18/II x x x x x Mez. p. 0,4 1,9 1,7 1,8 1,8 1,5 x x x x x

Zdroj: Makroekonomická predikce České republiky - listopad 2018 (vlastní úpravy)

Teoreticky by nyní stačilo stanovit okamžiky, kdy je mezera produktu rovná nule, (a reálný produkt se tudíž shoduje s potenciálním) a k těmto okamžikům pak např. na webu Českého statistického úřadu odečíst míru nezaměstnanosti. Realita je ale podstatně komplikovanější, poněvadž zde do hry vstupují dva klíčové faktory:

1. Na tvorbě produktu se vedle práce podílí též kapitál, a jak je patrné z obr. 2, míra jeho využití se v jednotlivých čtvrtletích lišila, někdy i velmi výrazně. O tyto odchylky je nutné mezeru produktu upravit, aby výsledný údaj odrážel výhradně vliv faktoru práce.

46

Obr. 2: Využití výrobních kapacit v průmyslu v ČR v letech 1995-2018

Zdroj: MF ČR (2018)

2. Využití výrobního faktoru práce je ukazatelem „míra nezaměstnanosti“ vystiženo jen částečně. Vedle toho je nutné brát v potaz též ukazatel „míra ekonomické aktivity obyvatelstva“, jež se ve sledovaném období 1995-2018 měnila až o 10 procentních bodů (viz obr. 3). Odchylky v míře ekonomické aktivity je rovněž nutné eliminovat, resp. zjištěné údaje převést na míru ekonomické aktivity jednotnou pro celé období 1995-2018.

Obr. 3: Míra ekonomické aktivity v ČR v letech 1995-2018

Zdroj: Vlastní úpravy dle ČSÚ (2018)

2 Metody

Na základě výše uvedeného lze formulovat jednoduchý matematický model, uvádějící do souvislosti všechny uvažované činitele, tedy velikost mezery produktu, její ovlivnění nestejným využitím kapitálu a rovněž vliv měnící se míry ekonomické aktivity obyvatelstva. Pro vytvoření příslušného modelu byly uvažovány následující předpoklady:

a) Při tvorbě HDP se trvale uplatňovaly konstantní výnosy z rozsahu, přičemž mezní produktivita práce a mezní produktivita kapitálu se sobě rovnaly.

b) Využití kapitálu bylo za sledované období 1995-2018 uvažováno v jednotné výši 84,1 % instalovaných kapacit, tedy na úrovni, kterou graf 2 uvádí jako průměrné využití kapitálu za celé období. Rozdíly ve skutečném využití kapitálu oproti

47

průměrné hodnotě byly v souladu s předpokladem ad a) kompenzovány koeficientem X, který je určen jako převrácená hodnota podílu skutečné a průměrné hodnoty využití kapitálu v daném čtvrtletí.

c) Míra ekonomické aktivity byla ve sledovaném období 1995-2018 uvažována v jednotné průměrné výši 64,6 %, s tím, že rozdíly ve skutečné míře ekonomické aktivity oproti průměrné úrovni byly v souladu s předpokladem ad a) kompenzovány koeficientem Z, který je určen jako převrácená hodnota podílu skutečné a průměrné hodnoty míry ekonomické aktivity v daném čtvrtletí.

Výsledný matematický vztah, pomocí něhož lze určit čtvrtletní mezeru produktu očištěnou o proměnlivé využití kapitálu a rozdílnou míru ekonomické aktivity obyvatelstva, pak nabývá podoby:

KMP = MP . X . Z (1)

kde KMP je kompenzovaná mezera produktu, neboli mezera produktu očištěná o vliv proměnlivého využití kapitálu a měnící se míry ekonomické aktivity, MP je skutečná mezera produktu, uvedená v makroekonomické predikci a rovněž v Tab. 1, X je koeficient kompenzující proměnlivé využití kapitálu na střední hodnotu 84,1 % a Z je koeficient kompenzující v čase se měnící míru ekonomické aktivity na střední hodnotu 64,6 %.

Po získání hodnot kompenzované mezery produktu je možné pro takto nalezené hodnoty použít veřejně publikované údaje o míře nezaměstnanosti a následně podle Okunova zákona vypočíst přirozenou míru nezaměstnanosti. Okunův zákon lze matematicky vyjádřit různými způsoby; vyjděme např. z formulace (Helísek, 2000):

Y / Y* = 1 + c (u* - u) (2)

kde Y je reálný produkt, Y* potenciální produkt, c je empiricky zjištěný koeficient lineární závislosti (pohybuje se obvykle v intervalu od 0,02 do 0,03; pro další výpočty je uvažována jeho dolní hranice, neboli 0,02), u* je přirozená míra nezaměstnanosti a u je skutečná míra nezaměstnanosti. Jednoduchou úpravou rovnice (2) dostaneme výraz:

(Y / Y*) - 1 = 0,02 (u* - u) (3)

kde (Y/Y*) - 1 představuje mezeru produktu v desetinných číslech, přesněji řečeno kompenzovanou mezeru produktu (KMP) vypočtenou pomocí rovnice (1). Po následném vynásobení KMP číslem 100 dostaneme KMP v procentech, jak je zvykem ji udávat. Poněvadž míra nezaměstnanosti je rovněž standardně udávána v procentech, tak hledanou přirozenou míru nezaměstnanosti následně vypočítáme podle vzorce:

u* = 0,5 KMP + u (4)

3 Rozbor problému

S využitím dat z makroekonomických predikcí a údajů Českého statistického úřadu lze za pomoci rovnic (1), (2) a (3) dospět k následujícím výsledkům (viz. Tab. 2) ohledně výše přirozené míry nezaměstnanosti v České republice:

48

Tab. 2: Mezera produktu a skutečná a přiroz. míra nezaměstnanosti v ČR 1995-2018 Kvartál 95/I 95/II 95/III 95/IV 96/I 96/II 96/III 96/IV 97/I 97/II 97/III u 4,3 4,0 4,1 3,7 3,7 3,8 4,0 4,1 4,3 4,5 5,0 u* 4,82 4,81 5,25 5,04 5,40 5,51 5,51 5,42 5,17 4,71 4,75 Kvartál 97/IV 98/I 98/II 98/III 98/IV 99/I 99/II 99/III 99/IV 00/I 00/II u 5,4 5,9 5,9 6,8 7,3 8,4 8,4 9,0 9,0 9,5 8,7 u* 4,80 5,35 5,20 5,89 6,33 7,21 7,22 8,14 8,50 9,45 9,14 Kvartál 00/III 00/IV 01/I 01/II 01/III 01/IV 02/I 02/II 02/III 02/IV 03/I u 8,6 8,3 8,5 8,0 8,3 7,8 7,7 7,0 7,2 7,3 7,6 u* 9,34 8,98 9,19 8,59 8,60 7,85 7,61 6,54 6,64 6,65 6,70 Kvartál 03/II 03/III 03/IV 04/I 04/II 04/III 04/IV 05/I 05/II 05/III 05/IV u 7,5 8,0 8,1 8,7 8,2 8,2 8,2 8,4 7,8 7,8 7,8 u* 6,50 6,95 7,06 7,70 7,05 7,31 7,56 7,91 7,50 7,75 7,99 Kvartál 06/I 06/II 06/III 06/IV 07/I 07/II 07/III 07/IV 08/I 08/II 08/III u 8,0 7,1 7,0 6,5 6,0 5,3 5,1 4,8 4,7 4,2 4,3 u* 8,58 8,27 8,36 7,99 7,50 6,89 6,78 6,70 6,76 6,23 6,03 Kvartál 08/IV 09/I 09/II 09/III 09/IV 10/I 10/II 10/III 10/IV 11/I 11/II u 4,4 5,8 6,3 7,3 7,3 8,1 7,1 7,1 6,9 7,2 6,7 u* 5,44 3,94 3,87 5,07 5,16 6,29 5,85 6,19 6,24 6,59 6,30 Kvartál 11/III 11/IV 12/I 12/II 12/III 12/IV 13/I 13/II 13/III 13/IV 14/I u 6,5 6,4 7,1 6,7 7,0 7,2 7,4 6,7 6,9 6,7 6,8 u* 6,15 5,90 6,39 5,69 5,69 5,59 5,54 4,96 5,22 5,18 5,72 Kvartál 14/II 14/III 14/IV 15/I 15/II 15/III 15/IV 16/I 16/II 16/III 16/IV u 6,0 5,9 5,7 6,0 4,9 4,8 4,5 4,4 3,9 4,0 3,6 u* 5,07 5,13 5,18 5,85 5,36 5,26 5,04 4,94 4,36 4,55 4,05 Kvartál 17/I 17/II 17/III 17/IV 18/I 18/II 18/III 18/IV 19/I 19/II 19/III u 3,5 3,0 2,8 2,4 2,4 2,2 2,3 x x x x u* 3,95 3,45 3,24 2,84 2,83 2,63 2,72 x x x x

Zdroj: Vlastní výpočty dle Makroekonomické-predikce 2018

Jak je patrné, vypočtené hodnoty přirozené míry nezaměstnanosti vykazují poměrně značné kolísání. To je ve zdánlivém rozporu s představou o přirozené míře nezaměstnanosti jako vytrvale stabilní úrovni, měnící se jen ve velmi dlouhodobém horizontu v důsledku hysterezí na trhu práce (Tobin, 1980), tzv. „vydržovacích poměrů“ (Mindorf, 1985) či tzv. „vykazovacích efektů“ (Dornbusch - Fischer, 1994). Avšak česká ekonomika ve svém transformačním období v 90. letech ještě nebyla standardní tržní ekonomikou, z jejíhož fungování odvozují své závěry výše uvedení ekonomové. Odvětvová skladba ekonomiky ČR zaznamenávala v 90. letech velké strukturální změny (útlum těžkého průmyslu, později zánik velkých podniků typu ČKD, Poldi atd.) a situace na trhu práce tomu zákonitě odpovídala. Vzhledem k tomu, že součástí přirozené míry nezaměstnanosti je strukturální nezaměstnanost, dá se tím snadno vysvětlit její strmý nárůst v druhé pol. 90. let, kdy zanikala dříve tradiční pracovní místa v těžkém průmyslu a namísto nich vznikaly pracovní příležitosti nové, převážně v terciárním sektoru. Tím vznikala strukturální nezaměstnanost, jež se začala výrazně snižovat až o několik let později po uskutečněné rekvalifikaci alespoň u části pracovních sil.

Dále je třeba si uvědomit, že ve vzájemných interakcích makroekonomických ukazatelů se prosazuje jistá setrvačnost a že tedy situace na trhu práce neodráží vývoj HDP v reálném čase, nýbrž až se značným, zpravidla několikaměsíčním zpožděním. Např. při nástupu recese, kdy HDP klesá, míra nezaměstnanosti se po řadu měsíců chová zdánlivě indiferentně; tento jev je způsobován existencí pracovní legislativy, neumožňující propustit nadbytečné pracovníky okamžitě, ale až po uplynutí příslušné výpovědní doby. Naopak při ekonomickém oživení firmy svůj výstup nejprve zvyšují dokonalejším využitím výrobních kapacit při nezměněné pracovní síle a až teprve

49

s odstupem, daným probíhajícími konkurzy a výběrovými řízeními, začínají nabírat nové pracovníky. Proto se tedy empirická závislost mezery produktu a míry nezaměstnanosti dle Okunova zákona v reálném čase od sebe odchyluje a vede ke zdánlivým disproporcím.

Ze všech těchto důvodů je třeba za hlavní indikátor vývoje přirozené míry nezaměstnanosti považovat ani ne tak spojnicovou křivku v grafu na obr. 4, odrážející údaje z Tab. 2, jako spíše z ní vycházející trendovou linii, jež vykazuje v dlouhodobém, třiadvacet let trvajícím časovém horizontu sestupnou tendenci:

Obr. 4: Vývoj přirozené míry nezaměstnanosti v ČR v letech 1995-2018

Zdroj: Vlastní výpočty

Z grafu je patrné, že přirozená míra nezaměstnanosti se ve sledovaném období snížila z původních cca 7 % na současných 4 až 5 procent. To plně odpovídá jak teoretickým postulátům, tak i realitě pracovního trhu v ČR. Přirozená míra nezaměstnanosti sestávající z nezaměstnanosti strukturální a frikční není nějaká neměnná konstanta; její výše je ovlivňována řadou skutečností. Strukturální nezaměstnanost se od 90. let snížila, na čemž měla nepochybně podíl celá řada skutečností; počínaje schválením zákona o zaměstnanosti č. 435/2004 Sb., kladoucím prioritní důraz na prosazování aktivní politiky zaměstnanosti, přes odvětvové změny české ekonomiky (sektor služeb, který v současnosti dominuje tvorbě HDP, se s výjimkou jazykové vybavenosti pracovníků vesměs vyznačuje nižšími nároky na specializovanou profesní kvalifikaci oproti odvětvím těžkého průmyslu, zejména strojírenství) až po strukturní změny výuky ve školství, v němž se v současnosti klade velký důraz právě na jazykovou přípravu žáků a studentů, požadovanou obzvláště v sektoru služeb.

Co se týče druhé složky přirozené míry nezaměstnanosti, kterou je nezaměstnanost frikční, tak ta se nepochybně v dlouhodobém horizontu snižuje rovněž, což je způsobováno zlepšujícími se informacemi o situaci na pracovním trhu. Zdánlivě by to mohlo být považováno za rozpor s obecně uznávanou tendencí k nárůstu frikční nezaměstnanosti v zemích s výraznými projevy tzv. „sociálního státu“. Důvodem je zejména štědrá sociální politika, vedoucí k dobrovolné nezaměstnanosti, jež

50

představuje podstatnou součást nezaměstnanosti frikční, a dále zneužívání této sociální politiky, kdy formální registrace na úřadu práce a pobírání podpory v nezaměstnanosti je doprovázeno tzv. prací načerno.

Faktem ale je, že oproti zemím jako Francie, Švédsko apod. jsou sociální dávky v České republice neporovnatelně nižší a uvedené negativní jevy, i když samozřejmě existují i zde, tak ale se projevují v daleko menší míře.

4 Diskuze

Model pro výpočet přirozené míry nezaměstnanosti, uplatněný v této stati, vedl k výsledkům, jež jsou sice jen aproximací skutečných hodnot přirozené míry nezaměstnanosti, ale které, jak bylo ukázáno v grafu 4, přesto odrážejí teoretické předpoklady o závislosti přirozené míry nezaměstnanosti na dostupnosti informací o situaci na pracovním trhu a účinnosti uplatňování aktivní politiky zaměstnanosti jako nástroje pro řešení zejména nezaměstnanosti strukturální. Na základě získaných hodnot přirozené míry nezaměstnanosti lze tudíž konstatovat, že dosažené výsledky mají vypovídací hodnotu a jsou dostatečně validní.

Porovnáme-li získané údaje s výsledky v obdobných studiích z poslední doby, zjistíme, že vedou k podobným závěrům.

Podle autorek Kadeřábkové a Jašové (2011) se NAIRU v letech 2001-10 vyvíjela následovně (viz obr. 5):

Obr. 5: Vývoj NAIRU v ČR v letech 2001-2010

Zdroj: Kadeřábková, Jašová (2011, s. 519)

Komparaci zde komplikuje odlišný sledovaný časový interval, ale je zřejmé, že v grafu 5 se v letech 2001-2010 NAIRU snižovala od výchozích cca 6,3 % do konečných cca 5,3 %.

To je v dobré shodě s výsledky dosaženými modelem z tohoto článku, podle něhož ve stejném období přirozená míra nezaměstnanosti klesala z 6,3 % na 5,5 %.

Jiné empirické výsledky přináší výzkum E. Jašové (2010), podle níž NAIRU v letech 1999 až 2008 (s predikcí do r. 2010) nabývala intervalových hodnot, uvedených v obr. 6:

51

Obr. 5: Skutečná míra nezaměstnanosti a NAIRU v ČR v letech 1999-2010

Zdroj: Jašová (2010, s. 21)

Zde jsou výsledky naopak zřetelně odlišné, což má svou primární příčinu v rozdílné metodice, neboť zde nebyla NAIRU odvozována jako spojitá trendová přímka (u modelu z této stati) či křivka (v grafu na obr. 5) v rámci časového intervalu, nýbrž jako množina horizontálních přímek v rámci kratších období.

Nicméně i zde je vidět shoda v tom, že po roce 1999 má NAIRU zřetelně sestupnou tendenci.

Pro dosažení přesnějších výsledků by nicméně bylo možné sestrojit model podstatně náročnější, a to zejména z hlediska předpokladu ad a) v kapitole 1. Metodika. Tedy jinými slovy, vzít v potaz reálný vývoj výnosů z rozsahu ve výrobě (resp. v jednotlivých výrobních odvětvích, neboť ta se od sebe tvarem produkční funkce nepochybně liší), skutečný průběh křivky rostoucího výstupu v daných odvětvích (zejména se zřetelem na to, zda je ta která výroba pracovně či kapitálově náročná), a vývoj mezní produktivity kapitálu.

Takto vytvořený rozvinutý model výpočtu přirozené míry nezaměstnanosti by pro Českou republiku vycházel z údajů o tvorbě HDP (hrubá přidaná hodnota podle odvětví), publikovaných na stránkách ČSÚ, a do metodiky výše uvedeného jednoduchého modelu by vnesl následující korekce:

Namísto rovnice (1) pro výpočet Kompenzované mezery produktu by musela být použita soustava rovnic ve tvaru:

OMP = KO . MPK . X . Z (5)

kde OMP je odvětvová mezera produktu, neboli mezera produktu jednotlivých národohospodářských odvětví očištěná o vliv proměnlivého využití kapitálu v daném odvětví a měnící se míry ekonomické aktivity, KO je meziroční přírůstek kapitálu v daném odvětví, MPK mezní produktivita kapitálu v daném odvětví, vyjádřená jako podíl přírůstků produkce daného odvětví a přírůstku hrubého fixního kapitálu v odvětví za jednotlivá čtvrtletí, X je koeficient kompenzující proměnlivé využití kapitálu na střední hodnotu 84,1 % a Z je koeficient kompenzující v čase se měnící míru ekonomické aktivity na střední hodnotu 64,6 %.

52

Kompenzovaná mezera produktu by pak byla následně určena jako vážený průměr za jednotlivá hospodářská odvětví, přičemž váhou jednotlivých odvětví by byl jejich podíl na tvorbě HDP České republiky.

Rovnice (2), udávající velikost mezery produktu podle Okunova zákona, by teoreticky měla vycházet z konkrétních podmínek dané (české) ekonomiky, s tím, že do rovnice by měla být zakomponována empiricky zjištěná citlivost mezery produktu na fluktuacích trhu práce. Problémem je, že neexistuje ucelená časová řada empirických podkladů pro určení koeficientu lineární závislosti c v rovnici (2), a z toho důvodu by bylo nutné i nadále uvažovat hodnotu koeficientu c ve výši 0,02.

Rovnice (3) by byla použita v nezměněném tvaru, přičemž parametr KMP (kompenzovaná mezera produktu) v rovnici by byl váženým průměrem soustavy rovnic (5).

Aplikace výše popsaného komplexního modelu na výpočet přirozené míry nezaměstnanosti by nesporně byla velmi ambiciózním úkolem, zejména z důvodu mimořádně komplikovaného sběru podkladových dat; volně dostupné údaje o hrubé tvorbě kapitálu jsou totiž k dispozici za ekonomiku jako celek (nikoliv za jednotlivá národohospodářská odvětví). V principu ovšem není nemožné tato data získat.

Závěr

Jednoduchý matematický model, využívající pro výpočet přirozené míry nezaměstnanosti hodnoty mezery produktu, vede k výsledkům potvrzujícím teoretické předpoklady založené na důvodech nezaměstnanosti frikční i strukturální. Frikční nezaměstnanost může být snížena lepšími informacemi o situaci na pracovním trhu (vliv online informací dostupných na internetu je ve sledovaném období 1995-2018 neoddiskutovatelný, navíc vznikly specializované zprostředkovatelské pracovní agentury), strukturální celou řadou nástrojů aktivní politiky zaměstnanosti (rekvalifikace, veřejné práce, podpora malého soukromého podnikání apod., přičemž na aktivní politiku zaměstnanosti se v současné době klade nesrovnatelně větší důraz, než v 90. letech). Přestože se jedná o model založený na řadě zjednodušujících postulátů, vzhledem k dlouhodobé datové řadě zahrnující období 23 let se výstupy získané jeho prostřednictvím jeví jako dostatečně validní.

Další zpřesňování dosažených výsledků by bylo možné, pokud by faktory pokládané v prezentovaném modelu axiomaticky za konstanty, jako typ výnosů z rozsahu, shoda mezní produktivity práce a kapitálu apod. byly uvažovány jako proměnné. Ani takovýto komplexní model by však nic zásadního nezměnil na vypovídací schopnosti údajů zjištěných modelem jednoduchým, pouze by umožnil dokonaleji specifikovat údaje zjištěné v této stati.

Reference

Český statistický úřad (2018). Míry zaměstnanosti, nezaměstnanosti a ekonomické aktivity - říjen 2018. Dostupné na https://www.czso.cz/csu/czso/cri/miry-zamestnanosti-nezamestnanosti-a-ekonomicke-aktivity-rijen-2018 [5.12.2018].

Dornbusch, R., Fischer. S. (1994). Makroekonomie. Praha: SPN a Nadace Economics.

Friedman, M. (1968): The Role of Monetary Policy. American Economic Rewiew. (1), s. 1-17.

Helísek, M. (2000). Makroekonomie. Slaný: Melandrium.

53

Jašová, E. (2010). Míra nezaměstnanosti neakcelerující inflaci a hospodářský cyklus v prostředí České republiky - historie a možný vývoj do konce roku 2010. Working Papers VŠE Praha, 12/2010

Kadeřábková, B., Jašová, E. (2011). Analýza ukazatele NAIRU na sektorové úrovni. Politická ekonomie (4), s. 508-525.

Mankiw, G. (2003). Macroeconomics. New York: Worth Publishers.

McAdam, P., McMorrow, K. (1999). The NAIRU Concept - Measurement Uncertainties, Hysteresis and Economic Policy Role. ECE Papers No. 136, IX/1999.

Ministerstvo financí ČR (2018). Makroekonomická predikce České republiky - listopad 2018. Dostupné na https://www.mfcr.cz/cs/verejny-sektor/makroekonomika/makroekonomicka-predikce/2018/makroekonomicka-predikce-listopad-2018-33386 [5.12.2018]

Mindorf, P. (1985). Unemployment. Cause and Cure. Oxford: Blackwell.

Modigliani, F., Papademos, L. (1975). Targets for Monetary Policy in the Coming Year. Brookings Papers on Economic Activity, s. 141-165.

Okun, A. (1962): Potential GNP: Its Measurement and Significance. The Political Economy of Prosperity, Proceedings of ASA, Cowles Foundation Paper 190.

Pošta, V. (2008). NAIRU a přirozená míra nezaměstnanosti - teoretický pohled. Výzkumná studie č. 1/2008. Ministerstvo financí ČR.

Tobin, J. (1980). Stabilization Policy Ten Years After. Brookings on Economic Activity, (1), s. 19-71.

Kontaktní adresa

RNDr. Ing. Jiří Dobrylovský, Ph.D. Vysoká škola ekonomie a managementu, katedra ekonomie Nárožní 2600/9A, 158 00 Praha 5 E-mail: [email protected] Tel. číslo: +420223003845


54

WORKING CAPITAL AND ITS IMPACT ON BUSINESS PERFORMANCE

Irena Honková

Abstract: A number of studies indicate that the decision on the level of working capital affects business performance. Managing optimal working capital brings businesses sustained reduction of working capital, improves liquidity and increases profitability through effectively managed capital. The aim of this paper is to conduct empirical research of Czech companies on the relationship between working capital and the performance of a company. An OLS model is used that can calculate quadratic regression for calculating the optimum amount of working capital. Models are created that describe the dependency of earnings on the components that determine working capital. By deriving the quadratic function of performance, an inflection point is found, indicating a theoretical net trade cycle of 134 days for Czech companies with inventory and 70 days for Czech companies with no inventory. However, there is a level of working capital at which a higher working capital begins to be negative in terms of profitability due to the additional interest expenses and, hence, the higher probability of bankruptcy and credit risk of firms. Thus, firm managers should aim to keep as close to the optimal level as possible and try to avoid deviations from it that destroy profitability.

Keywords: Working Capital, Net Trade Cycle, Ordinary Least Squares (OLS) Method

JEL Classification: M21.

Introduction

Research conducted by the PricewaterhouseCoopers consultancy (PwC, 2012) highlighted the inefficient management of working capital in the practice of European businesses. The research was based on comparing working capital with a percentage of the sales performance of businesses, and included the 4000 largest European companies from 34 countries, including the Czech Republic, during the years 2007 - 2011. The research found that the largest European businesses lost about 400 billion EUR due to inefficient management of working capital. This is equivalent to about a third of their earnings. Companies in southern Europe could gain up to 40% of their earnings, with British and Irish companies gaining 22%. The results confirmed that, on the contrary, companies that are able to manage working capital effectively are doing better.

The authors Madhou, A., Moosa, I. & Ramiah, V. (2015), Kim and Chung (1990) seem to think that the decision about the amount of working capital affects corporate performance. Managing an optimal working capital provides businesses with (PWC, 2007) a sustained reduction of working capital, improving liquidity and increasing return on assets through effectively managed capital.


Terminology in the definition of working capital is unclear, as different Czech authors often use two terms – “working capital” and “net working capital” - interchangeably.

In the narrower sense, some, such as Hrdý & Krechovská (2013), define working capital as current assets and net working capital after deducting short-term liabilities.

55

Due to the terminology of foreign authors, the term working capital will be used in this paper for the difference between current assets and short-term liabilities.

As has been said, working capital is defined as the difference between current assets and short-term liabilities. It is a part of long-term financial resources that is used to cover short-term assets. It serves to finance the normal day-to-day operation of a business (Jindřichovská, 2013).

Investments in receivables and inventories represent a significant proportion of a company's assets, while liabilities to suppliers are a significant source of funding for most businesses. Research (Honková, 2016) carried out in 2014 found that 48 % of the total capital were commitments to suppliers. In the UK, the business loan represents about 41 % of the total debt of medium-sized enterprises (Baňos, 2014). Working capital should achieve positive values because of the positive effect on the liquidity of a company (Fazzari & Petersen, 1993). According to the authors Blinder and Maccini (1991), larger inventories can minimize delivery costs and price fluctuations and act as a means of preventing production breaks and losses due to lack of material or products. It enables businesses to provide better customer service and protects from increased production costs (Schiff& Lieber, 1974). Kroflin & Kratz (2015) traditionally show high attention to liquidity. Authors Ding, S., Guariglia, A. & Knight, J. (2013) display high sensitivities of investment in working capital to cash flow and low sensitivities of investment in fixed capital to cash flow. On the other hand, if businesses provide trade credit, they can increase their sales (Brennan, Maksimovic, & Zechner, 1988); trade credit enables customers to check and use the goods or services prior to payment and strengthens supplier-customer relationships (Wilner, 2000, Smith & Tirell, 1987). Trade credit also reduces the information inequality between the buyer and the seller. Shipley and Davis (1991) believe that trade credit is also an important criteria for choosing a supplier if it is otherwise difficult to differentiate the product.

The positive level of working capital has, of course, a However, working capital is not purely maximizing, as there are arguments that criticize too high a working capital. Storing too many goods increases warehouse and inventory insurance costs (Kim & Chung, 1990). A large amount of working capital raises the need for additional capital and therefore additional capital costs. Interest costs and credit risk are increasing. (Kieschnick, LaPlante, & Moussawi, 2011). There are also financial constraints that complicate the financing of working capital. Fazzari and Petersen (1993) believe that investments in working capital are more sensitive than fixed capital investments. Hill, Kelly and Highfield (2010) show that businesses with lower financial constraints achieve higher levels of working capital. Financial constraints include paid dividends (Faulkender and Wang, 2006), size and interest coverage (Whited, 1992), cash flow (Moyen 2004) and financial costs (Fazzari, Hubbard & Petersen, 1988).

There is also a conflict between liquidity and return. Increasing short-term liabilities leads to a reduction of liquidity; on the other hand, short-term liabilities have lower financing costs than long-term instruments. Flexibility is definitely an advantage of short-term commitments, with another advantage being lower funding costs. The disadvantage is the fact that their costs are not fixed in the long term and that they may sometimes be unavailable. (Jindřichovská, 2013)

56

Thus, it can be generally said that the more the enterprise invests in working capital, the greater the liquidity, however the lower the return on assets. If we were looking at return on assets as a determinant of improving the performance of a business, we could at first glance say that the lower the working capital, the more the enterprise increases its value. However, efforts to minimize working capital could lead to many problems.

The studies listed in Tab.1 show the relationship between working capital and return on assets.

Tab. 1: Studies focusing on the relationship between working capital and return on assets

Author, Country Period Profit measure

Results NTC on profitability

Jose et al. (1996), USA 1974-1993 ROE NTC↓ Shin & Soenen (1998), USA 1975-1994 ROA NTC↓ Wang (2002), Japan 1985-1996 ROE NTC↓ Deloof (2003), Belgium 1992-1996 Gross oper.

profit NTC→

Lazaridis &Tryfonidis (2006), Greece

2001-2004 Gross oper. profit

NTC↓

Meyer & Lüdtke (2006), Germany 2003 ROCE n/a García-Teruel & Martínez-Solano (2007), Spain

1996-2002 ROA NTC↓

Raheman & Nasr (2007), Pakistan 1999-2004 Net oper. profit NTC ↓ Karaduman et al. (2010), Turkey 2005-2008 ROA NTC↓ Banos & Caballero et al. (2012), Spain

2002-2007 ROA Concave

Enqvist et al. (2012), Finland 1990-2008 Gross oper. profit

NTC↓

Wöhrmann et al. (2012), Germany 2007-2010 ROCE NTC↓ Source: (own)

Working capital management studies usually have two views. The first view states that higher amounts of working capital allow companies to increase their sales and therefore their business performance. Authors Fazzari and Petersen (1993) suggest in their analysis, that investing in working capital is more sensitive to funding limitations than investing in fixed capital. Alternative studies state that higher working capital needs increased funding and hence other financial costs that reduce profitability and can lead to bankruptcy. (Wang, 2002, Faulkender & Wang, 2006) The combination of these positive and negative effects of working capital on the performance of a business leads to the prediction that there is no direct linear relationship between the working capital and the value of a business. It can be assumed that the reduction of working capital may increase the performance of a business to some extent, but after achieving its optimum, further reduction of working capital would lead to a reduction in the value of a business. Therefore, it can be assumed that the influence of working capital on the performance of a company is a function in the form of the inverted U, i.e. a concave quadratic function, by means of which an optimal level of working capital can be found. Thus, raising working capital to a certain level increases the company's performance, but once

57

the optimal level is reached, the relationship between working capital and business performance becomes negative.

In terms of performance measurement, we turn to the return on equity (ROE) indicator, based on the Du point analysis. There is a model that calculates the performance of a company, primarily in terms of working capital from Shin & Soenen (1998) and others by Agrawal & Knoeber (1996), Himmelberg, Hubbard & Palia (1999), Thomsen, Pedersen & , Kostakis & Ozkan (2009) and Wu (2011):

𝑄 , 𝛽 𝛽 𝑁𝑇𝐶 , 𝛽 𝑁𝑇𝐶 , 𝛽 𝑆𝐼𝑍𝐸 , 𝛽 𝐿𝐸𝑉 , 𝛽 𝑅𝑂𝐴 𝜆 𝜂 𝜀 , (1)

where:

Qit is business performance as expressed by the ROE indicator.

Working capital is expressed in this model using the net trade cycle (NTC) (Shin & Soenen (1998). Because of the mentioned quadratic function, the NTC is considered in the first and second powers. Other variables that affect business performance are according to the previous mentioned researches: size (SIZE), leverage (LEV) and return on assets (ROA).

The NTC is calculated as (accounts receivable/sales) * 365 + (inventories/purchases) *365 – (accounts payable/purchases) * 365. Thus, NTC deals with the management of accounts receivable, the management of inventories and the trade credit received, with a shorter NTC indicating a more aggressive working capital policy. Previous publications report the importance of considering there three components at the same time because they influence each other as well as profitability and value. Schiff and Lieber (1974), for example, indicate the importance of taking into account the interrelationship between inventory and accounts receivable policies. We regress the profitability and against NTC and its square. The inclusion of these two variables allows us to test both the profitability and risk effects. The quadratic relation proposed in Eg.1 presents and breakpoint that can be derived by differentiating the firm´s profitability variable with respect to the NTC variable and making this derivative equal to 0. Since we expect NTC and corporate performance to relate positively at low levels of working capital and negatively at higher levels, the hypothesis is that β2 is negative, because it would indicate that firms have an optimal working capital level that balances the costs and benefits of holding working capital and maximizes their performance.

We measure firm size (SIZE) as the natural logarithm of sales; leverage (LEV) by the ratio of total debt to total assets; and the measurement of return on assets (ROA) is through the ratio earnings before interest and taxes over total assets. The parameter λt is a time dummy variable that aids to capture the influence of economic factors (e. i. GDP, inflation, unemployment, political and legislative background) that may also affect corporate performance but with companies cannot control. ηi is the unobservable heterogeneity or the firm´s unobservable individual effects, so we can control for the particular characteristics of each firm. Finally, εi,t is the random disturbance. We also control for industry effects by introducing industry dummy variables.

The aim of this paper is to determine the coefficient β0 and the variables β1 - β5, thus creating a model of influence of working capital on the company's performance.

In contrast to Shin & Soenen (1998), Agrawal and Knoeber (1996), Himmelberg, Hubbard & Palia (1999), Thomsen, Pedersen & Kvist, models are created separately for

58

no-inventory businesses and for inventory-holding businesses, as the most significant component of NTC's working capital is substantially influenced by inventory. The aim is to create a profit maximization model using an optimal level of net trade cycle.

2 Methods

The data in this paper is from the MagnusWeb database. It is a database of Czech companies of 2016 and 2015. All enterprises that keep their accounts and have published their financial statements in the monitored years were included in the research. This is a significant indicator for calculating working capital. The companies were divided according to whether or not they held inventory. A sample of companies with inventory was 3254. A sample of companies with no inventory was 1190. Due to incomplete data, this number was reduced to 1953 companies with inventory and 958 companies without inventory. Due to the sufficient number of samples, the assumption of the normality of distribution was assumed.

The data was edited in MS EXCEL and subsequently exported to the GRETL program, where statistical calculations were performed.

As mentioned in the previous chapter, the function of the influence of working capital on the company's performance is a function in the form of inverted U, i.e. a concave quadratic regression. Quadratic regression is the case of polynomial regression where the degree of polynomial Pk is equal to two. As such, this is a special case of linear regression. The set of values is interleaved (approximated) by the quadratic function (parabola). Polynomial coefficients (parabola) are calculated using the Ordinary Least Squares (OLS) method.

Data are interleaved with a parabola, or a second order polynomial P 2(x)= ax2+bx+c. The sum of deviations squares ei = yi − P2(xi) function F depends on parameters a, b, c in particular β0-β5. The minimum of the functional F can be found by partial derivations (at the local extreme are equal to zero):

0 (2)

The multiple regression model works with more than one explanatory variable, in our case (1). The task is to estimate parameters β0, β1, β2, β3, β4, β5. These parameters indicate how much, on average, the estimated value Qi,t increases, if an explanatory variable increases by one, assuming that the other variables do not change.

3 Problem solving

Using the OLS method, multiple regression was calculated , where the dependent variable was the ROE indicator representing the performance of the business. Equations (3) and (4) represent a function of profit depending on the items that determine working capital. Equation (3) is for businesses with inventory and equation (4) for businesses with no inventory. Companies with inventory put a greater emphasis on NTC and ROA, companies with no inventory on size and financial leverage.

𝑄 , 54,5751 0,0205𝑁𝑇𝐶 , 7,6146 𝑒 0,5 𝑁𝑇𝐶 , 3,7335 𝑆𝐼𝑍𝐸 ,

0,0773 𝐿𝐸𝑉 , 0,7162 𝑅𝑂𝐴 𝜆 𝜂 𝜀 ,

(3)

59

𝑄 , 67,0733 0,0146𝑁𝑇𝐶 , 1,0336 𝑒 0,4 𝑁𝑇𝐶 , 5,6722 𝑆𝐼𝑍𝐸 ,

0,3284 𝐿𝐸𝑉 , 0,1794 𝑅𝑂𝐴 𝜆 𝜂 𝜀 ,

(4)

Except the sq_NTC variable and the coefficient β0, all variables are positively dependent on the profit, which means that the profit increases with increasing variables.

We test the hypothesis β = 0, that the model generally explains the ROE well. Critical t-test values fall into critical areas in the following cases: for businesses with inventories for all indicators except for the ROA indicator, for non-inventory enterprises these are const, NTC and sq_NTC constants (Tab 2 and Tab. 3).

Tab. 2: OLS Statistical Values - Businesses with inventories

The mean value of the dependent variable −0,230163

Determinative deviation of the dependent variable 473,1739

Sum of squares of residues 7,23e+08

Determinative error of regression 471,9309

Determination coefficient 0,006776

Adjustment Determination Coefficient 0,005247

F(5,) 4,431652

P-value(F) 0,000503

Logarithm of credibility −24648,56 Source: own

Tab. 3: OLS Statistical Values - Businesses without inventories

The mean value of the dependent variable 6,650177

Determinative deviation of the dependent variable 186,4057

Sum of squares of residues 31557071

Determinative error of regression 182,0664

Determination coefficient 0,050999

Adjustment Determination Coefficient 0,046015

F(5,) 10,23211

P-value(F) 1,44e-09

Logarithm of credibility −6342,122 Source: own

ANOVA scattering analysis contains estimated values using the SSR model, unexplained random portion (SSE), and total SST sum (Tab 4. a Tab. 5). The created model is better for businesses with no inventory.

60

Tab. 4: ANOVA Analysis – Businesses with inventories

Sum of squares df Middle quadrate

Regression (SSR) 4,93506e+006 5 987012

Residue (SSE) 7,23391e+008 3248 222719

Full (SST) 7,28326e+008 3253 223894 Source: own

Tab. 5: ANOVA Analysis – Businesses without inventories

Sum of squares df Middle quadrate

Regression (SSR) 1,69588e+006 5 339176

Residue (SSE) 3,15571e+007 952 33148,2

Full (SST) 3,3253e+007 957 34747,1 Source: own

The problem that arises when testing model parameters is multi-collinearity. It results from a correlation between independent variables. The easiest way to determine if there is a strong correlation between the variables is to look at the correlation matrix. As can be seen in our model (Tab. 6 and Tab. 7), the values are not high as to not cause multi-collinearity problems. According to (Koop, 2008) the value is binding if there is a correlation in an absolute value greater than 0.9.

Tab. 6: Correlation Analysis - Businesses with inventories ROA ROE Leverage l_SIZE NTC

1,0000 0,0768 0,0007 0,0770 0,1006 ROA

1,0000 0,0159 0,0236 0,0084 ROE

1,0000 0,0219 -0,0054 Leverage

1,0000 0,1356 l_SIZE

1,0000 NTC Source: own

Tab.7: Correlation Analysis - Businesses without inventories ROA ROE NTC Leverage l_SIZE

1,0000 0,2013 0,0484 0,0026 0,0791 ROA

1,0000 0,0331 0,0309 0,0804 ROE

1,0000 -0,1011 0,0554 NTC

1,0000 0,0169 Leverage

1,0000 l_SIZE Source: own

T-test is used to test the significance of variables (Tab. 8 and Tab. 9). It tests whether it is possible to reject the hypothesis that β = 0 and therefore say that statistically, the given coefficient is not insignificant at the chosen materiality level.

61

Tab. 8: T-test Businesses with inventories

Variable Coefficient 95 confidential interval

Const -54,5761 (-128,648, 19,4960)

ROA 0,716165 (0,397802, 1,03453)

l_SIZE 3,73353 (-1,84715, 9,31420)

NTC 0,0204506 (-0,0244794, 0,0653806)

sq_NTC -7,61461e-007 (-2,11102e-005, 1,95873e-005)

Leverage 0,0773368 (-0,123213, 0,277887) Source: own

Tab. 9: T-test Businesses without inventories

Variable Coefficient 95 confidential interval

const -67,0733 (-119,024, -15,1225)

ROA 1,17940 (0,808187, 1,55061)

Leverage 0,328362 (0,0418209, 0,614904)

NTC 0,0146489 (-0,00741492, 0,0367126)

l_SIZE 5,67215 (1,05438, 10,2899)

sq_NTC -1,03395e-006 (-1,10953e-005, 9,02744e-006) Source: own

4 Discussion

How above mentioned, working capital should achieve positive values but it is not purely maximizing because of additional costs (Kim and Chung, 1990) (Kieschnick, LaPlante, & Moussawi, 2011). On the one hand, there is liquidity, continuity of production and lower financial constraints and on the other hand profitability. Both of these influences act against one another. Working capital includes return on assets (ROA) and liquidity. Both of these influences act against one another. Thus, the ROE profit maximization function with working capital is a concave quadratic function. The more the company invests in short-term assets, the better the liquidity of the company. The company may invest in short-term assets to improve its liquidity - cash and securities - but these assets generate a small return. Therefore, the company can reduce the risk of insolvency (low liquidity) only by reducing return on assets and vice versa (Jindřichovská, 2013).

Since we expect NTC and corporate performance to relate positively at low levels of working capital and negatively at higher levels, the hypothesis is that β2 is negative, because it would indicate that firms have an optimal working capital level that balances the costs and benefits of holding working capital and maximizes their performance. Models (3) and (4) that were developed describe the dependence of profit (ROE) on components determining working capital (net trade cycle, size, leverage, ROA). The concavity of this function was proved by negative parameters in models (3) and (4).

62

Model (1) allows us to determine when profit maximization occurs by derivation the inflection point becomes as -β1 / 2β2. For the model (3) "Businesses with inventories" it is calculated as:

𝑁𝑇𝐶 ,

∗ ,134 𝑑𝑎𝑦𝑠 (5)

a for the model (4) “Businesses without inventories” it is calculated as:

𝑁𝑇𝐶 ,

∗ ,71 𝑑𝑎𝑦𝑠 (6)

For model (3) – “businesses with inventories”, the optimal level of net trade cycle is calculated as 134 days. For model (4) “businesses with no inventory”, the optimal level of net trade cycle is calculated as 70 days.

Conclusion

The aim of this paper was to conduct empirical research into the relationship between working capital and business performance (ROE). A number of studies, generally agreeing that the amount of working capital affects the performance of a business, researches the relationship between the amount of working capital and business performance. Model (1) shows this dependence. The variables include net trade cycle, size, leverage and ROA of the business. On a representative sample of Czech companies, a model (3) was created for enterprises that do not have stocks and a model (4) for enterprises with stocks. These models show the dependence of working capital, enterprise size, indebtedness and ROA profitability on the resulting ROE profitability. While the dependence of these variables on ROE is statistically insignificant for the purposes of ROE maximization (since ROE depends on many other variables), the data from these models were used to calculate the optimum working capital, ie, net trade cycle at which highest profitability (ROE).

The OLS model was used, which calculated a quadratic regression of calculating the optimal amount of working capital.

The developed models (3) and (4) describe the dependence of profit on components determining working capital. However, it was calculated that except the sq_NTC variable and the coefficient β0, all variables are positively dependent on the profit, which means that the profit increases with increasing variables. The negative value sq_NTC has confirmed the concave shape of this quadratic working capital function. The inflection point was derived from this quadratic function, indicating a theoretical net trade cycle of 134 days for Czech companies with inventory and 70 days for Czech companies with no inventory.

However, there is a level of working capital at which a higher working capital begins to be negative in terms of profitability due to the additional interest expenses and, hence, the higher probability of bankruptcy and credit risk of firms. Thus, firm managers should aim to keep as close to the optimal level as possible and try to avoid deviations form it that destroy profitability.

Results suggest that managers should be concerned about working capital level because it seems to affect the corporate performance.

63

References

Agrawal, A. & Knoeber, C.R. (1996). Firm performance and mechanisms to control agency problems between managers and shareholders. Journal of Financial and Quantitative Analysis, 31, pp. 377-397.

Banos-Caballero, S., Garcia-Teruel, P.J., & Martinez-Solano, P. (2014). Working capital management, corporate performance and financial constraints. Journal of Business Research, 67, pp. 332-338.

Banos-Caballero, S., Garcia-Teruel, P.J., & Martinez-Solano, P. (2012). How does working capital management affect the profitability of Spanish SMEs? Small Bus Econ, 39 (1), pp. 517-529.

Blinder, A. S. & Maccini L.J. (1991). The resurgence of inventory research: What have we learned? Journal of Economic Surveys, 5, pp. 291-328.

Brennan, M., Maksimovic, V. & Zechner, J. (1988). Vendor financing. Journal of Finance, 43, pp. 1127-1141.

Dellof, M. (2003). Does working capital management affect profitability of Belgian firms? Journal of Business Finance and Accounting, 30, pp. 573-587.

Ding, S., Guariglia, A. & Knight, J. (2013). Investment and financing constraints in China: Does working capital management make a difference? Journal of Banking & Finance, 37, pp. 1490-1507.

Enqvist, J., Graham, M., & Nikkinen, J. (2012). The impact of working capital management on firm profitability in different business cycles: evidence from Finland. Working paper.

Faulkender, M. & Wang, R. (2006). Corporate financial policy and the value of cash. Journal of Finance, 61, pp. 1957-1990.

Fazzari, S. M., & Petersen, B. (1993). Working capital and fixed investment: New evidence on financing constraints. The Rand Journal of Economics, 24, pp. 328-342.

Fazzari, S.M., Hubbard, R. G., & Petersen, B.C. (1988). Financing constraints and corporate investment. Brookings Papers on Economic Activity, 1, pp. 141-195.

Florackis, C., Kostakis, A., & Ozkan, A. (2009). Managerial ownership and performance. Journal of business Research, 62, pp. 1350-1357.

García-Teruel, P.J. & Martínez-Solano, P. (2007). Effects of working capital management on SME profitability. International Journal of Managerial Finance, 3, pp. 164-177.

Hill, M. D., Kelly, G. & Highfield, M. J. (2010). Net operating working capital behaviour: A first look. Financial Management, 39, pp. 738-805.

Himmelberg, C., Hubbard, R. & Palia, D. (1999). Understanding the determinants of managerial ownership and the link between ownership and performance. Journal of Financial Economics, 53, pp. 353-384.

Honkova, I. (2016). Use of External Sources of Financing in the Construction Industry. Scientific papers of the University of Pardubice. Pardubice, 1(36), pp. 42-54.

Hrdý, M. & Krechovská, M. (2013). Podnikové finance v teorii a praxi. Praha: Wolters Kluwer ČR.

Hrdý, M. & Strouhal, J. (2010). Finanční řízení. Praha: Wolters Kluwer ČR.

Jindřichovská, I. (2013). Finanční management. Praha: C.H.Beck.

Jose, M. L., Lancaster, C., & Stevens, J. L. (1996). Corporate return and cash conversion cycles. Journal of Economics and Finance,20, 33-46.

Karaduman, H. A., Akbas, H. E., Ozsozgun, A., & Durer, S. (2010). Effects of working capital management on profitability : the case of selected companies in the Istanbul Stock Exchange. International Journal of Economics and Finance Studies, 2, pp. 47-54.

Kieschnick, R., LaPlante, M. & Moussawi, R. (2011). Working capital management and shareholder wealth. [online] Working paper. Available at: http://ssrn.com/abstract-1431165 [accessed 13.04.2018].

64

Kim, Y.H. & Chung, K. H.(1990). An integrated evaluation of investment in inventory and credit: A cash flow approach. Journal of Business Finance & Accounting, 17, pp. 381-390.

Knauer, T. & Wöhrmann, A. (2013) Working capital management and firm profitability. Journal Management Control, 24, pp. 77-87.

Koop, G. (2008). Introduction to Econometrics, 1 ed. John Wiley & Sons.

Kroflin, P. & Kratz, N. (2015). Working capital management as a routine: An action based access to the topic. BEH – Business and Economic Horizons, 11 (3), pp. 173-182.

Lazaridis, I. & Tryfonidis, D. (2006). Relationship between working capital management and profitability of listed companies in the Athens Stock Exchange. Journal of Financial Management and Analysis, 19, pp. 26-35.

Madhou, A., Moosa, I. & Ramiah, V. (2015). Working Capital as a Determinant of Corporate Profitability. Review of Pacific Basin Financial Markets and Policies, 18(4).

Meyer, S. & Lüdtke, J. P. (2006). Der Einfluss von Working Capital auf die Profitabilität und Kreditwürdigkeit von Unternehmen. Finanz Betrieb, 8, pp. 609-614.

Moyen, N. (2004). Investment-cash flow sensitivities: constrained versus unconstrained firms. Journal of Finance, 59, pp. 2061-2092.

PricewaterhouseCoopers (PwC). (2007). Řízení pracovního kapitálu. [online] Best Practice. Available at: http://www.pwc.com/cz/cs/podnikove-poradenstvi/provozni-aktivity/best-practise-rizeni-pracovniho-kapitalu.pdf [accessed 13.04.2018].

PricewaterhouseCoopers (PwC). (2012). Evropské firmy neumí řídit svůj pracovní kapitál - Za posledních pět let je to stálo zhruba 10 biliónů Kč. [online] Available at: http://www.pwc.com/cz/cs/tiskove-zpravy/2012/evropske-firmy-neumi-ridit-svuj-pracovni-kapital.jhtml [accessed 13.04.2018].

Raheman, A. & Nasr, M. (2007). Working capital management and profitability – case of Pakistani firms. International Review of Business Research Papers, 3, pp.279-300.

Schiff, M., & Lieber (1974). A model for integration of credit and inventory management. The Journal of Finance, 29, pp. 133-140.

Schmidt, R. & Tirell, M. (1997). Financial systems, corporate finance and corporate governance. European Financial Management, 3, pp. 333-361.

Shin, H. H. & Soenen, L. (1998). Efficiency of working capital and corporate profitability. Financial Practice &Education, 8, pp. 37-45.

Shipley, D. Davis, L. (1991). The role and burden-allocation of credit in distribution channels. Journal of Marketing Channels, 1, pp. 3-22.

Thomsen, S., Pedersen, T., & Kvist, H. (2006). Blockholder ownership: Effects on firm value in market and control based governance systems. Journal of Corporate Finance, 12, pp. 246-269.

Van Horne, J. C., Wachowicz, J. M. (2005). Fundamentals of financial management. England: Harlow.

Wang, Y.J. (2002). Liquidity management, operating performance, and corporate value: Evidence from Japan and Taiwan. Journal of Multinational Financial Management, 12, pp. 159-169.

Whited, T. M. (1992). Debt, liquidity constraints, and corporate investment: Evidence from panel data. Journal of Finance, 47, pp. 1425-1460.

Wilner, B. S. (2000). The exploitation of relationship in financial distress: The case of trade credit. Journal of Finance, 55, pp. 153-178.

Wöhrmann, A., Knauer, T., & Gefken, J. (2012). Kostenmanagement in Krisenzeiten: Rentabilitätssteigerung durch Working Capital Management? Zeitschrift für Controlling and Management (Sonderheft 3), 56, pp. 80-85.

65

Wu, H. L. (2011). Can minority state ownership influence firm value? Universal and contingency views of its governance effects. Journal of Business Research, 64, pp. 839-845.

Contact Address

Ing. Irena Honková, Ph.D. University of Pardubice, Faculty of Economics and Administration, Institute of business economics and management Studentská 95, 532 10 Pardubice, Czech Republic Email: [email protected] Phone number: +420466036163


66

SELECTED DETERMINANTS OF TOURISM DEMAND – PANEL COINTEGRATION ANALYSIS

Tomáš Jeřábek

Abstract: The aim of the paper is to investigate the effect of income of tourists, tourism price, trade openness and international outbound tourists in tourism origin countries on the tourism demand to the Czech Republic as well as to South Moravia Tourist Region of the Czech Republic from tourists coming from thirteen different countries is analysed. This paper applies panel co-integration analysis through a common correlation effect model (CCE). According to the obtained estimation results, income and trade openness have a significant positive influence on international arrivals to the Czech Republic, while for tourism price the results indicate that it has a negative influence on inflows, as expected. The effect of international outbound tourists on tourism demand in the Czech Republic is negative, against expectations, but it is not statistically significant. For tourism demand in the SMT region, income has a significant positive influence on international arrivals to the South Moravia tourist region, as it regards relative price, the negative sign of the coefficient is the same as the expected one. The effects of trade openness and international outbound tourists on tourism demand in the SMT region have expected positive signs but they are not statistically significant.

Keywords: Tourism Demand, Panel Cointegration Analysis, Real Exchange Rate, Gross Domestic Product, Trade Openness, International Outbound Tourists.

JEL Classification: F14, F47, Z3.

Introduction

Modelling and forecasting tourism demand has a long history. However, due to the development of a mixed autoregressive model (ARIMA) for the analysis of one-dimensional time series, it has started more pronounced since the 1970s. In the 1990s, ARIMA models added more advanced techniques for analysing multidimensional time series, namely vector autoregressive (VAR) models, error correction models (ECM), and vector ECM (VECM). One of the main advantages of these techniques is the ability to analyse causal relationships between the tourism demand as a dependent variable and its influential factors - explanatory variables. The applications of all these models can be seen in studies of the last decade. For example, Bonham et al. (2009) apply VECM to predict tourism in Hawaii. Chaiboonsri et al. (2009) through VECM seek to find short-term and long-term relationships between foreign tourists traveling to Thailand and economic variables such as international arrivals, gross domestic product, goods and services, transport costs and exchange rates. Gautam (2014), using cointegration techniques and the VECM model, examines long-term relationships between economic growth and tourism development in Nepal.

The above studies explore data in the form of time series that are capable of delivering a large number of results, but many observations remain hidden, such as more complicated spatial dependencies. The solution is to use so-called panel data. Panel data is spatial data that is repeatedly retrieved over multiple time periods. The

67

exploration of long-term relationships within panel data is the main purpose of this text. The paper represent a little-used approach to tourism demand analysis based on panel cointegration analysis. For example, Falk (2015) applied a panel ECM model to analyse the impact of the euro’s depreciation on the Swiss overnight stays in selected Austrian ski resorts. The author confirms the higher sensitivity of tourism demand to the development of the analysed exchange rate, in other words, Swiss visitors to Austrian ski resorts are strongly affected by CHF/EUR exchange rates. Ongan et al (2017) investigates the effects of real exchange rates and income on tourist arrivals from seven countries to the USA. According to the test results, the effect of income as a measure of GDP as well as the effect of RER on the number of tourists arriving in the USA is positive.

The aim of the paper is to investigate the effect of income of tourists, tourism price, trade openness and international outbound tourists in tourism origin countries on the tourism demand to the Czech Republic as well as to South Moravia Tourist Region of the Czech Republic by foreign tourists arriving from selected European, Asian and American countries. The rest of the text is divided into four parts. The first part provides a short problem statement. The second part presents the model apparatus used in the paper. In the third part, empirical model and results are presented. The fourth part provides discussion of results.


Knowledge of factors influencing the tourism demand in the Czech Republic is crucial for both managers and policy makers. The best known determinants are income and price variables. Income is a key variable in demand models. As an income measure, Gross Domestic Product (GDP) and Gross National Product, in nominal or real terms and in their per capita form, can all be considered, see Song and Li (2008). Important price variable is the exchange rates between the origin country and alternative destinations. The real exchange rate (RER) is defined as the ratio of the price level abroad (the Czech Republic) and the domestic (tourism origin country) price level, where the foreign price level is converted into domestic currency units via the current nominal exchange rate; see Li et al. (2005) or Seo et al. (2009) for RER applications in demand tourism models. Compared to previous similar studies, the harmonized consumer price index (HCPI) for hotels and restaurants is used to derive the real exchange rate instead of the general CPI that may not capture the price level of products and services that tourists often encounter.

For the Czech Republic, econometric models were applied in the studies of Babecká (2013) and Indrová et al. (2017). International tourism demand was measured in both studies through attendance at collective accommodation facilities. Babecká (2013) analyses the international tourism demand in the Czech Republic through the gravity model. Real GDP of tourism origin countries, real exchange rate and the distance of origin country from the Czech Republic and five dummy variables were considered as the tourism demand determinants. Indrova et al. (2015) monitored the impact of the development of income and price variables on the tourism demand in the Czech Republic. For this purpose, they used a dynamic panel with Arellano-Bond GMM estimator. Real GDP per capita for origin countries, RER, trade openness,

68

lagged value of tourism demand as dependent variable and two dummy variables are considered as determinants.

The previous papers did not take into account the cross-sectional dependences, i.e. unobserved time-specific heterogeneity which makes error terms contemporaneously colerated. In tourism demand context, the cross-sectional dependence can be induced by the presence of common shocks affecting tourism demand from observed countries and this can lead to different country-specific responses to these shocks. This approach has not been used in the tourism literature where cross-sectional dependence is usually ignored. However, cross-sectional units are almost certainly influenced by national or global shocks such as business cycles or national fiscal and monetary policies. Ignoring cross-sectional dependence of errors can have serious consequences.

This study examines the effect of four determinants on tourism demand from the France (FR), Austria (AT), the Netherlands (NL), Germany (DE), the Slovak Republic (SK), Spain (ES), Italy (IT), Sweden (SE), Israel (IL), Japan (JP), USA (US), Poland (PL) and Hungary (HU) to the Czech Republic and to the South Moravia Tourist (SMT) region in 2003:Q1- 2017:Q4, i.e. from the first quarter of 2003 to the last quarter of 2017. These countries were included in the selection by meeting two conditions, such as availability of data and a greater number of visitors from the country.

The monitored time series of dependent variables (tourism demand for Czech Republic and tourism demand for SMT region) are presented in Fig. 1. There can be seen a similar development of most of the series, i.e. it can be assumed the existence of long-term relationships. Furthermore, in the crisis period, the number of tourists is decreasing, indicating possible long-term relationships between the number of tourists and some macroeconomic indicators.

Fig. 1: Occupancy (in log) in CAE by selected countries in the Czech Republic

Source: Czech Statistical Office, elaborated by author.

In view of the aim of this paper, the explanatory (independent) variables are income as the gross domestic product (GDP) per capita at constant 2010 prices in the local currency. Tourism price as the real exchange rate (RER) based on the harmonized consumer price index (HCPI, 2010=100) for restaurants and hotels. Trade openness as the ratio of exports plus imports to GDP. In addition to determinants above, international outbound tourists as variable is used. International outbound tourists (IOT) are the number of departures that people make from their country of usual residence to any other country for any purpose other than a remunerated activity

69

in the country visited. The effect of departures number changes in tourism origin countries is important to analyse. None of the tourism studies do not consider this variable. Some of the studies consider total population of the tourism origin country. However, the population tends to be highly correlated with income. To take into account the cross-sectional dependence between units, unobserved dynamic common shocks are approximated using the method proposed by Pesaran (2006). This approach have the benefit that it do not require selection of a set of observed proxies. Source of data for the number of tourists was the Czech Statistical Office, real GDP per capita, nominal exchange rates as well as consumer price indices were obtained through the FRED database, administered by the Federal Reserve Bank of the USA. The data on exports and imports were collected from World Bank data source as well as international outbound tourists data. All time series were seasonally adjusted using the X-13ARIMA-SEATS software developed by U. S. Department of Commerce, U. S. Census Bureau. All statistical analysis were performed using Gretl and Eviews 8. Data were logarithmically transformed for further analyses.

Four hypothesis were formulated to study the effects of variables to interantional tourism demand in the Czech Republic and in the SMT region: (1) It is expected that income (as the real GDP per capita) has a positive impact on the number of tourists (as tourism demand). (2) If RER is greater than one, it means that foreign (Czech) goods are actually more expensive, i.e. the foreign (Czech) price level is higher than the price level in the origin destinations of the monitored tourists. Thus, it can be expected a negative relationship between price level and tourism demand which follows the law of demand. (3) The positive effect of trade openness between the Czech Republic and tourism origin countries is expected. (4) An increasing number of departures from destinations can be seen as an increase in the interest of tourists from these countries to travel. Furthermore, if this increase also leads to an increase in the number of tourists from these countries to the Czech Republic, this is evidenced by the attractiveness of the Czech Republic for these tourists. Given the choice of the tourism origin countries, it can be assumed that the number of departures from these countries will have a positive effect on tourism demand.

2 Methods

This paper applies panel cointegration analysis through a common correlation effect model (CCE). In panel analysis, it is first important to test the independence of individual panels and homogeneity of the estimated coefficients. For this purpose, we consider a standard panel data model, see Banerjee and Carrion-i-Silvestre (2017), for instance.

Cross-sectional dependence means the correlation between residuals across all cross-section units. The reason for analyzing cross-sectional dependence is the fact that its existence (or non-existence) affects the specific choice of test to determine the presence of the unit root within the variables, see Sharif and Hamzah (2015). Information about the presence of a unit root is important in cointegration analysis, which is one of the focus of this paper.

Two test statistics are used to verify the above hypotheses. Specifically, LM statistic proposed by Breusch and Pagan (1980) and CD statistic proposed by Pesaran (2004). It is also necessary to determine whether the slope coefficients in panel data

70

model are homogenous. We are interested in whether these coefficients are the same (homogenous) or different (heterogenous) across cross-section units. Standard for this purpose is the test procedure provided by Swamy (1970), which works with the assumption of a relatively small number of cross-sectional units. For this reason, the extension of the given procedure provided by Pesaran and Yamagata (2008) is applied here, specifically, both delta statistics ( and ) are used.

In order to check for the presence of cointegration relationships between the variables, it is important to test whether the variables contain unit root and therefore they are non-stationary, and are denoted as I(1). If the variables do not contain the unit root, they are stationary, I(0). In the panel analysis, two types of unit root tests can be applied, see Hurlin and Mignon (2007). The first type tests, sometimes referred to as the first generation tests, when the cross-correlation is not confirmed, for example, the Im-Pesaran-Shin test proposed by Im et al. (2003). For the second generation of panel unit root tests, the presence of cross-sectional dependence among the error terms is allowed within the panel, for example, the cross-sectionally augmented IPS test proposed by Pesaran (2007).

Following the determination of the stationary of the variables, the long-term cointegration relationship between the variables can be analysed by using the Durbin-Hausman (D-H) panel cointegration test proposed by Westerlund (2008). This test possible to conduct cointegration analysis in case the independent variables are I(1) or I(0), but dependent variable must be non-stationary in the sense of I(1). The test works with two statistics, Durbin-Hausman Group (DHG) and Durbin-Hausman Panel (DHP). In DHG, Westerlund (2008) allows the cointegration coefficients to differ between cross-sections, while the coefficients are considered to be the same for all cross-sections, in DHP. For these statistics, it is tested that there is no cointegration between variables in the null hypothesis. The alternative hypothesis for DHG indicates that there is cointegration relationship for at least some cross-sections. Rejection of the null hypothesis for DHP implies that there is a cointegration relationship for the whole of the panel, see Westerlund (2008).

Following the finding of cointegration between the variables, cointegration coefficients are estimated by using the Common Correlated Effects (CCE) method proposed by Pesaran (2006). One of the advantages of the CCE method is that cointegration coefficients can be estimated for each horizontal section unit. This makes an opportunity to individual interpretations cointegration coefficients for each section unit (country), see Banerjee and Carrion-i-Silvestre (2017).

3 Problem solving and results

The cross-sectional dependence was analysed for individual variables across the countries analysed as well as the whole model. The results of the cross-sectional dependence are presented in Tab. 1. Since, in all cases, the p-value is less than 0.01, the null hypothesis of the cross-sectional dependence was rejected at 1% significance level. Hence, it was concluded that there is cross-section dependence among the countries constituting the panel. In other words, the shock to one of the monitored countries is likely transmitted to the other countries.

71

Furthermore, Tab. 1 presents the homogeneity test of the estimated coefficients. According to the results, since the p-values of the tests were smaller than 0.01, null hypothesis of homogeneity was rejected and alternative hypothesis was accepted. Thus, it can be considered heterogeneous, in other words, it makes sense to look for their estimates for all cross-section units.

Tab. 1: Cross-sectional dependence test and homogeneity test Cross-sectional dependence test

Tests lnTD_cz lnTD_smt lnRER lnGDP lnTO lnIOT Model (CZ)

Model (SMT)

LM 182.61*** 193.24*** 264.85*** 263.48*** 243.61*** 196.93*** 1270.17*** 1277.21***

CD 18.10*** 15.12 *** 13.08 13.90*** 16.74*** 15.02*** 122.85 122.24***

Homogeneity test Tests Czech Republic SMT Region

Δ 9.08 8.99

Δadj 9.97 9.12 *** - statistically significant at 1% level,** at 5% level, * at 10% level.

Source: Own elaboration.

For further analysis, it is necessary to determine the presence of unit roots in the panel data. Due to the existence of the cross-sectional dependence, the second generation test was used, namely CIPS test proposed by Pesaran (2007). The results in Tab. 2 shows that for all logarithmized variables, the null hypothesis was rejected at 1% level of significance. Thus, the variables contain the unit root, i.e. I(1). Furthermore, first differencing the time series removed the unit root, they were stationary, i.e. I(0). For these reasons, the cointegration analysis can be used.

Tab. 2: CIPS Unit Root Test Results Variable lnTD_cz ln TD_smt lnRER lnGDP lnTO lnIOT

Constant -2.07 -2.01 -2.19* -1.91 -2.02 -1.63

Critical value 1%: -2.45; 5%: -2.25; 10%: -2.15

Constant+trend -2.69* -2.43 -2.58 -2.06 -2.52 -1.99

Critical value 1%: -2.93; 5%: -2.76; 10%: -2.66

Variable dif_lnTD_cz dif_ln TD_smt dif_lnRER dif_lnGDP dif_lnTO dif_lnD

Constant -8.98*** -7.76*** -8.75*** -6.11*** -10.02 -4.21

Constant+trend -9.05*** -8.12*** -8.86*** -7.07*** -10.25 -5.19 Source: Own elaboration

In order to confirm the existence cointegration relationships, a D-H test is applied. It offers two statistics, namely group statistic (DHG) and panel statistic (DHP). Tab. 3 presents the results of both these statistics, including their significance represented by the p-values. For both Durbin-Hausman panel and Durbin-Hausman group cointegration tests, the null hypothesis was rejected at 1% level of significance. In this case, the alternative hypothesis assuming that there are cointegration relationships in some countries and throughout the panel is accepted. Thus, cointegration test results show that there are some long-run relationships between tourism demand and independent variables and these relationships may be different for different countries. After deciding that there are long-run relationships, cointegration coefficients can be estimated.

72

Tab. 3: Durbin-Hausman Cointegration Test Results

CZ

(test statistic)SMT

(test statistic) Durbin-Hausman

(DHG) 7.84*** 8.98*** Durbin-Hausman (DHP) 3.44*** 4.21***

Source: Own elaboration.

Based on the above analysis, there are cointegration relationships between tourism demand (for the Czech Republic and for the SMT region) and GDP in the local currency (as income), real exchange rate, trade openness or international outbound tourists from tourism origin country.

The CCE results for the panel are reported in Tab. 4. The results show that the negative effect of the real exchange rate on the tourism demand is statistically significant, at 5% level. Thus, with the rising exchange rate, demand is decreasing, namely with 1% increase in the real exchange rate decreases tourism demand in the Czech Republic and the SMT region by 0.5% and 0.75%, respectively. The income effects as a measure of GDP on the tourism demand in the Czech Republic and SMT region are positive (0.85) at 10% level of significance and 5% level of significance, respectively. In other words, a 1% increase in GDP per capita in the tourism origin countries increases tourism demand by 0.85%. Trade openness is positively related to tourism demand in the Czech Republic, 1% increase in trade openness raises tourism demand by 0.066%. The results indicate that tourism demand is inelastic to bilateral trade and tourism demand is not very responsive to changes in trade volume between Czech Republic and respective countries. For tourism demand in the SMT region, trade openness is not statistically significant. Finally, the effect of international outbound tourists on tourism demand in the Czech Republic and SMT region, respectively is not statistically significant.

Tab. 4: CCE estimates for the Panel lnGDP lnRER lnTO lnIOT

Czech Republic 0.852* -0.500** 0.066* -0.039 (0.166) (0.074) (0.029) (0.055)

South Moravia Tourist Region

0.845** -0.747** 0.011 0.107 (0.137) (0.099) (0.081) (0.128)

Standard deviations are in round brackets. Source: Own elaboration

Tab. 5 presents country-specific CCE estimates. It is clear from the table that the expected positive relationships between the income of tourists and tourism demand in the Czech Republic and the SMT region are statistically significant for 11 countries (all except Japan and Hungary) and 7 countries (all expect France, the Netherlands, Spain, Israel, Japan and USA), respectively. The expected statistically significant negative relationship between RER and tourism demand in the Czech Republic and SMT region have been found for 11 countries (all expect Italy, Japan and Hungary) and for 10 countries (all expect Spain, Israel and Hungary), respectively. The effect of the trade openness on tourism demand for the Czech Republic is statistically significant and positive for France, the Netherlands, Germany, Spain, Italy and Hungary. The negative statistically significant effect have been found for Austria and

73

Japan. In the SMT region tourism demand, the negative statistically significant effect have been found for Spain and Hungary, the positive statistically significant effect for the Netherlands and Italy. The effect of international outbound tourists on tourism demand in the Czech Republic is negative statistically significant for France and Spain, the positive statistically significant for the Slovak Republic and Hungary. In the SMT region tourism demand, the effect of international outbound tourists is statistically significant for Spain (positive) and Italy (negative).

Tab. 5: CCE estimates for the Countries

Czech Republic South Moravia Tourist Region

lnGDP lnRER lnTO lnD lnGDP lnRER lnTO lnIOT

FR 1.267* -0.416** 0.097* -0.476* 0.781 -1.437** -0.130 -0.170

(0.235) (0.029) (0.019) (0.098) (0.993) (0.194) (0.158) (0.367)

AT 1.470* -0.183* -0.148 -0.423 2.047** -1.844** -0.009 -0.243

(0.433) (0.046) (0.093) (0.240) (0.287) (0.088) (0.038) (0.150)

NL 1.497* -0.783** 0.178** -0.016 0.563 -1.225** 0.173* 0.777

(0.283) (0.050) (0.023) (0.017) (0.437) (0.180) (0.043) (0.393)

DE 0.900** -1.114** 0.280* 0.211 1.270* -2.587** 0.343 -1.019

(0.137) (0.051) (0.076) (0.254) (0.310) (0.188) (0.205) (0.725)

SK 0.908* -0.793** 0.138 0.387** 1.575*** -1.615*** 0.045 0.040

(0.221) (0.094) (0.133) (0.057) (0.040) (0.080) (0.023) (0.013)

ES 1.541** -0.965** 0.098* -0.247** 0.490 0.465 -0.238 0.467*

(0.127) (0.068) (0.021) (0.030) (0.387) (0.156) (0.095) (0.087)

IT 0.650* 0.035 0.279** -0.037 1.293** -0.485** 0.193* -0.223*

(0.140) (0.037) (0.028) (0.040) (0.160) (0.040) (0.038) (0.057)

SE 0.007* -0.655* 0.095 0.000 0.577* -0.480** -0.002 -0.020

(0.001) (0.187) (0.038) (0.000) (0.107) (0.040) (0.035) (0.017)

IL 1.777* -1.885** -0.11 -0.403 0.453 -0.130 -0.095 0.513

(0.407) (0.094) (0.046) (0.213) (0.383) (0.120) (0.045) (0.223)

JP 0.003 0.0401 -0.255 -0.167 -0.143 -0.395** 0.188 0.630

(0.277) (0.037) (0.080) (0.245) (0.293) (0.068) (0.083) (0.243)

US 0.803** -0.202** 0.053* -0.027 0.463 0.415** 0.051 0.237

(0.127) (0.011) (0.015) (0.050) (0.363) (0.040) (0.049) (0.143)

PL 0.722** -0.720*** -0.070 0.023 0.323** -0.345** -0.148 0.080

(0.057) (0.026) (0.048) (0.062) (0.060) (0.038) (0.050) (0.060)

HU 0.473 -0.445 0.220* 0.673** 1.293* -0.055 -0.233 0.323

(0.150) (0.161) (0.040) (0.100) (0.233) (0.026) (0.150) (0.153) Source: Own elaboration

According to the comparative elasticities of the variables, the income elasticity (for tourism demand in the Czech Republic) is above unity for France, Austria, the Netherlands, Spain, Israel, with the highest elasticity coefficient of 1.78 for Israel. Thus, the tourism demand in the Czech Republic from these countries is income elastic. While for Germany, the Slovak Republic, Italy, Sweden, USA, Poland is the tourism demand in the Czech Republic from these countries income inelastic (elasticity coefficients are less than one). Tourism demand in the SMT region is income elastic for Austria with highest elasticity coefficient of 2.05, i.e. a 1% increase in GDP per capita in the tourism origin countries increases tourism demand from

74

Austrian tourists by 2.05%. Further, for Germany (1.27%), the Slovak Republic (1.58%), Hungary (1.29%) and Italy (1.29%). Tourists from these countries are more sensitive to the change in their income. On the other hand, tourism demand in the SMT region is income inelastic for Sweden and Poland that have elasticity coefficients of 0.58 and 0.32, respectively. Tourists from remaining countries do not respond to GDP changes since their coefficients are non-significant.

As far as the RER is concerned, Germany and Israel have elasticity coefficients of -1.11 and -1.89, respectively. For instance, a 1% increase in the tourism prices decrease the number of tourists from these two countries to the Czech Republic by levels of 1.11% and 1.89%, respectively. Thus, the tourism demand in the Czech Republic from these countries is price elastic. In terms of tourism demand in the SMT region, this demand is price elastic from France, Austria, the Netherlands, Germany and the Slovak Republic with the highest elasticity coefficient of -2.59 for German tourists. On the other hand, the lowest elasticity coefficients of -0.18 and -0.35 belong to Austria (in the Czech Republic) and Poland (in the SMT region), respectively.

Tourism demand in the Czech Republic is inelastic to bilateral trade. The trade openness has been found to be significant for France, the Netherlands, Germany, Spain, Italy, USA and Hungary. While, Germany has the highest elasticity coefficients of 0.28, the lowest elasticity coefficient of 0.05 belongs to USA. In terms of tourism demand in the SMT region, the trade openness has been found to be significant for the Netherlands and Italy. While the Italy has the highest elasticity coefficient of 0.193, the Netherlands have elasticity coefficients of 0.173.

In terms of tourism demand in the Czech Republic, the number of the international outbound tourists for tourism origin countries has been found to be positive, as expected, and significant for Hungary and the Slovak Republic but negative and significant for France and Spain. While Hungary has the highest elasticity coefficient of 0.67, the elasticity coefficients for France and the Slovak Republic are -0.48 and 0.39, respectively. The lowest elasticity coefficient of -0.25 belongs to Spain. With regard to tourism demand in the SMT region, the number international outbound tourists for tourism origin countries has been found to be positive and significant for Spain but negative and significant for Italy. For instance, a 1% increase in the number of departures that Spanish tourists make from their country of usual residence to any other country increase the number of tourists from Spain to the SMT region by levels of 0.47%.

4 Discussion

The main contributions of this paper are as follows: First, the effect of the examined four determinants on international tourism demand in the Czech Republic taking the cross-sectional dependence presence into account have been found. Second, The effect of these determinants on tourism demand in the South Moravia Tourist Region of the Czech Republic have been found.

From the findings of this study, results revealed that tourism demand in the Czech Republic as well as in SMT region is significantly and positively affected by income of foreign tourists but low value of the coefficient show that tourism demand is slightly dependent on the economic situation of the generating countries. This is consistent with

75

the result of Indrová et al. (2015), Surugiu et al. (2014) or Ongan et al (2017) which indicated that as the country’s income increases, more of its residents can afford to visit other countries, and therefore tourist arrivals are a positive function of income or directly related to income. In addition, our study shows that tourism demand in the Czech Republic and SMT region from some foreign tourists is income elastic.

As it regards relative price, the variable reduces the number of tourists to travel and the negative sign of the coefficient is the same as the expected one. The estimated price elasticity is suggesting that tourism demand is price inelastic. It holds for tourism demand in the Czech Republic as well as in the SMT Region. Different studies have used different measures of price explaining the differences in estimating price elasticities. As a standard, the real exchange rate adjusted by CPIs is used. In this paper, HCPI for hotels and restaurants is used to derive the real exchange rate instead of the general CPI. Found price inelasticity is consistent with the result of Indrová et al. (2015) or Yazdi and Khanalizadeh (2017). Eilat and Einav (2004) suggest that tourism demand to less developed countries is not very sensitive to fluctuation in prices, explained by the fact that prices in these countries are relatively low. However, this demand is price elastic from France, Austria, the Netherlands, Germany and the Slovak Republic.

As the empirical results suggest, increasing trade openness between Czech Republic and tourism origin countries has contributed to the increasing amount of tourists from these countries visiting the Czech Republic. Thus, the trade openness variable has an expected positive sign. Because of low coefficient elasticity, it is not very responsive to changes in trade volume between Czech Republic and tourism origin countries. These results are partly in agreement with Indrová et al. (2015) – they suggest that the trade openness is insignificant in explaining tourism demand. Statistically significant influence between the trade openness and tourism demand is confirmed by Surugiu et al. (2014). The effect of international outbound tourists on tourism demand in the Czech Republic and SMT region, respectively is not statistically significant.

Conclusion

The aim of the paper is to investigate the effect of income of tourists, tourism price, trade openness and international outbound tourists in tourism origin countries on the tourism demand to the Czech Republic as well as to South Moravia Tourist Region of the Czech Republic from tourists coming from thirteen different countries is analysed, namely France, Austria, the Netherlands, Germany, the Slovak Republic, Spain, Italy, Sweden, Israel, Japan, USA, Poland and Hungary over the period 2003:Q1-2017:Q4. This paper applies panel co-integration analysis through a common correlation effect model (CCE). LM and CD tests were used for testing cross-section dependence between the countries. The results showed existence of cross-section dependence between the countries. The stationarity of the series was tested using the CIPS unit root test. The cointegration relationships between the series were analysed by the Durbin-Hausman Cointegration Test which takes cross-section dependence into account.

According to the obtained estimation results, income and trade openness have a significant positive influence on international arrivals to the Czech Republic, while for tourism price the results indicate that it has a negative influence on inflows, as

76

expected. The effect of international outbound tourists on tourism demand in the Czech Republic is negative, against expectations, but it is not statistically significant. For tourism demand in the SMT region, income has a significant positive influence on international arrivals to the South Moravia tourist region, as it regards relative price, the negative sign of the coefficient is the same as the expected one. The effects of trade openness and international outbound tourists on tourism demand in the SMT region have expected positive signs but they are not statistically significant.

In order to get more specific results, the future research could to examine the dependence between tourism demand and selected determinants by using different approaches based on different copulas.

Acknowledgement

This paper was partially supported by the Research project Center of Transfer Innovation and Project Support of College of business and hotel management and by Tourist Authority South Moravia.

References

Babecká, O. (2013). Competitiveness and determinants of tourism. In: Global economic outlook-July. [online] Monetary and Statistics Department External Economic Relations Division of Czech National Bank, pp. 13–19. Available at: https://www.cnb.cz/miranda2/export/sites/www.cnb.cz/en/monetary policy/geo/geo_2013/gev_2013_07_en.pdf

Banerjee, A, Carrion-i-Silvestre, J.L. (2017). Testing for Panel Cointegration Using Common Correlated Effects Estimators. Journal of Time Series Analysis, 38(4), pp. 610-636.

Bonham C., Gangnes, B., Zhou, T. (2009). Modeling tourism: A fully identified VECM approach. International Journal of Forecasting, 25(3), pp. 531-549.

Breitung, J., Roling, Ch. Salish, N. (2016). LM-Type Tests for Slope Homogeneity in Panel Data Models. Econometrics Journal, 19(2), pp. 166-202.

Breusch, T., Pagan, A. (1980). The Lagrange multiplier test and its application to model specification in econometrics. Review of Economic Studies, 47(1), pp. 239–253.

Chang, Ch-L., Hsu, H-K., McAleer, M. (2017). A Tourism Financial Conditions Index for Tourism Finance. Challenges, 8(23), pp. 1-17.

Eilat, Y., Einav, L. (2004). Determinants of international tourism: A three-dimension panel data analysis. Applied Economics, 36(12), pp. 1315-1327.

Falk, M. (2015). The sensitivity of tourism demand to exchange rate changes: an application to Swiss overnight stays in Austrian mountain villages during the winter season. Current Issues in Tourism, 18(1), pp. 465–76.

Ganchev, G.T. (2014). Tourism industry: Role of the real effective Exchange rate. Tourism and Management Studies, 10(Special Issue), pp. 174–179.

Gautam, B.P. (2014). Economic Dynamics of Tourism in Nepal: A VECM Approach. In: Munich Personal RePEc Archive. [online]. Available at: https://mpra.ub.uni-muenchen.de/58102/8/MPRA_pa per_58102.pdf

Chaiboonsri, C., Chaitip. P., Rangaswamy, N. (2009). Modelling International Tourism Demand in Thailand. Annals of the University of Petroşani, Economics, 9(3), pp. 125-146.

Hurlin, C., Mignon, V. (2007). Second Generation Panel Unit Root Tests. In: EconPapers. [online]. Available at: https://halshs.archives-ouvertes.fr/halshs-00159842/document.

77

Im, K.S., Pesaran M.H, Shin Y. (2003). Testing For Unit Roots in Heterogeneous Panels. Journal of Econometrics, 115(1), pp. 53-74.

Indrová, J., Strielkowski, W., Vencovska, J. (2015). Determinants of Tourism Demand in Czech Republic. Actual Problems of Economiics, 3(159), pp. 93-99.

Ongan, S., Isik, C., Özdemir, A D. (2017). The Effects of Real Exchange Rates and Income on International Tourism Demand for the USA from Some European Union Countries. Economies, 5(4), pp. 51-61.

Pesaran, M.H. (2004). General diagnostic tests for cross section dependence in panels. In: Cambridge Working Papers in Economics of University of Cambridge, no. 0435. [online]. Available at: https://doi.org/10.17863/CAM.5113.

Pesaran, M.H. (2006). Estimation and Inference in Large Heterogeneous Panels with a Multifactor Error Structure. Econometrica, 74(4), pp. 967-1012.

Pesaran, M.H. (2007). A Simple Panel Unit Root Test in The Presence of Cross-Section Dependence. Journal of Applied Econometrics, 22(2), pp. 265-312.

Pesaran, M.H., Yamagata, T. (2008). Testing slope homogeneity in large panels. Journal of Econometrics, 142(1), pp. 50-93.

Seo, J.H., Park, S.Y., Yu, A.L. (2009). The Analysis of the Relationships of Korean Outbound Tourism Demand. Tourism Management, 30(1), pp. 530–43.

Sharif, N.S., Hamzah, N.A. (2015). A Robust Panel Unit Root Test in the Presence of Cross Sectional Dependence. Journal of Modern Applied Statistical Methods, 14(2), pp. 159-171.

Song, H., Li, G. (2008). Tourism demand modelling and forecasting e a review of recent research. Tourism Management, 29(2), pp. 203-220.

Song, H., Kim, J.H., Yang, A S. (2010). Confidence intervals for tourism demand elasticity. Annals of Tourism Research, 37(2), pp. 377-396.

Swamy, P.A.V.B. (1970). Efficient inference in a random coefficient regression model. Econometrica. 38(2), pp. 311-323.

Westerlund, J. (2008). Panel Cointegration Tests of the Fisher Effect. Journal of Applied Econometrics, 23(2), pp. 193-233.

Xu, F. a He, Z. (2017). Testing slope homogeneity in panel data models with a multifactor error structure. Statistical Papers, 59(173), pp. 1-24.

Yazdi, S. K., Khanalizadeh, B. (2017). Tourism demand: A panel data approach. Current Issues in Tourism, 20(8), pp. 787-800.

Contact Address

Mgr. Tomáš Jeřábek, Ph.D., MBA College of business and hotel management, Department of Economy and Management, Bosonožská 9, 625 00 Brno Email: [email protected]


78

SOLUTION OF SOCIO-ECONOMIC EFFICIENCY OF PUBLIC PROJECTS UNDER THE CONDITIONS OF

VARIABLE AMOUNT OF SUBSIDY

Petr Jiříček, Stanislava Dvořáková

Abstract. The paper will deal with modelling cash-flows of public investment projects under the conditions of variable amount of subsidy from public funds. A deterministic model based on a rational fractional function will be used to assessment of socio-economic efficiency of public projects subsidized from the European funds. In the model, we will examine the influence of the amount of the subsidy and critical changes of socio-economic cash-flows generated by a project on assessing public investments. We will use the European Union methodology for assessing socio-economic evaluation of project utility by means of cost-benefit analysis. The results of the crisis scenario analysis of the model in the Maple program created on the basis of the concept of Economic net present value (ENPV) and Economic internal rate of return (ERR) will be presented in 2D projection and evaluated by means of the ENPV function depending on the project parameters. The paper will also present using the model for stress test of socio-economic efficiency of innovation project from sphere digitalization of public government, and broadens so far published research.

Keywords: Public Projects, Socio-economic Efficiency, the ENPV Function, ERR, Subsidy from European Funds, Cost-benefit Analysis.

JEL Classification: C20, H43.

Introduction The principles of socio-economic efficiency of public projects are provided in frame

of welfare economics. Public project efficiency evaluation uses in its mathematical basis commercial methods – the net present value method, and the internal rate of return method (IRR). The commercial net present value method (NPV) is after applying the theoretical principles of welfare economics (cash-flows are the quantification of utility) available for evaluation public projects of investment character for its stress on expressing time-rated utility. It corresponds to utilitarian bipolar concept of utility as the difference between the benefit and harm (loss) to the welfare for a social entity. The adjusted method of economic net present value ( ENPV ) is generally used as the preferred method for evaluating investment oriented projects from European funds. These projects are characteristic by a certain amount of subsidy to the supported subjects determined according to the objectives of the European Union cohesion policy in case of European funds. The second in order preferred method for evaluating investment character projects subsidized from the European funds is the so called method of economic internal rate of return ( ERR ). This is based on finding a real root of the ENPV function from the parameters of the proposed project, which corresponds with the condition of indifferent evaluation of public project efficiency from the investor's perspective. The social discount rate of a public project is then compared to the ERR that has been found. (European Commission, 2015)

79

1. Statement of a problem The order of preferences for using both methods for public projects corresponds

also with the approach for evaluating commercial investment projects financed from private sources (Levi, Sarnat, 1994, Brealey et al., 2011). The recommendation for evaluating public projects by the commercial method of IRR reflects the fact that the IRR method has, in comparison with the NPV method, several significant disadvantages. These are connected namely with the so called non-conventional cash-flows generated by some projects, when more than one change of cash-flow polarity occur. The definition of the conventional type of investment is mentioned by (Bussey, Eschenbach, 1992) as an investment that contains one or more negative cash outflows, followed by one or more positive cash inflows. A non-conventional investment is defined as an investment that intersperses the positive and negative cash flows. Problems of the IRR method which are generally valid also for the ERR modification for public projects were completely dealt with in the past e.g. by (Teichroew et al., 1965) in compliance with solving the polynomial by the so called Descartes' rule. (Hazen, 2003) has solved the problem of multiple real IRRs by linking the present value of the outstanding capital expenditure of a project with the difference between any IRRs and the cost of capital. Derivation mathematical rules for the construction of the NPV function and solution for nonconventional stream of cash-flow is in (Marek, Radová, 2006). (The topic of using IRR and NPV criteria has been summarised by (Magni, 2011). The generalized net present value (GNPV) method for nonconventional project evaluation is dealt with (Kulakov, Kulakova 2013). The generalized internal rate of return (GIRR) and generalized external rate of return (GERR) indices based on the generalized net present value (GNPV) are presented by (Kulakov, Kastro 2015).

Besides the above stated topic of nonconventional cash-flows, for the area of public projects another significant problem is determining the discount rate of public projects, which cannot be derived on the basis of market principles. (Marglin, 1963) deals with criteria for evaluating public projects by the cost-benefit analysis method and the social discount rate of public projects and with. For evaluating the efficiency of public projects, he uses the Optimal Rate of Investment (Marglin, 1967). Similarly, (Feldstein, 1964) uses the term Net Social Benefit for evaluating public projects utility. (Dasgupta et al. 1972) create the methodology of development projects evaluation for UNIDO. Economic analysis of public projects is dealt with by (Squire, Van der Tak, 1975). The present methodology of the European Union issues namely from works of (Florio 2006, 2007) and distinguishes two phases in public investment projects evaluation (European Commission, 2015) In the first phase projects are evaluated on a purely commercial basis of financial cash-flows by means of the so called FNPV (financial net present value) and FRR (financial rate of return), which corresponds to the above mentioned IRR. In the other phase of the economic analysis of a project, economic cash-flows are aggregated by means of the so called fiscal corrections of calculated financial cash-flows and by adding quantified negative and positive externalities. The resulting evaluation of social efficiency of public projects is carried out on the basis of discounting economic cash-flows by means of the ENPV and ERR (European Commission, 2015).

80

Projects are supported from the EU funds in the situation when a project generates negative FRR and FNPV values and so it is inefficient from purely commercial perspective. By adding positive cash-flows from positive externalities of proposed projects we then arrive, in the stage of project economic analysis, from negative values of FRR and FNPV at positive values on the basis of ERR and ENPV . A project built like this can obtain a subsidy from European funds because it shows positive utility and contributes to increasing social welfare and natural resources protection. In the real economic situation, however, there may be scenarios which can generate negative cash-flows in a project, usually proposed as a conventional one. Those will cause a nonconventional nature of the project, i.e. there may not be exactly one positive real-valued root of the ENPV function (i.e. the value of 0ERR ). Then it is not possible to decide conclusively about a project on the basis of the ERR method.

The presented paper aims at analysing the impacts of the change in the amount of a subsidy of a public project in selected economically justified scenarios of evaluating a project by means of the ERR , which is based on a modification of the commercial IRR method. This application contains both advantages and disadvantages of the commercial IRR method, described in the above mentioned specialist sources and summarized in (Magni 2011). For the description of public project behaviour under the conditions of variable amount of subsidy from European funds, we will use determinist model of the ENPV function within numeric solution of the model by the Maple program. These model scenarios will capture changes of behaviour of the ENPV function and its roots with a view of the amount of subsidy depending also on changes of further parameters, i.e. cash-flows generated by an investment in the building, operational and liquidation phases.

2. Methods

2.1 Notations For construction and analysing the ENPV model, the following notation and

assumptions are used (Dvořáková, Jiříček 2013).

ERR Economic rate of return

)(kfENPV Economic net present value displayed by means of the investment curve of project

k Social discount rate

t Years of project lifetime

n Investment project lifetime

nCFCFCF ,,0 Stream of cash flows

2.2 Assumptions Conventional projects – they are characterized by only one change of their polarity

in the sequence of the stream of cash-flows generated within the project (e.g. , ).

Nonconventional projects – they are characteristic by more than one change of project generated cash-flows in the sequence of the CF stream (e.g. , ).

81

Investment curve of the project – defined by the rationally fractional function

n

nn

tt

t

k

CF

k

CF

k

CFCF

k

CFENPV

1111 221

00

. (1)

It expresses the relation between economic net present value ENPV of the project (dependent variable) and the social discount rate k of a public project (independent variable). The investment curve will be analysed with a view of solving real-valued roots.

The degree ( n ) of the ENPV investment (1) – it represents the period of economic lifespan of a public project (in the model, n will equal 6 years).

The constant term ( 0CF ) – it represents capital expenditures of the project in the model. It will acquire non-positive values, which corresponds to a negative cash-flow reflecting project investment costs or as the case may be various (i.e. up to 100%) amount of subsidy from the donor.

Coefficients ( nCFCF ,,1 ) – in the model represent cash-flows caused by the investment and they can generally acquire both positive and negative values, while

11 ,, nCFCF represent the operational phase of the project and nCF the liquidation phase of the project.

Years of project's lifetime ( t ) – the model is built on the initial prerequisite of the possibility to separate the investment phase (year 0), the operational phase (years 1 to

1n ) and the liquidation phase (year n ) of the project.

Stream of CFs – the sequence of cash-flows generated by the project in the form of nCFCFCFCF ,,, 10 , in the model in the basic form 1,1,1,1,1,1,6CF , which corresponds

to realistically predictable flows of proposed public projects.

Economic internal rate of return of the project ( ERR ) – the root of the investment curve (1) of the project, i.e.

010

n

tt

t

ERR

CFENPV . (2)

3. Problem solving

3.1 Modelling the cash-flow of a public project For the proposed model from a set of variants of public investment projects, this

paper will only consider some chosen economically justified scenarios when an originally conventional project may change into a project of non-conventional character on the basis of social evaluation of utility. For the purpose of comparing modelling results, we analyse also two types of conventional projects – a classical conventional model of a public project whose stream of cash-flow contains a negative investment expenditure as well as positive utilities from the project, and a type when after a negative investment expenditure at the beginning of the operational phase of the investment there is a negative cash-flow (e.g. due to delay of the project implementation). We also analyse two types of non-conventional projects.

We display the course of the project investment curve )(kfENPV (see the definition (1)), i.e. the relation of the utility from the project to the social discount rate. We use 2D projection and seek the dependence of the curve course and root position

82

on the amount of the subsidy from public sources, manifesting in the change of 0CF for the defined cases of (Dvořáková, Jiříček 2013):

Classical conventional project ,CF

Conventional project with a crisis in the building phase ,CF

Nonconventional project with a crisis in the operational phase ,CF

Nonconventional project with a crisis in the liquidation phase ,CF

For these alternatives we chose basic scenario with sequence of cash flow in shape

In each of these alternatives we will simulate situation, when the public

investment project will be supported in level 0%, 30%, 70% and 100% subsidy from European funds.

3.2 The investment curve course in 2D projection First, we will analyse the course of the investment curves ENPV for classical

conventional projects and for a conventional project with a crisis in the building phase; then the ENPV functions for nonconventional projects will be displayed.

The course of the investment curve of a classical conventional project shows the influence of subsidy of investment costs on the position of the ENPV function roots at a given unit stream of cash-flows in the operational and liquidation phases of the project. The subsidy amount influences the shift of the curves along the y axis.

Fig. 1: The course of the investment curve of the project ENPV=f(k) according to the subsidy amount (a classical conventional project)

Source: authors

From the perspective of assessing the project efficiency by the ERR method, we look for positive real-valued roots. According to Descartes' rule of signs these conventional projects have one real-valued root greater than –1. There is a monotonously decreasing function with the asymptote in the value 0CF for k . In case of a classical conventional project (Fig. 1) the ENPV function does not have a final positive root at 100% subsidy, but ERR (see Tab. 1).

83

Fig. 2: The course of the investment curve of the project ENPV=f(k) according to the subsidy amount (conventional project – crisis in the building phase)

Source: authors

Tab. 1: Roots (ERR) of investment curves at various subsidy amounts (conventional projects)

subsidy parameter 0CF ERR (classical conventional

project) 1,1,1,1,1,1,0CFCF ERR (crisis in the building phase) 1,1,1,1,1,1,0 CFCF

0% –6 –1.67 0 –1.729 –0.082

30% –4.2 –1.705 0.112 –1.793 –0.01

70% –1.8 –1.787 0.508 –2.045 0.18

100% 0 –2 ∞ 0.966 ∞ Source: authors

In this case, the ERR method cannot be efficiently used for evaluation projects, in the positive area the root limits to +∞ and the project is thus efficient for any amount of the social discount rate chosen by the investor. The influence of a negative cash-flow due to the project building (Fig. 2) causes a change of the curve course. This curve is still monotonously decreasing. In cases of 30% and 0% subsidy the ERR method cannot be used either, because the roots ERR of the ENPV are negative (see Tab. 2).

The course of the investment curve of a nonconventional project in the situation of a crisis in the operational phase (change of 3CF – see Fig. 3) is not different in the right half of the curve from the course of the conventional project (Fig. 1). According to Descartes' rule of signs this project has three or one real-valued root greater than –1. The course of the ENPV function is quite different in case of a crisis in the liquidation project phase (Fig. 4), when the curves "turn over" in 1k to on the vertical asymptote. According to Descartes' rule this nonconventional project has two or no root greater than –1.

84

Fig. 3: The course of the investment curve of the project ENPV=f(k) according to the subsidy amount (nonconventional project – crisis in the operational phase)

Source: authors

Fig. 4: The course of the investment curve of the project ENPV=f(k) according to the subsidy amount (nonconventional project – crisis in the liquidation phase)

Source: authors

Tab. 2: Roots ERR of investment curves at various subsidy amounts (non-conventional projects)

subsidy parameter 0CF ERR (crisis in the operational

phase) 1,1,1,1,1,1,0 CFCF ERR (crisis in the liquidation

phase) 1,1,1,1,1,1,0 CFCF

0% –6 –1.784 –0.098 –1.784 –0.098

30% –4.2 –1.848 –0.013 –1.848 –0.013

70% –1.8 –2.2024 0.282 –2.2024 0.282

100% 0 –2.792 ∞ –2.792 ∞ Source: authors

85

4. Results Now we will discuss the results of using the proposed 2D model by means of a case

study depicting a crisis in the liquidation phase of a public investment project in the conditions of predicted cash-flows (instead of the unit cash-flow in the model). Here, the proposed model testing will be carried out by means of a real-life public project, (see Tab. 3) financed from the European funds from the Integrated Operational Programme from the period of 2007–13. Tab. 3: E-Government project cash flows

CF item 2010 2011 2012 2013 2014

Investment costs –576,150 –57,273,850 –300,000 0 0

Wage costs 0 0 0 –400,000 –400,000

Implementation 0 0 0 –830,000 –830,000

Technical support 0 0 0 –2,600,000 –2,600,000

Maintenance 0 0 0 –800,000 –800,000

Time saving – region 0 0 0 27,871,200 27,871,200

Time saving – state 0 0 0 340,000 340,000

Time saving – clients 0 0 0 12,000,000 12,000,000

Project cash flow –576,150 –57,273,850 –300,000 35,581,200 35,581,200

CF item 2015 2016 2017 2018 2019

Investment costs 0 0 0 0 0

Wage costs –400,000 –400,000 –400,000 –400,000 –400,000

Implementation –830,000 –830,000 –830,000 –830,000 –830000

Technical support –2,600,000 –2,600,000 –2,600,000 –2,600,000 –2,600,000

Maintenance –800,000 –800,000 –800,000 –20,800,000 –6,800,000

Time saving – region 27,871,200 27,871,200 27,871,200 27,871,200 27,871,200

Time saving – state 340,000 340,000 340,000 340,000 340,000

Time saving – clients 12,000,000 12,000,000 12,000,000 12,000,000 12,000,000

Project cash flow 35,581,200 35,581,200 35,581,200 15,581,200 29,581,200

CF item 2020 2021 2022

Investment costs 0 0 0

Wage costs –400,000 –400,000 –400,000

Implementation –830,000 –830,000 –830,000

Technical support –2,600,000 –2,600,000 –2,600,000

Maintenance –6,800,000 –6,800,000 –6,800,000

Time saving – region 27,871,200 27,871,200 27,871,200

Time saving – state 340,000 340,000 340,000

Time saving – clients 12,000,000 12,000,000 12,000,000

Project cash flow 29,581,200 29,581,200 29,581,200Source: (Olomoucký kraj, 2011)

86

It is a thirteen-year-long project with the aim to create a technology centre of the region in the system of electronic document and records management system, which should be part of e-Government centres within the CR public administration digitalization. The subsidy of the conventionally proposed project will be 85% of eligible budget costs (Olomoucký kraj, 2011).

Based on the proposed model, we will now carry out the project socio-economic efficiency evaluation within the scope of one of the model scenarios – the crisis scenario in the liquidation phase due to increased maintenance costs and positive externalities decrease (time saving for project stakeholders) related to the transition of part of the users to newer technologies (see Tab. 4). The year of 2022, i.e. the last year of the project where a decrease of the above-mentioned values is highly probable after twelve years of the project lifetime, was selected as the critical year.

Tab. 4: E-Government project cash flows, crisis in liquidation phase (in CZK) CF item 2020 2021 2022 Investment costs 0 0 0 Wage costs 400,000 400,000 400,000 Implementation 830,000 830,000 830,000 Technical support 2,600,000 2,600,000 2,600,000 Maintenance 6,800,000 6,800,000 16,800,000 Time saving – region 27,871,200 27,871,200 3,000,000 Time saving – state 340,000 340,000 0 Time saving – clients 12,000,000 12,000,000 0 Project cash flow 29,581,200 29,581,200 –17,630,000

Source: (authors according to Olomoucký kraj, 2011)

Fig. 5: The course of the ENPV function of the e-Government project

Source: authors

We can see that despite the impact of the crisis situation in the liquidation phase on project efficiency social evaluation, the project is stable even in this scenario (see Tab. 5), which is also reflected by the smooth course of the plane in the following Fig. 10.

As shown in Tab. 5, after modelling a crisis in the liquidation phase of the project there was no significant decrease of the project’s internal rate of return. It testifies little

87

sensitivity of the project to the changes of increasing maintenance costs and positive externalities resulting from time savings. The reason may be overstated cash-flow values, especially those from positive externalities in previous years. By further detailed examination of this phenomenon by the sensitivity analysis method in the next stage of modelling, we could find out the sensitivity of the economic internal rate of return of the project to cash-flow changes in individual years and determine the so-called cash breakeven points of the project.

Tab. 5: Results of testing the case study of the e-Government project; evaluation of social efficiency of a nonconventional model (crisis in liquidation phase)

Project social efficiency evaluation

Project proposal

Crisis in liquidation phase

ERR 0.401 0.393 Source: authors

5. Conclusion Efficiency evaluation of a public project of investment type is carried out on the

grounds of cost-benefit analysis which includes, beside evaluating commercial utilities of the project on financial basis (corresponding to commercial projects), also evaluation of social utilities according to theoretical principles of welfare economics. As stated in the article, basic criteria methods for evaluating public projects exploit the same mathematical principles and are based on the same mathematical foundations as commercial methods, i.e. on the grounds of net present value of the project and internal rate of return of the project.

In the area of public investment, the NPV method (according to the EU methodology the ENPV) is considered the preferential method as it reflects the basic principle of welfare economics regarding maximizing the social utility from their implementation, similarly as maximizing the market value of a company is the goal of commercial projects.

Just like in case of commercial projects, the IRR method (according to the EU methodology the ERR) is only the second most preferred method of public investment evaluation; the reason is well-known disadvantages of this method in case of nonconventional projects, mentioned in this article. Yet, some theoretical and practical concepts prefer, as stated in the article, the IRR (ERR) method.

The paper solves nonconventional public projects when negative cash-flows arise due to crisis situations in operational and liquidation phase of a public project. The paper also presents a comparison with public projects of conventional type as usually constructed by applicants for subsidy.

The model is applicable also for commercial projects due to identical mathematical basis of the ERR (IRR) methods. The model application based of cost-benefit analysis principles proves suitable especially for evaluating those commercial projects that cause negative externalities in the form of environmental damage by their implementation.

Acknowledgement

This contribution was supported by the College of Polytechnics Jihlava No. 1170/04/189 “Rozvoj 3D modelu pro ekonomické hodnocení investic“.

88

References

Brealey, R.A.; Myers, S.C.; Allen, F. (2011). Principles of Corporate Finance. New York: McGraw-Hill Irwin.

Bussey, L.E.; Eschenbach, T.G. (1992). The Economic Analysis of Industrial Projects. New Jersey: Prentice-Hall.

Dasgupta, P.; Marglin, S. (1972). Guidelines for Project Evaluation. New York: United Nations.

Dvořáková, S.; Jiříček, P. (2013). Modelling Financial Flows of Development Projects Subsidized from European Funds. In: 31st International Conference on Mathematical Methods in Economics, Jihlava: Vysoká škola polytechnická Jihlava, pp. 147-152

European Commission. (2015). Guide to Cost-Benefit Analysis of Investment Project. Brusel: DG Regional and Urban Policy.

Feldstein, M. (1964). Net Social Benefit Calculation and The Public Investment Decision. Oxford Economic Papers, vol.16(1), pp. 114–131.

Florio, M. (2006). Cost-Benefit Analysis and the European Union Cohesion Fund: On the Social Cost of Capital and Labour. Regional Studies, vol. 40(2), pp. 211–224.

Florio, M. (2007). Cost-Benefit Analysis and Incentives in Evaluation. The Structural Funds of the European Union. Cheltenham: Edward Elgar.

Hazen, G.B. (2003). A new perspective on multiple internal rates on return. The Engineering Economist, vol. 48(1), pp. 31–51.

Kulakov, N.Y.; Kulakova, A.N. (2013). Evaluation of Nonconventional Projects. The Engineering Economist 04/2013, vol. 58(2), pp. 137–148.

Kulakov, N.Y.; Kastro, A.B. (2015). Evaluation of Nonconventional Projects: GIRR and GERR vs. MIRR. The Engineering Economist 01/2015, vol. 60(3), pp. 183–196.

Levi, Ch.; Sarnat, M. (1994). Capital Investment and Financial Decisions. New Jersey: Prentice Hall International.

Magni, C.A. (2011). Average Internal Rate of Return and investment decisions: a new perspective. The Engineering Economist, vol. 55(2), pp.150–180.

Marek, P.; Radová, J. (2006). Funkce čisté současné hodnoty při konvenkčních a nekonvenčních peněžních tocích. Český finanční a účetní časopis, vol.1(4), pp. 23–33.

Marglin, S. (1963). The Social Rate of Discount and The Optimal Rate of Investment. Quarterly Journal of Economics, vol. 77(1), pp. 95–111.

Marglin, S. (1967). Public investment criteria. Benefit-Costs Analysis for Planned Economic Growth. London and Cambridge: Allen and Unwin and MIT Press.

Olomoucký kraj. (2011). Studie proveditelnosti projektu Technologické centrum a elektronické spisové služby. [online]. Available at https://docplayer.cz/1023389-Studie-proveditelnosti-projektu-technologicke-centrum-a-elektronicke-spisove-sluzby.html [Accessed 16-10-2018]

Squire, L.; Van der Tak, H. (1975). Economic Analysis of Projects. Baltimore: John Hopkins University Press.

Teichroew, D.; Robichek, A.A.; Montalbano, M. (1965). An analysis of criteria for investment and financing decisions under certainty. Management Science, vol. 12(3), pp. 151–179.

Contact Address

Petr Jiříček College of Polytechnics Jihlava, Department of Economic Studies Tolstého 16, 586 01 Jihlava, Czech Republic Email: [email protected]

89

Stanislava Dvořáková College of Polytechnics Jihlava, Department of Mathematics Tolstého 16, 586 01 Jihlava, Czech Republic Email: [email protected]


90

LIFE SATISFACTION OF CZECH SENIORS: WHAT ACTUALLY MATTERS?

Lea Kubíčková, Vladimír Hajko, Naďa Hazuchová, Martina Rašticová

Abstract: We estimate the influence of various factors on life satisfaction of Czech seniors in a large survey sample. We find that good health, more education and awareness of voluntary work participation, employee satisfaction and being currently employed are the main factors that contribute to being satisfied with the current quality of life in the group of Czech seniors. Surprisingly enough, increasing self-reported financial sufficiency is negatively associated with the quality of life. The main factors contributing to the life dissatisfaction are associated with being socially separated. The worst outcomes are recorded for those living in social homes and living alone. Any reported expectations of expected life changes (both positive and negative expectations) are associated with lower probability of life satisfaction.

Keywords: Seniors, Employee Satisfaction, Life Satisfaction, Well-being, Logit.

JEL Classification: J26, J28.

Introduction

The proportion of older people in the population is increasing and this long-enduring demographic trend (resulting from the so-called age pyramid) will only become more and more pronounced. The increasing number of seniors influences all areas of social and economic development across the EU. Demographic aging is a new challenge for 21st century society, both in terms of economic growth, and new forms of social integration (Čevela, Kalvach, Čeledová, 2012). The positive representations of ageing, which also include the image of active old age, form hand in hand with discussions pointing out the increasing number of older adults in the population as a negative phenomenon (Hasmanová Marhánková, 2017). Among the major emerging topics in this context in the individual countries is the life satisfaction of seniors.

Studies suggest the quality of life of elderly is influenced by physical health, mental status, personal conviction, social relationships, and interaction of many other social, health, economic and environmental conditions that affect human and social development. From the psychological point of view, it is mainly the subjective well-being of individuals and life satisfaction (Hnilicová, Benko, 2005). The possibility of declining life satisfaction levels over the retirement transition, retirees with low levels of individual pension income with a paid job are much better off – financially and psychologically – than retirees in the same position without a paid job (Dingemans, Henkens, 2014). The ageing process is characterized by physical and cognitive slowdown (Cullati et al. 2014), and health issues correlate with and determine life satisfaction (Gana et al. 2013). As individuals become older, health status is characterized by high heterogeneity (Mitnitski et al. 2017).

As stated by Gabriel and Bowling (2004), European research in particular has often focused on the positivist and functionalist aspect of aging (i.e., the attempt to describe available social welfare resources or to normalize the appropriate limits for "ensuring" an "adequate quality of life"). Attention has thus been paid almost

91

exclusively to the negative aspects of aging. The subjective perception of one’s satisfaction is, however, remarkably neglected – although, as noted by, for example, Kolosnitsyna, Khorkina and Dorzhiev (2017), the life satisfaction of most people can be represented by a U-shaped curve with the lowest satisfaction roughly in middle-age – an increase in life satisfaction can actually be expected in elderly. Instead, the general anticipation is a decline associated with negatively perceived aspects, for example in the deterioration of health.

Considering this persistent lack, this article analyzes the primary data to identify the factors which influence the perceived life satisfaction of seniors in the Czech Republic. Conclusions can be used in policy design and also provide the evidence on the the significant importance of these cognitive findings in addition to the normative approach – as recommended e.g. in Kolosnitsyna, Khorkina, and Dorzhiev (2017).


The issues of life satisfaction of seniors are investigated in Diener and Suh (1998), Borg et al. (2006), Cheng and Chan (2006) or Baird et al. (2010). Existing literature focuses mainly on psychological factors, health quality (especially frequent in medical research), or relatively broadly conceived quality of life estimates (QoL). While without a doubt all these areas relate to life satisfaction, nevertheless the perceived life satisfaction remains relatively outside the focus of interest. Šolcová and Kebza (2004, p. 21) argue that from a psychological point of view, the well-being is linked to the subjective assessment of the quality of life as a whole, a long-term state in which the satisfaction of the individual with his or her life is reflected. Gabriel and Bowling (2004) draw attention to the fact that most existing QoL models do not take into account the views and priorities of seniors and have not adequately tested their validity.

Personal well-being must be understood and measured through cognitive components (life satisfaction, moral principles), emotional ones (positive and negative emotions) and characterized by consistency in different situations and stability over time. Some studies focus only on "psychological well-being" – e.g. Hao (2008) deals with the psychological well-being of seniors. Surveys related to the life satisfaction of elderly and their determinants are quite rare. Kolosnitsyna, Khorkina and Dorzhiev (2017) conducted a survey of life satisfaction among Russian elderly and found the strongest correlation with indicators of family status, state of health, social status and personal income. However, they point to significant gender differences in life satisfaction factors, when having a job and higher education increases the satisfaction of women, but not men.

Multiple factors can influence the perceived satisfaction with the life situation of seniors. Baird et al. (2010)performed a comparison of two representative panel studies in Britain and Germany that assessed changes in satisfaction over the course of one’s life. In both panels they found that life satisfaction dropped sharply in people over 70. Similarly, Diener and Suh (1998) examined subjective well-being among many nations and found life satisfaction to be fairly stable. From the British results analyzed by Baird et al. (2010), however, a slight increase in life satisfaction could be observed from 40 to 70 years. Hamarat et al. (2002), however, did not find any differences in life satisfaction among groups of people between 45 and 89 years. Blanchflower and Oswald (2008), on the other hand, offer international evidence that there is a link

92

between life satisfaction and age. In addition, Walker (2005), Pinquart and Sorensen (2000, 2001) and Bourque et al. (2005) are inclined toward the claim that the determinants of life satisfaction may vary by age, gender, family status and cultural context, which in particular indicates the need for specific research within individual groups (especially those to be targeted by a specific policy, i.e. seniors).

2 Methods

As existing studies suggest, life satisfaction is influenced by a number of socio-economic characteristics in elderly (such as the income situation, the form of housing, whether the given person is still working, etc.) along with the state of health, social relations, place of residence but also unevenly represented interventions resulting from government or local policies (e.g. accessibility of social care centers, etc.). Because of the complex character of the issue on hand, we analyze the data from several points of view. We use three mutually complementary approaches.

First we examine the existing secondary data from Eurostat EU-SILC Survey (European Union - Statistics on Income and Living Conditions). In 2014, questions directly related to life satisfaction were added to the survey.

Primary data on satisfaction with the life situation of Czech seniors were collected as a part of TAČR Beta project No. TB05MOSV004 "Survey on the Quality of Life of Elderly Citizens in the Czech Republic". This article uses the primary quantitative data obtained from the questionnaire survey among the Czech seniors in 2016.

The total number of respondents surveyed was 3,045, while the number of complete observations obtained for quantitative analysis was 2931. Of these, 36.73% of respondents were men and 63.27% of respondents were women. The respondents' age ranged from 55 years and over. The largest part consisted of respondents aged 66-70, a total of 39.5%, almost a third (33%) of respondents were in the youngest category 55-65 years, less than a fifth (18.3%) of respondents were aged 71-75 And almost 10% of respondents were over 76 years of age (of which 3.1% of respondents were over 81 years of age). The original questionnaire contained more than 100 questions, but only 19 variables were quantitatively applicable. The description of the variables and their values is available in the appendix. Data are represented by categorical variables (X3, X4, X5, X6, X10, X116, X117), binary (dummy) variables (X1, X8, X11, X34, X35, X101, Ysafe, X115) and ordinal variables (X2, X39, X49, X100). For estimation, the categorical variableswere broken down into individual categories (note the values of the coefficients of the categorical variables in the model therefore express the contrast with the basic category. For example in Tab. 1, X117val2 represents the dummy variable with the value of 1 for observations of X117 = 2 (i.e. "How did you perceive retirement?" = "I cannot identify with it so far"), and 0 otherwise. The coefficient value for X117val2 reported in Tab. 1 thus records the effect of this category compared to the basic category (of negative perception). Note that default categories are marked with the asterisk (*) symbol in Tab. 3 in the appendix).

To identify the important factors that positively or negatively affect the subjective perception of life satisfaction of seniors in the Czech Republic, the article uses the so-called binomial logistic regression model.

93

3 Problem solving

EU-SILC data provide some information about the satisfaction of Czech seniors with the selected areas of life. Seniors are understood here to be a person aged 50 years or more, Fig. 1 illustrates in greater detail the level of seniors' satisfaction with selected areas of life. It is clear that most of the seniors are dissatisfied with the financial situation of households (37%) and with the amount of free time (28%), which is related to the fact that working seniors (persons over 50) are included in this analysis. On the contrary, they are most satisfied with housing (70%) and personal relationships (70%). Overall, it is clear from the EU SILC data that Czech seniors are more satisfied (52%) than dissatisfied (29%) and 16% of people have expressed a neutral attitude. The most satisfied are people in the 60-65 age group (54%).

Fig. 1: The level of satisfaction with the selected areas of life

Source: (EU-SILC, own calculations, 2018)

The satisfaction of seniors is closely related to seniors’ anticipation of the development of their life situation for the future. Fig. 2 shows the anticipated development of the overall life situation of Czech seniors. The EU SILC data suggests that most seniors assume that their situation will not change or it will worsen and only a small percentage across all the relevant age categories assume their life situation will improve.

94

Fig. 2: Predicted development of the overall life situation

Source: (EU-SILC, own calculations, 2018)

The identification of factors that positively or negatively affect the perception of the feeling of satisfaction of Czech seniors is based on the binomial logit model. The results of the model estimates (after top-down sequential elimination) are summarized in the following table (Tab. 1).

Tab. 1: logit model for X115A, N = 2931

coefficient Odds ratio standard error

p-value

Financial sufficiency (X100) -0.37196 0.6893834 0.042125 0.000 Dependence on social benefits (X101A) -0.68031 0.5064606 0.270226 0.012 Life situation will get worse (X116val3) -0.54858 0.5777705 0.249375 0.028 Life situation will get better (X116val4) -1.38541 0.2502202 0.574107 0.016 Positive perception of retirement (X117val1)

-1.200716 3.322494 0.432884 0.006

I cannot identify with retirement (X117val2)

-1.35062 0.2590804 0.385176 0.000

Awareness of voluntary work participation (X34A)

0.580033 1.786096 0.245019 0.018

Health quality (X39) 1.277264 3.586812 0.176704 0.000 Housing – social home (X4val1) -2.40202 0.0905353 0.440569 0.000 Housing – Rented apartment (X4val2) -0.80857 0.4454957 0.283746 0.004 Living alone in a social home (X5val4) -0.69175 0.5006999 0.247546 0.005 High school education (X6val2) 0.930934 2.536876 0.263308 0.000 University/college education (X6val3) 0.901156 2.462447 0.33014 0.006 Currently employed (X8A) 0.603184 1.82793 0.315751 0.056 _cons 1.387253 0.587531 0.018

Source: (authors, own calculations, 2018)

Significant variables positively influencing the perceived life situation are X39 (with a better health condition the probability of choice 1 increases, with an additional increase on the given scale, the odds ratio for satisfaction with their current life situation increases 3.5 times), X117val1 (i.e. a well-perceived retirement – a surprising

95

finding is that X117val2 indicates that the inability to identify with retirement reduces satisfaction with the current life situation, even compared to a poorly perceived retirement (Sic!)), knowledge of the possibility of the involvement of seniors in volunteering and whether the person has attained secondary or higher education (both cases with virtually the same influence, i.e. the education attained increases the odds ratio by approximately 2.5 times).

The remaining regressors depict a negative effect on the probability of choice 1 (= good life situation of the seniors). The most significant impact is then apparent in the X4 variable, i.e. the fact whether a person lives in a social facility (compared to their own home, with an odds ratio about 91% less than the odds ratio compared to the situation of seniors living in their own house) or in a rented flat (a 54% lower odds ratio than seniors living in their own home). The negative effects on the X116 variable categories indicate that the expectation of a change in the current situation reduces the satisfaction of seniors, both in cases of improvement and worsening.

Surprisingly, the growth of the perceived sufficiency of finances (X100) has a negative effect, as does the use of social security benefits (X101A). As expected, a person living alone (X5val4) has a negative effect on the probability of choice 1 for the explained variable (the odds ratio is roughly half that of cohabitation with a husband / wife).

The p-value of the Hosmer-Lemeshow test (0.432) indicates that the model matches the data well. The value of pseudo is 0.405 and the area under the ROC curve is 0.932, indicating the good classification capability of the model as a whole. However, given the significantly unbalanced share of the values of the explanatory variable, it is naturally desirable to pay attention to the exchange matrix (Tab. 2).

Tab. 2: Exchange matrix Observed

1 0 total classified 1 2400 12 2412 0 421 98 519 total 2821 110 2931

Sensitivity Specificity Correctly classified (total) 85.08% 89.09% 85.23%


The model as a whole is able to correctly classify a total of 85.2% of the cases (let us note, however, that given the uneven representation of values 0 and 1, this value is not very informative, because even the naive model always assuming choice 1 (i.e. with zero specificity!)) will be able to classify the vast majority of observations correctly). The more accurate indicators, i.e. the sensitivity and specificity of the chosen model with a threshold value maximizing Youden's J-index, are relatively high (over 85%), which already signifies the very good classification ability of this model both for cases of the value 0 = "bad" and 1 = "good" explained variables ("What is your current life situation?").

If the logit model is supplemented by a simple descriptive analysis of the results of the questionnaire survey, it can be stated that most Czech seniors (52.23%) assess the

96

current life situation as good or very good (see the graph in Fig. 4), more than forty percent (43.86%) perceive the situation as acceptable, only a marginal part of the respondents evaluate their life situation as bad or very bad (3.91%) – note the consistency with the aforementioned statement that the satisfaction with the life situation is relatively high for seniors, as indicated, for example, by the assumption of the U-curve describing the relationship between satisfaction and age, with the bottom at middle-age.

4 Discussion and conclusion

The results of secondary data analysis show that Czech seniors are mostly satisfied with their life situation – the most satisfied are seniors aged between 60 and 65 years. Most seniors in the Czech Republic, however, expect their life situation to remain unchanged (about 46%) or worsen (about 20%). Discontent can be traced, according to EU-SILC data, especially to the area of the financial situation. The qualitative surveys showed that the seniors are generally satisfied with their life situation (it can even be observed that the expectation of a change in the current life situation is not desirable) and all the groups surveyed in the primary data survey indicate that their overall life situation will be determined by the development of their state of health. This is in line with the findings of Gabriel and Bowling (2008).

Good relationships in the family have a positive effect on the feelings of satisfaction. Similar results can be found in Bryla et al. (2013), who consider participation in family meetings as a factor that significantly contributes to the quality of life of seniors. This is consistent with our findings that a significant negative effect on the feeling of satisfaction is also manifested in the fact of whether a seniors citizen lives alone or in a social care home.

Analyzing quantitative primary data has made it possible to refine the perception of the current life situation of seniors and the prediction of their development. The results are in line with the findings based on EU SILC survey data. Most seniors (52.23%) assess their life situation as good or very good, while on a positive note 22.76% of those who perceive their life situation positively think that it will not change and only 12.34% of these seniors are expecting a worsening of the life situation. It is possible to declare that, among Czech seniors, it is not possible to prove the influence of gender on the perception of the current life situation, which is in line with results in Borg et al. (2008) and in contrast with Gaymu and Springer (2010), who argued that the factors that can most affect life satisfaction are not the same for men and women. For Czech seniors, however, only a relationship between gender and the assessment of the development of the life situation was identified, when men are clear that their life situation will not change.

Logit model was used to identify factors that influence the perception of the life situation in more detail.

The significant and positive impact of the health status of seniors is confirmed, this finding is also supported by e.g. the results of Gaymu and Springer (2010) and Berg et al. (2009).

The most striking negative effect on the life satisfaction is whether some seniors person lives in a home for seniors, in social facilities or in a rented apartment. This

97

finding is in line with the conclusions of Borg et al. (2008), who point to the higher satisfaction of respondents living in a regular home. The fact whether an individual lives in his/her home alone also negatively affects the satisfaction of Czech seniors. This finding is in line with assertion in Kucharova (2002), that the feeling of loneliness leads to the fact that the person does not feel safe, and if the need for safety is not met as one of the key needs, it will affect the perception of the overall life situation negatively.

Although it is quite difficult to generalize the results on a European or global scale, since according to Bourque et al. (2005), life satisfaction determinants may differ in cultural context, and Gaymu and Springer (2010) also point out that the factors that most affect life satisfaction are not the same for the European countries under consideration, there is some substantial consistency with the existing results, especially the fact that health determinants and family relationships (for which cohabitation with a partner and place of residence can be beneficial proxies) are the determining factors for seniors.

Thus, it may be questioned whether the functionalist approach of government policies, often oriented towards the construction of social facilities, or social benefits aimed at adjusting the income situation of seniors are indeed what will lead to the greater satisfaction of their intended recipients.

Acknowledgement

This contribution was supported by TAČR Beta project No. TB05MOSV004 "Survey on the Quality of Life of Elderly Citizens in the Czech Republic".

References

Baird, B. M., Lucas, R. E., Donnellan, M. B. (2010). Life satisfaction across the lifespan: Findings from two nationally representative panel studies. Social Indicators Research, 99 (2), pp. 183–203.

Berg, A. I., Hassing, L. B., Nilsson, S. E., Johansson, B. (2009). “As long as I’m in good health”. The relationship between medical diagnoses and life satisfaction in the oldest-old. Aging Clinical and Experimental Research, Volume 21 (4–5), pp. 307–313.

Blanchflower, D., Oswald, A. (2008). Is well-being U-shaped over the life cycle? Social Science & Medicine, 66 (8), pp. 1733–1749.

Borg, C., Hallberg, I., Blomqvist, K. (2006). Life satisfaction among older people (65+) with reduced self-care capacity: the relationship to social, health and financial aspects. Journal of Clinical Nursing, 15 (5), pp. 607–618.

Borg, C., Fagerström, C., Balducc, C., Burholt, V., Ferring, D., Weber, G., Wenger, C., Holst, G., Hallberg, I. (2008). Life Satisfaction in 6 European Countries: The Relationship to Health, Self-Esteem, and Social and Financial Resources among People (Aged 65-89) with Reduced Functional Capacity. Geriatric Nursing, 29 (1), pp. 48 – 57.

Bourque, P., Pushkar, D., Bonneville, L., Béland, F. (2005). Contextual effects on life satisfaction of older men and women. Canadian Journal of Aging, 24 (1), pp. 341-44.

Bryła, M., Burzyńska, M., Maniecka-Bryła, I. (2013). Self-rated quality of life of city-dwelling seniors benefitting from social help: results of a cross-sectional study [online]. Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3819270/ [Accessed 06. 23. 2017].

Cullati, S., Rousseaux, E., Gabadinho, A., Courvoisier, D. S., Burton-Jeangros, C. (2014). Factors of Change and Cumulative Factors in Self-Rated Health Trajectories: A Systematic Review. Advances in Life Course Research, 19, pp. 14–27.

98

Čevela, R. Kalvach, Z., Čevelová, L. (2012). Sociální gerontologie: úvod do problematiky. Praha: Grada.

Cheng S. T., Chan, A. (2006). Relationship with others and life satisfaction in later life: do gender and widowhood make a difference? The Journals of Gerontology: Series B, 61 (1), pp. 46 -53.

Diener, E., Suh, E. M. (1998). Subjective well-being and age: An international analysis. In K. W. Schaie & M. P. Lawton (Eds.), Annual review of gerontology and geriatrics, 17, pp. 304–324. New York: Springer.

Dingemans, E., Henkens, K. (2014). Involuntary retirement, bridge employment, and satisfaction with life: A longitudinal investigation. Journal of Organizational Behavior, 35 (4), pp. 575-591.

Gabriel, Z., Bowling, A. (2004). Quality of life from the perspectives of older people. Aging and Society, 24 (5).

Gana, K., Bailly, N., Saada, Y., Joulain, M., Trouillet, R., Hervé, C., Alaphilippe, D. (2013). Relationship Between Life Satisfaction and Physical Health in Older Adults: A Longitudinal Test of Cross-Lagged and Simultaneous Effects. Health Psychology, 32 (8), pp. 896–904.

Gaymu, J., Springer, S. (2010). Living conditions and life satisfaction of older Europeans living alone: a gender and cross-country analysis. Ageing and Society, 30 (7), pp. 1153 – 1175.

Hamarat, E., Thompson, D., Steele, D., Matheny, K., Simons, C. (2002). Age differences in coping resources and satisfaction with life among middle-aged, young-old, and oldest-old adults. Journal of Genetic Psychology, 163, pp. 360–367.

Hasmanová Marhánková, J. (2017). Seductive Solutions, Inspiration, Easy-to-Remember Phrases, and Ambiguity: Why Is the Idea of Active Ageing so Successful? In Ł. Tomczyk & A. Klimczuk (Eds.), Selected Contemporary Challenges of Ageing Policy (7-25). Kraków: Uniwersytet Pedagogiczny w Krakowie.

Hao, Y. (2008). Productive activities and psychological well-being among older adults. Journal of Gerontology: Social Sciences, 63B, pp. 64 – 72.

Hnilicová, H. Bencko, V. (2005). Kvalita života vymezení pojmu a jeho význam pro medicínu a zdravotnictví. 2005. Praktický lekář, 85 (11), pp. 656-660.

Kolosnitsyna, M., Khorkina, N., Dorzhiev, H. (2017). Determinants of Life Satisfaction in Older Russians. Ageing International, pp. 1–20.

Kuchařová, V. (2002). Život ve stáří. Zpráva o výsledcích empirického šetření. Praha: SOCIOKLUB – VÚPSV.

Mitnitski, A., Howlett, S. E., Rockwood, K. (2017). Heterogeneity of Human Aging and Its Assessment. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 72 (7), pp. 877–884.

Pinquart, M., Sörensen, S. (2000). Influences of socio-economic status, social networks and competence on subjective well being in later life: a meta-analysis. Psychology and Ageing, 15 (2), pp. 187-224.

Pinquart, M., Sörensen, S. (2001). Gender differences in self-concept and psychological well being in old age: a meta-analysis. Journals of Gerontology: Psychological Sciences, 56B (4), pp. 195-312.

Šolcová, I., Kebza, V. (2004): Kvalita života v psychologii: Osobní pohoda (well- being), její determinanty a prediktory In: Kvalita života, pp. 21-32.

Walker, A. (2005). A European perspective on quality of life in old age. European Journal of Ageing, 2 (1), pp. 2-12.

Youden, W. J. (1950). Index for rating diagnostic tests. Cancer, 3 (1), pp. 32-35.

99

Contact Address

doc. Ing. Lea Kubíčková, Ph.D. Mendel University in Brno, Faculty of Business and Economics, Department of Marketing and Trade Zemědělská 1, 613 00, Brno, Czech Republic Email: [email protected] Phone number: +420545132323

Ing. Vladimír Hajko, Ph.D. Mendel University in Brno, Faculty of Business and Economics, Department of Business Economics Zemědělská 1, 613 00, Brno, Czech Republic Email: [email protected] Phone number: +420545132636

Mgr. Ing. Naďa Hazuchová, Ph.D. Mendel University in Brno, Faculty of Business and Economics, Department of Marketing and Trade Zemědělská 1, 613 00, Brno, Czech Republic Email: [email protected] Phone number: +420545132333

doc. PhDr. Martina Rašticová, Ph.D. Mendel University in Brno, Faculty of Business and Economics, Department of Law and Social Sciences Zemědělská 1, 613 00, Brno, Czech Republic Email: [email protected] Phone number: +420545132644


Appendix

Tab. 3: Variable names and encoding1 Gender Age Family status Housing Who do

you live with?

What is your highest attained education?

Are you currently employed?

X1A X2 X3 X4 X5 X6 X8A Male Less than 65 Widowed Social

home Partner Vocational

education No

Female 66-70 Divorced Rented apartment

*Husband/wife

High school Yes

1 Categorical variables: X3, X4, X5, X6, X10, X116, X117, note that * indicates the default category; Binary (dummy) variables: X1, X8, X11, X34, X35, X101, Ysafe, X115 Ordinal variables: X2, X39, X49, X100,

100

71-75 Single Rented house

Family/next of kin

University/college

76-80 *Married/registred partnership

Own apartment

Alone *Elementary

81 or more *Own house

Alone in social home

Alone in a rented place Type of settlement where you lived for the majority of your life?

Do you consider moving?

Are you aware of the possibility of voluntary work participation?

Do you participate in voluntary work?

How is your health?

Do you have a health condition limiting your activities?

Financial sufficiency (higher is better)

X10 X11A X34A X35A X39 X49 X100 Capital city No No No Very bad No 1 NUTS3 (region) city center

Yes Yes Yes Bad Yes, but it is not limiting

2

*City/town Average Yes and it is somewhat limiting

3

Village Good Yes and it is very limiting

4

Very good

5 6 7 8 9 10

Do you depend on social benefits?

How do you expect your overall life situation will develop?

How did you perceive retirement?

Do you feel safe?

How do you perceive your overall life satisffaction?

X101A X116 X117 YsafeA X115A No I cannot say Good No Bad Yes *Will not

change I cannot identify with it so far

Yes Good

It will get worse

As a natural part of life

It will get better

*Bad


101

ELEKTRONIZACE VEŘEJNÉ SPRÁVY JAKO NÁSTROJ OMEZOVÁNÍ KORUPCE

ELECTRONIZATION OF PUBLIC ADMINISTRATION AS A TOOL FOR THE CORRUPTION REDUCTION

Veronika Linhartová, Veronika Tvrdíková

Abstract: The paper deals with the influence of electronization on the reduction of corruption in public administration. According to a number of studies, corruption in public administration has an impact on a wide range of taxpayers. They may feel the consequences of corrupt behaviour, for example in the form of tax and price increases or poor quality of service. For this reason, corruption is considered one of the most serious problems as it affects a large part of society. A number of empirical studies have shown that one of the ways to limit corruption is the electronization of public administration. For analysis, a group of 113 countries from all over the world were used, regardless of their geographical jurisdiction or political establishment. The time period is from 2003 to 2017. In order to fulfil the goal, the graphical interpretation of Bag plots and linear regression analysis is used to assess the impact of selected variables on the level of corruption. The multidimensional regression model has confirmed the positive effect of electronization. However, the most significant effect on the level of corruption in the country had the performance of the economy in the examined period and selected set of countries.

Keywords: Corruption, Electronization, Bag Plots, Regression Analysis, CPI, EGDI.

JEL Classification: D73, E 6, H11.

Úvod

Korupce ve veřejné správě dopadá na široký okruh daňových poplatníků. Ti mohou pociťovat důsledky korupčního jednání například ve formě zvyšování daní a cen nebo nízké kvality poskytovaných služeb (Linhartová, V., Židová, E., 2016; Gans-Morse, J. et al. 2018). Právě z důvodu natolik širokého dopadu je považována za jeden z nejzávažnějších problémů společnosti (Mauro, 1995, 1996; Leite, Wedeman, 1999; Tanzi, Davoodi, 1997, 2000; Volejníková, 2012).

V obecném smyslu ovlivňuje možnost vzniku korupce řada socio-kulturních, ekonomických i politických faktorů. Mezi jinými to jsou výše úplatku, systém sankcí, korupční tradice, sociální chaos, hodnotový žebříček společnosti, chudoba, korupční vzory, vliv médií, kvalita veřejné správy, výše platů státních úředníků či normativní systém fungování veřejné správy (Chmelík, 2003). Řada empirických studií prokázala, že jednou z možností, jak omezit korupci, je elektronizace veřejné správy, která může některé z výše uvedených faktorů eliminovat a tím pozitivně působit na míru korupce v zemi (například Kimbro, 2002; Andersen, Rand, 2006; Kim, 2007; Bhatnagar, 2008; Shim, 2008; Andersen, 2009; Mistry a Jalal, 2012; Elbahnasawy, 2014).

Příspěvek se věnuje elektronizaci veřejné správy z hlediska jejího využití v oblasti omezování korupce ve veřejné správě. Cílem příspěvku je ověřit vliv elektronizace veřejné správy na omezování míry korupce ve veřejné správě ve zvoleném souboru 113 zemí v časovém období let 2003 až 2017. Index rozvoje e-governmentu (EGDI),

102

který kvantifikuje míru elektronizace veřejné správy v zemi, byl sestaven poprvé v roce 2003 a naposledy publikován v roce 2017. S ohledem na využité indikátory kvantifikující míru korupce v zemi a úroveň elektronizace veřejné správy se tak jedná v současnosti o nejširší dostupný datový soubor.

1 Vliv elektronizace na korupci ve veřejné správě

Dle řady studií (například Kimbro, 2002; Andersen, Rand, 2006; Kim, 2007; Bhatnagar, 2008; Shim, 2008; Andersen, 2009; Mistry a Jalal, 2012; United Nations, 2016) může digitalizace veřejné správy pozitivně působit na míru korupce ve veřejné správě. Jak uvádí United Nations (2018), elektronická vláda představuje využívání informačních technologií vládními agenturami (např. širokoplošné sítě, internet a mobilní výpočetní technika), které mohou transformovat vztahy s občany, podniky a dalšími vládními orgány. Tyto technologie mohou sloužit řadě různých cílů, jako je například lepší poskytování veřejných služeb občanům, lepší interakce s podniky a průmyslem, posílení občanské participace prostřednictvím přístupu k informacím nebo efektivnější vládní řízení. Výsledným přínosem může být méně korupce, větší transparentnost, větší pohodlí, růst příjmů i snížení nákladů na správu věcí veřejných. Díky elektronizaci dochází ke snižování kontaktu mezi úředníky a občany. To může mít za následek větší transparentnost činností, která omezuje možnost veřejného úředníka přijímat či dokonce vyžadovat úplatek. Elektronizace snižuje korupci externě tím, že zlepšuje vztahy s občany a interně účinnější a dostupnější kontrolou. Transparentnost může být zajištěna za předpokladu, že právní rámec země podporuje volný přístup k informacím. Mnoho zemí světa mělo v minulosti přísné zákony o státním tajemství, které byly zrušeny ve prospěch zákona o svobodě informací, a to zejména v Spojených státech amerických a v zemích Evropy. Se zvyšujícím se přístupem občanů k informacím musí vláda rovněž řešit rizika spojené se soukromím a bezpečností občanů (Bhatnagar, 2003).

Elbahnasawy (2014) konkrétně uvádí, že při udržení ostatních faktorů na konstantní úrovni, zvýšení elektronizace o jednu standardní odchylku (0,2 bodové zvýšení indexu elektronizace) vede ke snížení vnímání korupce o 0,25 až 0,43 bodu. Autor dále uvádí i další faktory, které korupci také ovlivňují. Například se jedná o výkonnost ekonomiky (měřená pomocí HDP). Každé zvýšení HDP o jeden standardní vzestup (12 739 dolarů na obyvatele) snižuje korupci o 0,04 až 1,01 bodu. Vliv má také rozsah a kvalita online služeb, kdy zvýšení standardní odchylky (0,24 bodu) snižuje korupci o 0,15 bodu. Korupci výrazně ovlivňuje také kvalita právního prostředí, která zahrnuje úroveň práva nebo justici sloužící mocným k prosazování vlastních zájmů (Katsios, 2015). Zvýšení standardní odchylky rule-of-law indexu (o 1,01 bodu) snižuje korupci o 0,45 až 0,61 bodu (Elbahnasawy, 2014).

2 Metody

Cílem příspěvku je analyzovat vliv elektronizace veřejné správy na omezování korupce. Pro analýzu byla použita skupina 113 zemí celého světa bez ohledu na jejich geografickou příslušnost, politické zřízení či ekonomickou vyspělost. Seznam zemí zahrnutých do analýzy je obsahem tabulky v příloze 1 příspěvku. S ohledem na využité indikátory kvantifikující míru korupce v zemi a úroveň elektronizace veřejné správy se jedná v současnosti o nejdelší časovou řadu a nejširší soubor zemí

103

umožňující ověřit platnost hypotézy o pozitivním vlivu elektronizace veřejné správy na korupci ve veřejném sektoru.

Jako ukazatel míry korupce byl pro analýzu zvolen Index vnímání korupce (CPI). CPI je zveřejňován od roku 1995 organizací Transparency International. Jedná se o Index, který je založen na vnímání korupce respondenty šetření, jimiž jsou domácí i zahraniční podnikatelé, analytici a zástupci odborné veřejnosti z hodnocených zemí. Index je publikován jednou ročně a průzkumy použité při jeho sestavování většinou obsahují otázky zaměřené na uplácení veřejných činitelů nebo provize při veřejných zakázkách. Vzorek zkoumaných zemí se v průběhu času měnil. Například v roce 1995 bylo zahrnuto 41 zemí, v roce 2017 jich bylo již 180. Měnící se počet hodnocených zemí je důvodem k tomu, že při posuzování jednotlivých zemí není důležité jejich pořadí v žebříčku, ale samotná hodnota CPI. V roce 2012 došlo ke změně škály hodnocení u indexu CPI (dříve od 0 do 10, nyní od 0 do 100). Z důvodu potřeby dlouhodobějšího srovnání bylo pro analýzu u dat týkajících se míry korupce od roku 2012 převedeno nynější hodnocení na škálu dřívějšího hodnocení, tedy od 0 do 10, kde 0 představuje zemi velmi zkorumpovanou a hodnota 10 značí zemi bez korupce (Transparency International, 2017, 2018).

Jako ukazatel hodnotící úroveň elektronizace v zemi byl zvolen Index rozvoje e-governmentu (EGDI). EGDI byl sestaven poprvé v roce 2003 a nabývá intervalu <0;1>, kde hodnota 1 představuje vysokou úroveň využití metod e-governmentu a nulová hodnota Indexu značí naopak nízkou míru aplikace zmiňovaných metod v oblasti veřejné správy. Měření EGDI je založeno na průzkumech sestavených za spolupráce OSN-UNDESA (United nations Department of Economic and Social Affairs) a Civic Resource Group, poradenskou firma poskytující technologická řešení v oblasti e-governmentu (UNITED NATIONS, 2016). Přestože korupce není primární složkou, kterou by E-Government Survey zkoumal, jak uvádí Unitied Nations ve své zprávě z roku 2018, jednou z hlavních složek kvalitní digitální správy je „Odpovědnost (Acconutability)“, která například kromě volného přístupu občanů k informacím či otevřené správy zahrnuje právě úroveň kontroly korupce v zemi.

V provedených analýzách je nejprve zkoumán vztah mezi EGDI a CPI ve dvou samostatných časových obdobích (tj. 2003 a 2017). Následně je zkoumáno, zda existuje vztah mezi změnou v EGDI a změnou CPI během tohoto období, tedy jak se změnily hodnoty zmíněných indexů ve sledovaném souboru zemí mezi roky 2003 a 2017. Sledování změn ve využití e-Governmentu a míry korupce ve veřejné správě umožní zjistit, zda jednotlivá opatření e-Governmentu během uvedených čtrnácti let ovlivnila míru korupce, třebas s určitým časovým zpožděním.

Za účelem naplnění cíle příspěvku bude využita grafická interpretace tzv. Bag plots a lineární regresní analýza, s jejíž pomocí lze posoudit vliv vybraných proměnných na míru korupce. Všechna testování jsou prováděna s 5% hladinou významnosti. Využité odhady parametrů jsou vyvozeny z metody nejmenších čtverců.

Bag plots jsou generalizované dvoudimenzionální grafy, které slouží ke grafické interpretaci statistických dat. Kombinace závisle a nezávisle proměnné jednotlivých zemí představují body v grafu. Ve vnitřní tmavé části leží 50 % pozorování (mezi prvním a třetím kvartilem) a medián zkoumaných pozorování, který je označen tmavým čtvercem. Vnější oblast bagu tvoří světlá část obsahující ostatní státy, které mají více odlišné hodnoty než státy v tmavé oblasti, nejsou to však odlehlé hodnoty.

104

Mimo tyto dvě oblasti leží odlehlé hodnoty, ty jsou označeny hvězdičkou. Graf také ukazuje další charakteristiky, a to například pozici státu v rámci hodnocených zemí. Orientace bagu ukazuje vztah mezi hodnocenými proměnnými. Pozitivní vztah mezi hodnocenými proměnnými lze předpokládat při rostoucím bagu, negativní vztah naopak při bagu klesajícím (Rousseeuw, P. J., Ruts, I., Tukey, J. W.,1999).

Lineární regresní analýza je metoda, pomocí níž se odhaduje hodnota závislé proměnné na základě znalosti nezávisle proměnných. V případě jedné nezávisle proměnné se jedná o jednoduchou regresi, která popisuje vztah mezi závisle proměnnou a jednou nezávisle proměnnou (tzv. regresor). Vícenásobná regrese na rozdíl od jednoduché se využívá v případě více nezávislých proměnných, tedy když závislá proměnná závisí na dvou a více regresorech. Vzorec lineární regrese lze vyjádřit následovně (Víšek, 1998):

y= α+ β1*x1+ β2*x2+⋯+βn*xn+ε (1)

Parametr y je závisle proměnná, parametry x značí nezávisle proměnné. Alfa (α) určuje vzdálenost průsečíku regresní přímky s osou y od počátku souřadnic (hodnota regresní funkce pro x = 0). Beta (β) jsou regresní koeficienty, které udávají, o kolik se změní závisle proměnná, když se zvětší hodnota nezávisle proměnné o jednotku. Index n značí počet nezávislých proměnných. Symbol ε značí reziduální rozptyl. Jedná se o grafické znázornění vzdálenosti bodů od přímky. Pro úspěšnou interpretaci vícenásobné regresní analýzy musí data splňovat základní předpoklady, které zajistí nezkreslené výsledky analýzy. (Víšek, 1998).

Odhad míry korupce v závislosti na hodnocení elektronizace pomocí jednoduché lineární regrese má tvar (Andersen, 2009; Andersen, Rand, 2006):

CPI= α+ β* EGDI+ ε, (2)

kde CPI je hodnocení míry korupce v zemi a EGDI je hodnocení elektronizace země.

Odhad změny v míře korupce vyvolaný změnou elektronizace lze zapsat jako následující model:

∆CPI= α+ β* ∆EGDI+ ε, (3)

kde ΔCPI je změna v míře korupce a ΔEGDI je změna v hodnocení elektronizace.

3 Rozbor problému

Ke grafickému znázornění vztahu mezi úrovní elektronizace a mírou korupce ve veřejné správě budou použity grafy v podobě Bag plots. Následující Obr. 1 zobrazuje Bag plot zvoleného souboru zemí. Využita jsou data z roku 2003. Na ose x je nanesen index hodnotící elektronizaci v zemi EGDI 2003, na ose y je index hodnotící míru korupce v zemi CPI 2003.

Lepší schopnost využívání informačních technologií ve veřejné správě je vyjádřena vyššími hodnotami EGDI. Nižší míra korupce v zemi se naopak vyznačuje vyššími hodnotami indexu CPI. Na první pohled viditelný pozitivní sklon bag plotu potvrzuje pozitivní vztah mezi mírou korupce a úrovní elektronizace ve zvoleném souboru zemí v roce 2003. V bag plotu je několik odlehlých zemí, ať už z pohledu hodnocení míry korupce či využívání elektronizace. Nejvíce zaostávající zemí v oblasti korupce je Bangladéš, která v roce 2003 dosáhla v hodnocení míry korupce hodnoty jen 1,3

105

z možných 10. Dalšími zaostávajícími zeměmi jsou například Sierra Leone, Etiopie nebo Mali. Spojené státy americké předběhly ostatní země v hodnocení elektronizace veřejné správy. Za rok 2003 hodnocení elektronizace v této zemi dosáhlo s velkým náskokem hodnoty 0,927, druhá nejlepší země Švédsko má hodnotu 0,840. Dle hodnocení míry korupce předčilo ostatní Finsko, a to s nejvyšší hodnotou 9,7 z 10.

Obr. 2 zobrazuje Bag plot zvoleného souboru zemí, pro data z roku 2017. Na ose x je naneseno hodnocení elektronizace EGDI 2017 a na ose y hodnocení míry korupce CPI 2017. U proměnných pro rok 2017 byl rovněž prokázán pozitivní vztah mezi mírou korupce a elektronizace ve veřejné správě, který je zřejmý z pozitivního sklonu. Bag plot opět vykazuje několik odlehlých hodnot. Nejvíce odlehlou zemí, co se týče využití elektronizace ve veřejné správě, je Bahrajn a Mali. I přes nízké hodnocení míry korupce dosahující hodnoty pouze 3,6 bodů, došlo v případě Bahrajnu ke zlepšení hodnocení elektronizace v oblasti veřejné správy na hodnotu 0,812. Oproti tomu země Mali je odlehlým bodem kvůli velmi nízkému využití elektronizace ve veřejné správě, což dokazuje nejnižší hodnota ze všech zkoumaných zemí 0,242. Dle hodnocení míry korupce je nejvíce odlehlou zemí Súdán, a to s hodnotou Indexu vnímání korupce jen 1,6.

Obr. 1: Bag plot CPI 2003 vs. EGDI 2003 Bag plot z CPI 2003 proti EGDI 2003

CPI 2003 Medián Odlehlé hodnoty

0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

EGDI 2003

0

2

4

6

8

10

12

CP

I 20

03

Argentina

Austrálie

Bangladéš

BotswanaEstonsko

Etiopie

Filipíny

Finsko

Indonésie

Island

Itálie

Lucembursko

MaliMexiko

Nový Zéland

Omán

Paraguay

Sierra Leone

Spojené státy americké

Zdroj: vlastní zpracování, Statistica 2012

Obr. 2: Bag plot CPI 2017 vs. EGDI 2017 Bag plot z CPI 2017 proti EGDI 2017

CPI 2017 Medián Odlehlé hodnoty

0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

EGDI 2017

1

2

3

4

5

6

7

8

9

10

CP

I 20

17

Angola

Bahrajn

Botswana

Kazachstán

Malajsie

Malawi

Mali

Mexiko Rusko

Senegal

Sierra Leone

Súdán


Následující Obr. 3 zobrazuje Bag plot zvoleného souboru zemí pro změny CPI a EGDI v letech 2003 až 2017. Na ose x je nanesena změna hodnocení elektronizace veřejné správy ΔEGDI a na ose y změna hodnocení míry korupce ΔCPI. Pro změny v hodnotách lze pozorovat mírný pozitivní sklon bag plotu, tudíž lze předpokládat pozitivní vztah mezi proměnnými, tj. elektronizací a mírou korupce ve veřejné správě. Graf obsahuje poměrně velké množství odlehlých hodnot. Nejvíce odlehlá hodnota patří zemi Bangladéš. Ta dosáhla největších změn v obou proměnných. Změna míry korupce ve veřejné správě mezi lety 2003 až 2017 byla 115 %, změna hodnocení elektronizace dosáhla téměř 195 %. Další výrazněji odlehlou zemí je Etiopie. Ve sledovaném období došlo v Etiopii k druhé nejvyšší změně hodnocení elektronizace, a to o 170 %. Mezi odlehlé hodnoty se dále řadí například Malajsie, Bahrajn, Sudán nebo Kuvajt, to především kvůli negativní změně v hodnocení míry korupce ve sledovaném období.

106

Obr. 3: Bag plot ΔCPI vs. ΔEGDI Bag plot z ΔCPI proti ΔEGDI

ΔCPI Medián Odlehlé hodnoty

-20 0 20 40 60 80 100 120 140 160 180 200 220

ΔEGDI 2003-2017

-60

-40

-20

0

20

40

60

80

100

120

140

ΔC

PI 2

003-

2017

AlbánieArgentina

Ázerbajdžán

Bahrajn

Bangladéš

Egypt

Etiopie

Ghana

Chorvatsko

Indonésie

Kanada Kuvajt

Malajsie

Nigérie

Omán

Papua‐Nová Guinea

Paraguay

PolskoRumunsko

Saudská arábie

Sierra Leone

Spojené státy americké

Srbsko

Súdán

Tunisko


K odhadu změny míry korupce v zemi na základě znalosti změny využití elektronizace veřejné správy bude využita jednoduchá lineární regrese. Popisuje závislost závisle proměnné (míra korupce) na nezávislé proměnné (využití elektronizace). Odhad procentní změny v míře korupce (ΔCPI) vyvolaný procentní změnou elektronizace (ΔEGDI) v letech 2003 až 2017 lze zapsat jako následující model:

ΔCPI = α + β * ΔEGDI + ε, (4)

kde ΔCPI je změna v hodnocení míry korupce v letech 2003 až 2017 a ΔEGDI je změna v hodnocení elektronizace ve veřejné správě ve stejných letech.

V příloze 2 je zobrazen lineární regresní model, který znázorňuje změny míry korupce a elektronizace ve zvoleném souboru zemí mezi lety 2003 a 2017. Na vodorovné ose je procentní změna EGDI mezi lety 2003 a 2017 a na svislé ose procentní změna CPI. Data jsou proložena lineární regresní přímkou, která je doplněná o pás spolehlivosti. Zemí, která dosáhla největšího pokroku v obou analyzovaných proměnných, byla ve sledovaném období Bangladéš (č. 9). U Malajsie (č. 61) či Bahrajnu (č. 8) došlo i přes zvýšení využití IT v oblasti veřejné správy ke zhoršení hodnocení míry korupce v zemi. Z grafického lineárního regresního modelu je zřejmé, že vyšší využití IT v oblasti veřejné správy nešlo ve sledovaném období u všech zemí ruku v ruce s nižší mírou korupce. I přesto model prokázal, že při neměnných podmínkách jednoprocentní změna v hodnocení elektronizace veřejné správy vedla ve sledovaném období a analyzovaném souboru zemí k nárůstu hodnocení míry korupce o více než 0,2 %. Uvedený regresní model vykazuje nízkou hodnotu koeficientu determinace, R2 = 0,079787. Na základě toho lze tvrdit, že variabilita vysvětlované proměnné (ΔCPI) je vysvětlena zmíněným modelem pouze z přibližně 8 %.

Odhadnutý model vysvětlující vztah mezi mírou korupce a elektronizací ve veřejné správě má tvar:

ΔCPI = 4,2577 + 0,238 * ΔEGDI + ε (5)

107

Jednoduchá lineární regrese prokázala určitou závislost změny míry korupce na změně ve využití elektronizace ve veřejné správě, přičemž při zlepšení hodnocení elektronizace za pomocí indexu EGDI o 1 % se zlepšuje hodnocení míry korupce dle CPI průměrně o 0,238 %. Nízký podíl vysvětlené variability naznačuje, že na změnu CPI mají vliv rovněž další v analýze neuvažované faktory. Z tohoto důvodu je vhodné tento model rozšířit o další vysvětlující proměnné. Odhad koeficientu závislosti změny míry korupce na změně využívání elektronizace vyšel statisticky významný (nenulový), tudíž proměnná ΔEGDI má v modelu své opodstatnění. Na základě studií (Elbahnasawy, 2014, Katsios, 2015) byly pro další analýzu určeny další 2 faktory, u kterých byl prokázán vliv na míru korupce v zemi. Tyto faktory jsou výkonnost ekonomiky země a kvalita právního prostředí, měření pomocí HDP a rule-of-law indexu. V následující vícerozměrné regresní analýze budou využity všechny tři nezávisle proměnné. Odhad změny v míře korupce (ΔCPI) vyvolaný změnami elektronizace (ΔEGDI), změnami výkonnosti ekonomiky (ΔHDP) a změnou stavu právního systému (Δrule-of-law index) v letech 2003 až 2017 lze zapsat jako následující model:

(6)

Následující Tab. 2 s výslednými hodnotami vícenásobné regrese za roky 2003 až 2017 zobrazuje závislost procentní změny míry korupce na procentní změně výše uvedených vysvětlujících proměnných. Výsledky z tabulky ukazují, že ve sledovaném období byla míra korupce ovlivňována především stupněm elektronizace veřejné správy a výkonností ekonomiky. Změna kvality právního prostředí neměla ve sledovaném období ve zvoleném souboru zemí zásadní vliv na míru korupce.

Tab. 2: Výsledky vícenásobné regrese pro ΔCPI vs. ΔEGDI, ΔHDP a Δrule-of-law index

N=113

R = 0,476809102, R2 = 0,227346919, p < 0,000003

b*

Sm.chybaz b*

b

Sm.chybaz b

t(109)

p-hodn.

Abs.člen

-11,0513 5,360245 -2,06172 0,041613

ΔEGDI

0,149709 0,089753 0,1263 0,075700 1,66802 0,098183

ΔHDP

0,404249 0,089785 0,2164 0,048068 4,50240 0,000017

Δrule-of-law

0,021126 0,084910 0,0000 0,000064 0,24880 0,803980


Model vícenásobné regrese má následující tvar:

(7) Z uvažovaných faktorů měla na míru korupce ve sledovaném období pro skupinu

analyzovaných zemí největší vliv výkonnost ekonomiky. Při 1 % změně v hodnocení výkonnosti ekonomiky došlo k 0,21% nárůstu v hodnocení míry korupce. Jednoprocentní změna v hodnocení elektronizace ve veřejné správě vedla k 0,12% růstu míry korupce.

4 Diskuze

Výsledky výše provedených analýz se v řadě závěrů shodují s empirickými studiemi (např. Andersen, 2009; Bhatnagar, 2008; Shim, 2008; Department of

108

Economic and Social Affairs, 2016; Andersen, Rand, 2006; Mistry and Jalal, 2012; Kimbro, 2002; Kim, 2007) potvrzujících možný pozitivní vliv elektronizace veřejné správy na míru korupce ve veřejném sektoru. Provedená jednoduchá regresní analýza rovněž prokázala tvrzení Elbahnasawy (2014) a Katsiose (2015), že prosté omezení diskreční pravomoci úředníků veřejné správy a není všelékem na boj s korupcí a tento fenomén je ovlivňován řadou dalších faktorů.

Již desítky let je zejména problematika vztahu korupce a ekonomické výkonnosti zemí předmětem řady teoretických a empirických studií. Z bag plotů je zřejmé, že ekonomicky vyspělé státy zpravidla vykazují v hodnocení indexu CPI vyšších hodnot, rovněž tak jsou lépe hodnocené z hlediska využití IT technologií v oblasti veřejné správy. Výsledky vícenásobné regrese pro celý soubor zkoumaných zemí pozitivní vliv výkonnosti ekonomiky na míru korupce v zemi skutečně prokázaly. V případě vztahu mezi změnou mírou korupce a změnou kvality právního prostředí výsledky naznačují, že mezi těmito proměnnými neexistuje vztah a kvalita právního prostředí nemá v analyzovaných zemích zásadní vliv na stav korupce v zemi. Takový závěr provedených analýz je naopak v rozporu se závěry studií autorů Elbahnasawy (2014) a Katsios (2015) o možném snižování korupce prostřednictvím kvality právního systému.

V rámci hodnocených zemí i mezi jednotlivými kontinenty existují značné rozdíly v přístupu k elektronizaci veřejné správy (Lněnička, M., 2015). Nejnižších hodnot dlouhodobě dosahuje průměr zemí afrického kontinentu, naopak nejvyšší je průměrná hodnota evropských zemí. Velká Británie, Nizozemí, Estonsko, státy severní Evropy, ale také Austrálie či Nový Zéland se dlouhodobě snaží o co největší rozšíření digitálních veřejných služeb. Estonsko je známým průkopníkem e-voleb, které poprvé vyzkoušelo již v roce 2005. On-line způsob voleb, kdy se volič prokazuje občanským průkazem s digitálním čipem, využívá Estonsko dodnes (Ardielli, E., Halásková, M., 2015). Elektronizace veřejné správy je pochopitelně závislá na stádiu ekonomického rozvoje země (UNITED NATIONS, 2016). Zabezpečení přístupu ke službám v elektronické podobě, přechod k více "samoobslužnému" občanství, které v průběhu doby zmenší velikost státu, vyžaduje dlouhodobé investice ze strany státu. V provedené analýze však nejvýraznějšího zlepšení v míře korupce ve sledovaném období dosáhla především Bangladéš, která mezi ekonomicky vyspělé státy nepatří. Tu organizace Transparency International (2018b) pozitivně hodnotí zejména v souvislosti s přijetím antikorupční strategie, která se zaměřuje na „profesionální integritu a nestrannost veřejných služeb; robustní přístup k informacím; rychlejší, širší a hlubší digitalizaci veřejné správy; zvýšený prostor pro občany a jejich zapojování se do věcí veřejných, svobodu sdělovacích prostředků, občanskou společnost a podporu nevládních organizací.“

Je naivní se domnívat, že pouhá digitalizace veřejné správy je všelékem v oblasti snižování míry korupce. Korupce má mnoho příčin, které se v jednotlivých zemích výrazně liší v závislosti na historickém vývoji, geo-politické či ekonomické situaci. Zejména v oblasti korupce ve veřejné správě se však jedná o více než užitečný nástroj jejího omezování, který může fungovat bez ohledu na ekonomickou vyspělost či geografickou polohu zemí.

Závěr

Přestože korupce sužuje vlády jednotlivých zemí již odnepaměti, s bojem proti tomuto fenoménu si většina z nich stále příliš neví rady. Zejména korupce v oblasti

109

veřejné správy je z důvodu závažnosti svých důsledků i možné šíře poškozených subjektů předmětem bouřlivých diskuzí i aktivit v oblasti protikorupční politiky. V současnosti poměrně moderním nástrojem, který podle řady autorů může přispívat k omezování korupce ve veřejné správě prostřednictvím regulace diskreční pravomoci úředníků, transparentního poskytování a uchovávání informací i vyšší efektivnosti probíhajících procesů, je zavádění informačních a komunikačních technologií do procesů veřejné správy.

V příspěvku byla ověřována platnost tvrzení o pozitivním vlivu elektronizace na omezování korupce. Analýza byla provedena na souboru 113 zemí světa v letech 2003 až 2017. Z provedené regresní analýzy vyplynulo, že zlepšení hodnocení elektronizace veřejné správy ve sledovaném období a zvoleném souboru zemí skutečně vedlo ke snížení míry korupce. V provedených analýzách však bylo potvrzeno, že elektronizace veřejné správy není jediným faktorem, který ovlivňuje stav korupce ve veřejné správě v zemi, což je patrné z nízké hodnoty koeficientu R2 v provedené regresní analýze. Vícerozměrný regresní model zohledňující kromě míry elektronizace veřejné správy i výkonnost ekonomiky a kvalitu právního prostředí pozitivní vliv elektronizace rovněž potvrdil. Zásadnější vliv na míru korupce v zemi měla však ve sledovaném období a zvoleném souboru zemí výkonnost ekonomiky.

Provedená analýza vlivu působení elektronizace veřejné správy na korupční prostředí v zemi tak navazuje na již existující studie (např. Andersen, 2009; Bhatnagar, 2008; Shim, 2008; Department of Economic and Social Affairs, 2016; Andersen, Rand, 2006; Mistry and Jalal, 2012; Kimbro, 2002; Kim, 2007) a potvrzuje možné snižování míry korupce v zemi za pomoci využití metod e-governmentu. S ohledem na zjištěné výstupy příspěvek poukazuje potenciál e-governmentu jako součásti protikorupční strategie.

Reference

Ardielli, E., Halásková, M., 2015. Assessment of E-government in EU Countries. Scientific Papers of the University of Pardubice, 22(34), pp.4-16.

Andersen, T. B., 2009. E-Government as an anti-corruption strategy [online]. [cit. 2018-12-15]. Dostupné z: https://doi.org/10.1016/j.infoecopol.2008.11.003

Andersen, T. B., Rand, J., 2006. Does E-Government Reduce Corruption? [online]. [cit. 2018-12-16]. Dostupné z: https://pdfs.semanticscholar.org/2854/ab54b65a26236ab0e7752bdd89794eaf3ed6.pdf

Bhatnagar, S., 2003. Transparency and Corruption: Does E-Government Help? [online]. [cit. 2018-12-15]. Dostupné z: https://faculty.iima.ac.in/~subhash/pdfs/CHRIDraftPaper2003.pdf

Elbahnasawy, N. G., 2014. E-Government, Internet Adoption, and Corruption: An Empirical Investigation [online]. [cit. 2018-11-07]. Dostupné z: https://www.sciencedirect.com/science/article/abs/pii/S0305750X13002891

Gans-Morse, J. et al. 2018. Reducing bureaucratic corruption: Interdisciplinary perspectives on what works. World Development, 105, pp.171-188.

Chmelík, J., 2003. Pozornost, úplatek a korupce. Praha: Linde, ISBN 80-7201-434-X.

Κatsios, S., 2015. The shadow economy and corruption in Greece. South-Eastern Europe Journal of Economics, 4(1), pp. 61-80.

Kim, C., 2007. A cross-national analysis of global E-government, Public Organization Review, vol. 7, n. 10:.317-329. Dostupné z: http://dx.doi.org/10.1007/s11115-007-0040-5.

Kimbro, M., 2002. A cross-country empirical investigation of corruption and its relationship to economic, cultural, and monitoring institutions: an examination of the role of accounting and financial

110

statements quality, Journal of Accounting, Auditing, and Finance vol. 17, n. 4:.325-349. [online]. [cit. 2018-12-15]. Dostupné z: http://dx.doi.org/10.1177/0148558X0201700403.

Leite, C., Weidmann, J. Does Mother Nature Corrupt – Natural Resources, Corruption and Economic Growth. IMF Working Paper, Washington, D.C.: International Monetary Fund, [online]. 1999 [cit. 2019-03-18]. Dostupné z WWW: <http://www.imf.org/external/pubs/ft/wp/1999/wp9985.pdf>.

Linhartová, V., Židová, E., 2016. The impact of corruption on economic growth in oecd countries. Scientific Papers of the University of Pardubice. Series D, Faculty of Economics & Administration, 23(37).

Lněnička, M., 2015. E-government Development Index and its Comparison in the EU Member States. Scientific Papers of the University of Pardubice, p.75.

Mauro, P., 1995. Corruption and Growth, Quarterly Journal of Economics, Vol. 110, No. 3, pp. 681 – 712. ISSN 1531-4650.

Mauro, P., 1996. The Effects of Corruption on Investment, Growth and Government Expenditure. IMF Working Paper, Washington, D.C.: International Monetary Fund, [online]. [cit. 2019-03-29]. Dostupné z WWW: <http://www.imf.org/external/pubs/ft/wp/1996/wp0098.pdf>.

Mauro, P., 1997. Why Worry About Corruption?, IMF Economic Issues, Vol. 6, 19 p. ISBN 1-55775- 635-X.

Mistry, J. J., Jalal, A., 2012. An empirical analysis of relationship between e-government and corruption. The International Journal of Digital Accounting Research, 12, 145-176. [online]. [cit. 2018-12-25]. Dostupné z: http://www.uhu.es/ijdar/10.4192/1577-8517-v12_6.pdf.

Rousseeuw, P. J., Ruts, I., Tukey, J. W. ,1999, The Bagplot: A Bivariate Boxplot, The American Statistician, 53:4, 382-387, DOI: 10.1080/00031305.1999.10474494

Shim, D. Ch., Eom, T. H., 2008. E-Government and Anti-Corruption: Empirical Analysis of International Data [online]. [cit. 2018-12-15]. Dostupné z: https://www.tandfonline.com/doi/abs/10.1080/01900690701590553

Tanzi, V., Davoodi, H., 1997. Corruption, Public Investment, and Growth. IMF Working Paper, Washington, D.C.: International Monetary Fund, [online] [cit. 2019-03-29] Dostupné z WWW: <http://www.imf.org/external/pubs/ft/wp/wp97139.pdf>.

Tanzi, V., Davoodi, H., 2000. Corruption, Growth and Public Finances. IMF Working Paper, Washington, D.C.: International Monetary Fund, [online] [cit. 2019-03-29]. Dostupné z WWW: <http://www.imf.org/external/pubs/ft/wp/2000/wp00182.pdf>.

TRANSPARENCY INTERNATIONAL. CORRUPTION PERCEPTIONS INDEX 2017 [online], 2018. [cit. 2018-12-15]. Dostupné z: https://www.transparency.org/news/fea-ture/corruption_perceptions_index_2017#table

TRANSPARENCY INTERNATIONAL. Our organization [online]. 2018a [cit. 2018-07-11]. Dostupné z: https://www.transparency.org/whoweare/organisation

TRANSPARENCY INTERNATIONAL. Right to information in Asia Pacific [online]. 2018b [cit. 2019-03-11]. https://www.transparency.org/whatwedo/publication/right_to_information_in_asia_pacific

UNITED NATIONS E-GOVERNMENT SURVEY 2016 [online], 2016. [cit. 2018-12-11]. ISBN 978-92-1-123205-9. Dostupné z: http://workspace.unpan.org/sites/Internet/Documents/UNPAN96407.pdf

UNITED NATIONS E-GOVERNMENT SURVEY 2018 [online], 2016. [cit. 2019-03-03]. ISBN 978-92-1-123205-9. Dostupné z: http://workspace.unpan.org/sites/Internet/Documents/UNPAN96407.pdf

Vishny, R. W., Shleifer, A., 1993. Corruption. The Quarterly Journal of Economics, 108(3), 599 – 617.

Víšek, J. A., 1998. Statistická analýza dat. Praha: Vydavatelství ČVUT, ISBN 8001017354.

Volejníková, J., 2012. Korupce jako racionální ekonomická volba. Scientific Papers of the University of Pardubice - Series D, Faculty of Economics and Administration, 17 (2), pp. 147-159.

111

Kontaktní adresa

Ing. et Ing. Veronika Linhartová, Ph.D. Univerzita Pardubice, Fakulta ekonomicko-správní, Ústav ekonomických věd Studentská 84, 532 10 Pardubice, Česká republika E-mail: [email protected] Tel. číslo: +420466036126

Bc. Veronika Tvrdíková Univerzita Pardubice, Fakulta ekonomicko-správní, Ústav ekonomických věd Studentská 84, 532 10 Pardubice, Česká republika

Příloha č. 1: Seznam analyzovaných zemí 1. Albánie 30. Honduras 59. Madagaskar 88. Saudská Arábie

2. Alžírsko 31. Chile 60. Maďarsko 89. Senegal

3. Angola 32. Chorvatsko 61. Malajsie 90. Sierra Leone

4. Argentina 33. Indie 62. Malawi 91. Singapur

5. Arménie 34. Indonésie 63. Mali 92. Slovensko

6. Austrálie 35. Írán 64. Maroko 93. Slovinsko

7. Ázerbájdžán 36. Irsko 65. Mauricius 94. Spojené arabské emiráty

8. Bahrajn 37. Island 66. Mexiko 95. Spojené království VB a Severního Irska

9. Bangladéš 38. Itálie 67. Moldavská republika 96. Spojené státy americké

10. Belgie 39. Izrael 68. Mosambik 97. Srbsko

11. Bělorusko 40. Jamajka 69. Namibie 98. Srí Lanka

12. Bolívie 41. Japonsko 70. Německo 99. Súdán

13. Bosna a Hercegovina

42. Jihoafrická republika 71. Nigérie 100. Španělsko

14. Botswana 43. Jordán 72. Nikaragua 101. Švédsko

15. Brazílie 44. Kamerun 73. Nizozemí 102. Švýcarsko

16. Bulharsko 45. Kanada 74. Norsko 103. Tanzanie

17. Česká republika 46. Katar 75. Nový Zéland 104. Thajsko

18. Čína 47. Kazachstán 76. Omán 105. Trinidad a Tobago

19. Dánsko 48. Keňa 77. Pákistán 106. Tunisko

20. Dominikánská rep. 49. Kolumbie 78. Panama 107. Turecko

21. Egypt 50. Kongo 79. Papua-Nová Guinea 108. Uganda

22. Ekvádor 51. Kostarika 80. Paraguay 109. Ukrajina

23. Estonsko 52. Kuvajt 81. Peru 110. Uruguay

24. Etiopie 53. Kypr 82. Polsko 111. Vietnam

25. Filipíny 54. Kyrgyzstán 83. Portugalsko 112. Zambie

26. Finsko 55. Libanon 84. Rakousko 113. Zimbabwe

27. Francie 56. Litva 85. Rumunsko

28. Ghana 57. Lotyšsko 86. Rusko

29. Guatemala 58. Lucembursko 87. Řecko

112

Příloha č. 2: Bodový graf z ΔCPI proti ΔEGDI



113

CLASSICAL MANAGEMENT BIASES AND BEHAVIORAL APPORACH COMPREHENSION

Ivana Mandysová

Abstract: The article at the outset conducted on the theoretical basis, incorporating relevant authors and theories, has pointed to a number of weaknesses of neoclassical theory, which would be hardly compatible with real management practice and manager behavior. Article critically evaluated theoretical background in management sciences and examined manager in behavioral and heuristic theories. Analysis offered an interpretation of managerial decision-making as a product of personal preferences and characteristics, limited rationality, lack of information, mental shortcut, and imitation. Analysis described practical behavior of managers and then correlated these to each other and to explanatory factors as well. The re-orientation towards recognition of managerial behavior is developed and illustrated. Conclusions were derived from the generalization of empirical observations. Empirical research is based on in-depth qualitative study and subsequently it derives from established theoretical patterns. It explores and explains the way managers make economic decisions. The emerging theory strongly emphasizes the strengthening of the human potential inside a manager.

Keywords: Management, Behavioral Management, Manager´s Decision-making, Heuristics, Qualitative Research.

JEL Classification: D81, D91, L2, M1.

Introduction

Management theory has taken various paths of development in recent years. Increasing number of discussions on management pivots around the ideas of manager´ competencies and capabilities, personal preferences and characteristics, limited rationality, lack of information, knowledge-creating and other managerial routines, decisions, and behavioral practices. Handling disadvantages and acquiring capabilities are crucial for the ability of a firm to function. New ideas featured in this article by investigating the behavioral patterns of managers can contribute to update the classical theories.

1 Theoretical background

1.1 Classical theory of the firm, its characteristics, biases

The neoclassical theory of the firm has grown on quite orthodox opinions on the functioning of firms and the role they play on the market. The traditional understanding of firm theory is based on the assumption that it is based on the profits maximization, and that it is the only and main goal. Hence, management of the firm is pursuing profits, this single goal. Therefore, it does not allow anyone in the firm management to pursue a different goal than the firm's profit.

It can be said that the theory meets the goals for which it was created. It serves to understand the behavior of the market as a whole, to explain the impact of certain factors like changes in market prices, to which this theory is sufficient, and its

114

shortcomings are not important. This simplified theory of the firm, represented by Drucker (2008), Koontz and Weihrich (2015), makes it easy to understand the economy as whole.

The neoclassical theory of the firm assumes that the owner of the business is also a manager at the same time. Theory unites the one who bears the risks of doing business and puts his money into it with the one who manages a firm. Owner and manager as one person signs for all contracts for the firm's activities and decides about its future business orientation. In the case of a one-man firm, it is possible that the owner itself will control the firm. However, merging of ownership and management of the firm is not a way of management in practice. Nowadays, it is quite common to own a firm but not to manage it at the same time.

Another factor that is not sufficiently taken into account by the neoclassical theory is the manager´s control and oversight of the firm's activities. Today dominate the kind of companies where ownership is separated from control. These are firms or entities in which control and management are entrusted to the firm's board of directors, which is composed of owners, but also of other management personnel.

Biases of the neoclassical theory can be summarized, that it looks at a firm purposely in a very specific way in order to explain easily various changes caused by multiple factors.

1.2 Classical versus behavioral theories

According to the classical theory of the firm, companies solve problems by identifying a given problem, then setting up different solutions, from which they then select and implement the best possible solution. Neoclassical theory is based on the assumption that managers have the maximum available information and all accurate and up-to-date calculations. However, this situation in today's market can hardly ever happen.

According to the behavioral theories, managers simplify their decision-making process. They set short-term goals and look for alternatives to solutions that would satisfy these goals, rather than focusing on finding one best possible solution. Rather, they achieve targets gradually, even with the risk of potential loss as a result of choosing the wrong alternative of choice. Above all, they try to make rational decisions, use human behavioral approach to follow decision-making processes.

The neoclassical theory is based on the fact, that all entities in the company pursue only one goal, maximizing profit, as an unchanging fact. The theory considers the assumption that there are many entities inside the firm with different, often conflicting interests. The neoclassical theory acknowledges the fact that managers may have their own private interests but at the same time, theory assumes that they will either voluntarily ignore them or that a possible conflict of interest between the firm and manager will be covered by employee contract.

Behavioral theories take the opposite view. They assume that there is a constant conflict of interest between the firm and other entities. The permanent conflicts of interests among owners and managers are subject to speculation and action in the firm. This conflict will never disappear and must be solved over and over again. It is therefore a major problem to identify an unambiguous firm goal as each side has a

115

different interest. According to behavioral theories, the coherence of the interests inside the firm is rarely achievable and difficult to sustain.

Behavioral theory takes into account the human factor that has according to Šebestová (2007) a major influence. It examines not only the behavior of the firm, but also the behavior of particular individuals like hired managers.

1.3 Behavioral theories

Behavioral theories are alternative theories of a firm that explain the behavior of individual firms. They look for the factors that affect the business, its management, the firms´ decision-making, and the goals it pursues. They were founded in the 1960s on the assumption that the maximization of business profits is not the only target function of the business. In 1963, Cyert and March, one of the earliest experts in economic research work on behavioral theories, defined the main ideas underlying behavioral theories, and the authors themselves led to their practical validation.

Theory of limited rationality comes from observation that rational factors in decision-making are significantly limited by the amount of information available. Businesses are run by managers, who have complex relationships with business owners, which may be individuals, a group of people, or other businesses. This creates a complex ownership and managerial structure, bringing together organizational and communication problems that complicate the transfer of information. This causes lack of information, as it is not always possible to decide and manage accurately and timely. To make calculations is often costing and time consuming.

Cyert and March (1963) set out the main goals of behavioral theory. They clearly defined what the behavioral theory serves for. Authors have validated theoretical background by comparing theoretical assumptions with observations in different companies.

Behavioral theories display and explain how the firms and their management decides. It seeks to specify a set of rules and motivation assumptions that could be used to describe the manager's individual decisions.

Based on this, behavioral theories try to identify the possibilities for individual manager to make decisions. Based on the firm’s goals, the theory specifies a set of management decision-making rules related to production, price, customers, costs, suppliers, and location. Management decision-making takes account of firm process optimization, but at the same time puts emphasis on set goals.


Neoclassical theory is insufficient and does not give a true picture of the reality surrounding the firms, it does not explain nature and principles of their management operation. The neoclassical theory does not take into account the factors like an ownership form, performance of control, decision making and managing under uncertainty conditions. The theory was not set in order to investigate and explain the behavior of individual firms or managers and to deal with their internal processes.

Because the neoclassical theory of the firm does not correspond with the reality, more and more theorists gradually began to present various alternative theories dealing

116

with the behavior and nature of the existence and management of the firm. One of these theories is the theory of behavioral management.

Managerial theories are based on the principle that a firm managed by hired managers can have a different goal than to maximize their profits. When the firm owner is not present in the firm and the person who manages and controls the firm has all powers, he can pursue alternative interests. These interests are different then the business owners have. Thus, the two sides in the firm are often in mutual conflict. They follow the interests important to them from their point of view. Neoclassical theories are in conflict with reality and therefore it is necessary to amend theories by empirical research and evaluate theory consequently.

The main aim of the study is to verify the adequacy of behavioral management theory by examining it in companies and compare the results achieved with other management approaches.

3 Methods

Behaviorists proceed inductively, they do not examine how managers should behave, but rather how they really do. Conclusions are derived from the generalization of empirical observations. Behavioral method describes the behavior of managers and then correlates these to explanatory factors. Hence, this is the method of presented study when outcome hypotheses are created in the end of research.

Manager´s behavioral variables are in large part about subjective perceptions and not precise measurements. Accordingly, behavioral variables are used. Managerial action variables are defined as follows: transaction costs, competitiveness, limited rationality, imitation, size, suppliers’ availability, human resources, finance availability and location.

3.1 Measurements

Behavioral variables are based on the qualitative and quantitative research of entrepreneurs, micro and small firms, their behavior and their management. For some variables, it is difficult to express them numerically, so in these cases a scale of score we determined according to analysis of the qualitative sources, which is requirement for used statistical methods. (Yasai-Ardekani, 1986)

To that end study chooses a qualitative research method, which involves the collection and analysis of qualitative data.

Qualitative research proceeds inductively and statements are generated only during the research as part of trying to understand a new problem. The reason for choosing the qualitative method here is that this issue is quite complex. The main benefit of the method is to provide in-depth insight into 54 enterprises.

3.2 Data collection

The investigated 54 firms come from the Pardubice region. The numbers of SMEs in the Czech Republic and in the Pardubice region (Zpráva o vývoji malého a středního podnikání a jeho podpoře, 2018) were identified. The research was carried out over the past five years, mapping 20 years of history of the individual business. Interviewed managers are a versatile group of people, all with valuable and

117

comprehensive insights of their firm´ genesis, development and daily problems. Direct personal confrontation with the examined object and repeated long term research is an argument for the reliability.

Managers and entrepreneurs were proud of the results they have reached and they were willing to share their experience in empirical research. They have described how they started the business and how it developed, mentioning environmental obstacles they had to face as well as inner processes and inner doubts they had to go through. Six of them were women, two of which were wholesalers and four hair salon owners. The sample fits with NACE structure, hence major proportion is metal production. Although at the beginning managers have lacked technical skills, know-how, business experience, and capital, they proved exceptional flexibility and hard work.

Respondents were recorded and then the text was literally transcribed. The framework of the interview was established, and questions were directed to crystallize the situations that evoked a need for a managerial decision and subsequent choice of action. Their dynamics and causal mechanisms were emphasized.

The interviews searched for links and moments that particularly affected the managerial decision. Case study text was analyzed and integrated properly by prevailing assumptions and their meaning taking into account theoretical background. Data were combined in order to identify links and lastly use the processed material to amend existing theory. Processed data were connected to relevant theories finding a match or mismatch to be in accordance with main one.

3.3 Statistical methods

This measure was developed by Yasai-Ardekani (1989) and Kutner et al., (2005), used mathematical measuring analysis. In order to test effects of the management style and manager´s behavior, a correlation is used.

A Pearson´s correlation coefficient is used. A number displayed in a Tab. n. 1 indicates the extent to which two variables are linearly related. (Bryman, Lewis-Beck, 2003).

Variables:

As the representational variables for manager´s behavior and his individual perception were chosen indicators following method of Soofi et al. (2000). These variables were analyzed, in order to explain and confirm basic correlations. Variables are dichotomous (except Size variable) and represent the factors as managers themselves perceive them. Qualitative managers´ statements related to this topic were analyzed and for each firm there was set value 0 for NO answer or a value 1 for YES answer.

Variable Profit maximization represents management style where value 0 is aiming towards maximum profit and value 1 is aiming towards satisfactory profit level.

Variable Location represents the factor of the location choice, as the managers themselves perceive it. There was set value 0 for disadvantageous location or value 1 for advantageous location.

118

Variable Competitiveness expresses the ability of the manager to successfully identify, grasp and use the right opportunity for growth, to expand and improve and strengthen the market position of the firm towards competitors (Starzyczna, 2015). Value 0 is for weak competitiveness and value 1 is for strongly perceived competitiveness.

Variable Imitation comes when behavior of peer firms are mindlessly imitated by managers of other firms no matter how wrong they may be. For instance if the recent evidence suggest that competitors have made money by choosing certain possibility, others will do exactly the same, managers simply imitate. Value 0 is for no imitation practice, value 1 is for active imitation practice.

Variable Size was measured as the logarithm of the number of employees.

Variable Transaction costs is a cost in making an economic transaction, especially cost to get information. Many theorists like De Nisi (2011) define transaction costs as cost of time spent on contracting, etc. Value 0 is for perceived low cost and value 1 is for perceived high cost.

Variable Suppliers availability is also a dummy variable and shows limited availability of convenient suppliers in the region. For 0 there is a good availability, entrepreneur does not need to create supra-regional ties of suppliers; for 1 the businesses experience the lack of suitable suppliers and they have to look for them in the more distant environments or even abroad.

Variable Human resources displays limited availability of qualified employees in the region (Augier, Teece, 2006), (Dvouletý, 2017). For every firm there is determined value 0 or 1. Value 1 means that firm is not able to find enough qualified employees in the surrounding region and is forced to search for human resources at the supra-regional level while on the other hand value 0 confirmed that entrepreneur could hire all the employees only within the region.

Variable Finance availability represents rate of limited availability of the finance in the region. It is measured from the point of view of the entrepreneurs, who indicate the circumstances and the rate of availability of the various finance resources in the region. For better interpretation of the results of the analysis, this variable is as well expressed with dummy variable 0 and 1. Value 0 means that availability of finance is not limited and the level of available finance resources in region is satisfying while value 1 confirmed statement that the availability of finance resources if limited for the specific entrepreneur.

Variable decision under Limited rationality expresses the limited ability of manager to choose the optimal possibility out of several possibilities and decision making under uncertainty. Task optimization is taken into consideration. Value 0 is for perceived weak task optimization and value 1 is for strong task optimization.

119

4 Outcomes and results

Relationships between all the variables are displayed in the Tab. 1 through correlation coefficients, which express positive or negative deviation via plus or minus signs of the coefficient. There are also means and standard deviations. (Kutner, et al., 2005).

To replicate the procedure of Yasai-Ardekani (1989), his proposed research model was tested via correlation analysis. Correlations were calculated together with means and standard deviations are portrayed in Tab. 1. Higher means shows Size, lower means show Finance availability, Human resources, Transaction costs and Limited rationality.

As can be seen in Tab. 1, variables tested on bivariate correlations under 0,39 are bellow critical values. Higher correlations show variables Profit – Competitiveness, Size – Competitiveness, Suppliers availability – Imitation practices, Human resource management – Competitiveness, Limited rationality – Transaction costs.

Tab. 1: Means, Standard Deviations, and correlations for all variables

Variables Means Standard

Deviations 1 2 3 4 5 6 7 8 9 10 1. Profit maximization 1.80 0.89 1

2. Location 1.48 0.64 .10 1

3. Competitiveness 1.84 0.83 .39 .29 1

4. Imitation 0.84 0.37 -.22 .16 -.08 1

5. Size 2.60 1.83 .01 .00 .40 .01 1

6. Transaction costs 0.56 0.50 -.02 .16 .31 .05 .12 1

7. Suppliers availability 0.88 0.32 .06 .08 .08 .51 -.10 -.08 1

8. Human resources 0.60 0.49 .37 .23 .43 .09 .13 .10 .20 1

9. Finance availability 0.40 0.49 -.18 .15 .06 .13 -.20 -.10 .30 .17 1

10. Limited rationality 0.52 0.50 .32 -.03 -.09 -.20 .28 .48 -.11 .20 -.20 1

All correlations above 0,30 are significant at p < 0,05 and all correlations above 0,39 are significant at p < 0,01. The number of observation is 54.

Source: own

Significance allowed to state outcome hypotheses:

H1 The more manager seeks to maximize profit, the more competitive is the firm.

H2 Bigger firm reaches better competitiveness.

H3 Suppliers´ availability is high for firm with imitation practices.

H4 The firm that is able to manage relevant human resources is more competitive.

H5 With limited rationality of manager the transaction cost are growing.

5 Evaluation and discussion

The theory accuracy was verified examining the observed firms, deciding on management strategies, and subsequently compared with the original theories. Behavioral theory served as a tool for analyzing decision-making among several alternatives of business strategy. A large number of recommendations in the field of long-term management strategy are designed to influence decision-making principles

120

in various ways. On this basis, the theory is also used to deduce the likely consequences of alternative management strategies.

Evaluation comes from using the basic statements of the behavioral theory for describing the certain style of a management of individual manager (Cyert, March, 1963) and deducing behavior of a manager. Supposed behavior compared with actual observed manager´s behavior allowed to create outcome hypotheses based on correlations (Mandysová, 2017).

Empirical research has proved that decision-making under conditions of uncertainty is a very common situation for every manager today. However, the neoclassical theory of the firm does not consider this, it assumes the perfect awareness of all market players. Every manager of every firm on the market searches for the most accurate information about their internal and external environment as a basis for their management decision-making. Internal information, such as business performance data, can be obtained relatively easily, while the availability of external information is not always easy. Research proved that it is largely dependent on the manager's ability to decide under limited rationality, to apply different methods and tools, such as imitation, postponing suppliers´ payments, suppliers´ quality diversification, market research, competition analysis, and customer analysis future trends estimates, etc.

According to research outputs managerial decision-making is enriched by new goal setting. Firm size influences its competitiveness. In a big firm, the profit maximization is the main goal. In smaller firm, for example, maximizing profit, production, rewards, etc., is replaced by an attempt to achieve a satisfactorily level with all objectives (Vodák and Strišš, 2005). Thus, there is some explanation of satisfactory levels for the managerial objectives, reducing and influencing managerial decisions made under limited rationality and causing transaction costs growth. Outcomes showed that company size and its growth enables to maximize profit, to deal more effectively with competitive pressures (Myšková and Doupalová, 2015) and to reach higher competitiveness. In supplier´s policy, managers´ behavior depends on personal preferences and imitation practices. This will ensure greater flexibility in dealing with perceived pressures.

At present, managers are forced to tackle increasing amount of new and complex challenges and tasks associated with market globalization, IT technologies, innovation, environmental requirements, advanced and high-level competition, and so on. Managers have to solve several problems at the same time and must therefore look for a comprehensive and original solution. The ordinary management procedures used before are inefficient, outdated, and are not suited to addressing the situation that the business is currently facing. One of the possible ways to tackle these new challenges is the heuristic approach. Managers decide following their personal preferences, decision-making according to Kahneman (2012) is a product of personal preferences and characteristics, limited rationality, risk perception. In accordance with Killingsworth et. al. (2016), managers use heuristics, they often meet satisfactory level and due to lack of information imitate and decide by mental shortcut.

121

Conclusion

The method of analysis undertaken in this research allowed examining the management style as impact of individual characteristics and preferences. The analysis first conducted at the theoretical basis including relevant authors and theories, has pointed to a number of weaknesses of neoclassical theory, which would be hardly compatible with real practice and management´ behavior.

The article linked manager´s individual perceptions and actions to his specific response. It allowed greater precision. Attention in the research of manager´s behaviour and decision-making was focused on actions and practices under limited rationality and insufficient information. It was proved that managers react sometimes in an inadequate way. Surprisingly, many choices of the decision can be used to mitigate the situation (Koráb, 2016). Managers´ behavior as the frequency of changing ties is directly affected by the limited determinants such as availability of suppliers, human resources and finance sources.

The research integrated social and human factors, including values such as confidence, prestige, a feeling of home, insecurity, fear. Identifying the actual behavior and daily work, it creates a description of a manager’s decision in terms of a specific series of semi-decisions used to reach a choice. It discovers inner mechanisms as well as their genesis and incorporates them into the classical behavioral theory of the firm.

References

Augier, M., Teece, D. J. (2006). Understanding complex organization: the role of know-how, internal structure, and human behavior in the evolution of capabilities, Industrial and Corporate Change. Oxford: University Press, No. 15, no. 2, pp. 395-416, ISSN: 09606491

Bryman, A., Lewis-Beck, M., (2003). The SAGE Encyclopedia of Social Science Research Methods. Sage Publishing: London.

Cyert, R., March, J. G. (1963). A behavioral Theory of the Firm. Oxford: Blackwell.

De Nisi, A. (2011). Managing Performance to Change Behavior. Journal of Organizational Behavior Management, No. 31, Issue 4

Drucker, P., (2008). The essential Drucker: the best of sixty years of Peter Drucker’s essential writings on management. New York: Harper Collins.

Dvouletý, O. (2017). Does the self-employment policy reduce unemployment and increase employment? Empirical evidence from the Czech regions. Central European Journal of Public Policy, No. (2), 11-22.

Kahneman, D. (2012). Thinking, Fast and Slow. New York: Penguin Press.

Killingsworth, K. et al. (2016). A Behavioral Interpretation of Situation Awareness: Prospects for Organizational Behavior Management, Journal of Organizational Behavior Management, No. 36(4), pp.301-321

Koontz, H., Weihrich, H., (2015). Essentials of management: an international, innovation, and leadership perspective. Tenth edition. Chennai: McGraw-Hill Education (India) Private Limited. ISBN 978-93-392-2286-4.

Koráb, V., (2016). What determines innovation in small and medium enteprises? A case study from the Czech Republic, Transformations in Business and Economics. 15(2): s. 78-88. Brno; VUT v Brně

Kutner, M. H. et al. (2005). Applied linear statistical models (5th ed.). Boston: McGraw-Hill.

122

Mandysová, I., (2017). A behavioral theory of the firm: Specifics of Czech entrepreneurial behavior. E+M, Ekonomie a Management. No. 21. pp. 85-100.

Myšková, R., Doupalová, V., (2015). Approach to Risk Management Decision-Making in the Small Business. Procedia Economics and Finance, No. 34, C, pp. 329-336 https://doi.org/10.1016/S2212-5671(15)01637-8

Soofi, E. S., Retzer, J. J., et al. (2000). Framework for measuring the importance of variables with applications to management research and decision models. Decision Sciences, No. 31 (3), pp. 595-625.

Starzyczná, H. et al. (2015). Fluence of Product Leaflets on Purchasing Behaviour in Times of Crisis. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, No. 63(4), pp.1357-1363

Šebestová, J., (2007). Analysis of endogenous factors influencing small and medium sized enterprises: The case of the Moravian-Silesian region. Ekonomický časopis, No. 55, no. 4, pp. 411-424. ISSN: 00133035

Vodák, J., Strišš, J. (2005). Corporate identity as one of the important management activity. Scientific papers of the University of Pardubice, No. 3 Series B, 11

Walker, R., (1989). A requiem for corporate geography: new directions in industrial organization, the production of place and uneven development. Geografiska Annaler. No. 71 B, 1, pp. 43-68.

Yasai-Ardekani, M., (1989). Effects Of Environmental Scarcity And Munificence On The Relationship of the context to organizational structure, Academy of Management Journal. No. 32, 1 ProQuest pp. 131-156

Zpráva o vývoji malého a středního podnikání a jeho podpoře v roce 2018. [online]. Available from: https://www.komora.cz/legislation/78-18-zprava-o-vyvoji-maleho-a-stredniho-podnikani-a-jeho-podpore-v-roce-2018/

Contact Address

Ivana Mandysová, Ph.D. University of Parduibce, Faculty of Economics and Administration, Institute of Administrative and Social Sciences Studentská 84, 532 10 Pardubice, Czech Republic Email: [email protected] Phone number: +420466036528


123

THE PROBLEM OF DIFFERENT PAIRWISE COMPARISONS SCALES IN THE AHP FRAMEWORK

Jiří Mazurek

Abstract: In pairwise comparisons, several scales are used to compare objects. Perhaps the most known is Saaty’s fundamental scale for the AHP/ANP from 1 to 9 (with reciprocals), but other scales with 3, 5 or 10 items are also used in practice. Since the AHP/ANP is scale invariant, the following problem arises: for example, a preference of one object over other object expressed by the value 2 means something else for the scale from 1 to 3 (it expresses the medium preference) and the scale from 1 to 100 (in this case the preference is almost negligible). Therefore, the need of a normalization for pairwise comparison scales arises. The aim of the article is to propose a suitable transformation of a general linear scale for pairwise comparisons to a unit real interval that preserves several natural and desirable properties.

Keywords: AHP, Normalization, Pairwise Comparisons, Comparison Scale.

JEL classification: D81, D83.

Introduction

Pairwise comparisons (PCs) belong among the most common tools for multiple criteria decision making since the introduction of the analytic hierarchy process (AHP) and the analytic network process (ANP) by T. L. Saaty in 1977 and 1980 respectively, see Saaty (1977 and 1980). The list of successful AHP/ANP applications is rapidly growing, see e. g. Vaidya and Kumar (2006), Subramanian and Ramanathan (2012), Kramulová and Jablonský (2016), or Lidinská and Jablonský (2018). It should be noted that PCs method has a long history predating the papers by T. L. Saaty, with the first mention of the PCs method dating back to the work of the Catalan scholar and monk Ramon Lull in the 13th century, while the first modern work on PCs can be attributed to L. L. Thurstone and his Law of Comparative Judgments, see Thurstone (1927).

In the AHP/ANP, the so called fundamental (linear) scale from 1 to 9 (with reciprocals) is used for pairwise comparisons. However, other linear scales with 3, 5, 7, 8 or 10 items were also proposed, see Koczkodaj (1993) or Koczkodaj et al. (2016). In particular, Fülop et al. (2010) and Koczkodaj et al. (2016) provide strong arguments to use the scale only up to three. Variety of other studies suggested the use of non-linear scales such as logarithmic scales, exponential scales, or scales based on a logistic function, see for example Lootsma (1993), Donegan et al. (1992), Ma and Zheng (1991) or Salo and Hämäläinen (1997). Ishizaka and Labib (2011) provide a comprehensive review of scales of pairwise comparisons up to year 2010.

Studies on scale comparisons are rather seldom, see Dong et al. (2008), Elliot (2010), Triantaphyllou et al. (1994), or Starczewski (2017). Dong et al. (2008) provided comparisons on several scales for pairwise comparisons, and concluded that with respect to their algorithms the best scale was the geometrical one. Elliot (2010) experimentally compared three different scales with the result that none of the scales captured accurately the preferences of all individuals. Triantaphyllou et al. (1994)

124

compared 78 scales to conclude that no single scale could outperform all the other scales. Starczewski (2017) examined the effect of a scale (he compared the fundamental scale, extension scale and geometric scale) on a priority vector, and found that scales with more options lead to a better (more precise) evaluation of a priority vector. Franek and Kresta (2014) compared Saaty’s scale to other scales for both consistent and inconsistent pairwise comparison matrices. According to the authors, Saaty’s scale is still favorable, but if a decision maker demands higher consistency, he/she should use root square or logarithmic scales.

The AHP/ANP is scale invariant, which means that the result (a priority vector) does not depend on a scale used. However, when using different scales for PCs, the same value from a given scale has a different meaning. This problem was explicitly expressed by Koczkodaj (2015) and dubbed the “pairwise comparisons rating scale paradox”. Koczkodaj (2015) also offered a solution to this paradox, a normalization of a rating scale via linear transform. However, the proposed linear transformation in Koczkodaj (2015) has a main drawback as it does not preserve consistency of PCs.

Therefore, the aim of this paper is to introduce a new (power) transform of a (linear) rating scale to a unit interval that has several desirable properties, namely it preserves consistency of PCs, the ranking of objects and the most inconsistent triad.

The paper is organized as follows: section 1 provides brief introduction to pairwise comparisons and the problem with different scales, in section 2 the normalization is proposed along with several of its properties, section 3 provides a numerical example and several aspects of the paper are discussed in section 4. Conclusions close the article.


1.1 Preliminaries

Let 1,..., nC c c , 2, nNn , be a set of compared objects (concepts, entities). Let

1 / ,ija m m denotes the relative importance of an object ic over object jc . Then the

(square) matrix nnijaA )( , , 1, ...,i j n is called a pairwise comparison matrix (PCM).

The matrix A is reciprocal if and only if:

1 /ij jia a , , 1,...,i j n (1)

The matrix A is consistent (or, alternatively, pairwise comparisons are said to be consistent) if and only if:

ij jl ila a a , , 1,...,i j l n (2)

The final weights of objects (priority vector) w in the AHP/ANP, which were pairwise compared, is derived by the eigenvalue method, and satisfies the following condition:

wAw max

where max is the maximum (positive) eigenvalue of the matrix A.

As pairwise comparison matrices are often inconsistent, various inconsistency indices were proposed. Perhaps the most known are Saaty’s consistency index C.I. and

125

consistency ratio C.R., see Saaty (1980, 2008). For other inconsistency indices proposed in the literature see e.g. Alonso and Lamata (2006), Brunelli and Fedrizzi (2015) or Brunelli (2017).

In this paper Koczkodaj’s inconsistency index (KII) is used to measure PCM inconsistency, which is defined as follows:

Definition 1. Koczkodaj’s inconsistency index (KII), Koczkodaj (1993, 2014): Let ( )n be the set of all ordered triples (“triads”) , ,ij jk ika a a satisfying (2) for

, , 1, 2,...,i j k n . Then:

( )max min 1 , 1 ij jkik

nij jk ik

a aaKII

a a a

(3)

The KII expresses inconsistency of a pairwise comparison matrix in terms of the most inconsistent triad, and 1,0KII .

Let m,1 with reciprocals (which will be further omitted in the text) denote alternative scales for pairwise comparisons, where 2, mRm . In the context of this study whether the scale is discrete or continuous is not important. For Saaty’s scale m = 9.

1.2 The problem of different scales

Originally, T. L. Saaty proposed to use the linear scale from 1 to 9 for the pairwise comparisons, see Saaty (1977). The strength of preference or importance for the Saaty’s scale is shown in Tab. 1.

However, when at least two different scales are used for pairwise comparisons, the following problem emerges.

Let’s consider two pairwise comparisons scales, 3 1, 2,3S , and 100 1, 2,...,100S .

If, for instance 2ija , then its meaning for the scale 3 1, 2,3S and the scale

100 1, 2,...,100S is different. In the former case, the preference 2ija means medium

preference of the object i to the object j, while in the latter case the preference is almost negligible, see Fig. 1. Therefore, the scale for pairwise comparisons cannot be neglected. This is true especially in situations when different scales are used simultaneously, or, when results of pairwise comparisons of the same set of objects with different scales need to be compared. In such cases, scale normalization is necessary.

Fig. 1. A comparison of two different scales

Source: author

126

Tab. 1. Saaty’s fundamental scale. Intensity of importance Definition

1 Equal importance 2 Weak or slight 3 Moderate importance 4 Moderate plus 5 Strong importance 6 Strong plus 7 Very strong importance 8 Very, very strong importance 9 Extreme importance

Source: Saaty (1977, 1980, 2008)

2 Methods

2.1 Normalization of a pairwise comparison scale

To avoid the problem described in the previous section, normalization of the comparison scale is applied.

Let the general pairwise comparison scale 1,GS m , 2, mRm . Then, by the

normalization, this scale is transformed into the 1,2 unit interval of real numbers.

The transformation f (the normalization) should satisfy the following (obvious) conditions:

i) : 1, 1, 2f m ,

ii) f is strictly increasing. iii) (1) 1f and (m) 2f .

In Koczkodaj (2015), a linear transform f is proposed for normalization:

1 2( )

1 1

mf x x

m m

(4)

However, the linear transform (4) has a serious drawback: it does not preserve consistency of pairwise comparisons.

Consider for example the consistent triad (2, 3, 6), and let m = 9. Then, after the

linear transform, the triad (9/8, 10/8, 13/8) is not consistent, because 9 10 13

8 8 8 .

2.2 Properties of the proposed power normalization

Definition 2. Let A(a )ij be a pairwise comparison matrix, 1 / ,ija m m , 2, mRm .

Let f be the power transform:

( ) k

ij ijf a a , ln 2

lnk

m (5)

Proposition 1: The power transform (5) satisfies conditions i)-iii).

Proof is obvious.

127

Proposition 2: The power transform (5) preserves consistency: if ij jl ila a a , then also

( ) ( ) ( )ij jl ilf a f a f a .

Proof: Let ij jl ila a a . Then ( ) ( )kk k k

ij jl ij jl ij jl il ilf a f a a a a a a f a .

Proposition 3: Let A be a PCM of the order n. Let 0 0 0 0( , , )T x y z be a triad (from A) with maximum KII. Let A* be the matrix A transformed by (5). Then the KII of the transformed triad 0 0 0 0( , , )T x y z (from A*) is also maximal.

Proof: For a triad 0 0 0 0( , , )T x y z one of two possible cases holds: either 0 0

0

1x y

z

, or

0 0

0

1x y

z

. Without loss of generality suppose that 0 0

0

1x y

z

. Also, because KII is

maximal for T0, 0 0

0

x y

z

x y

z

holds for all triads ( , , )x y z . After transformation (5),

where k > 0, we get: 0 0 0 0

0 0

kk k

k

x y x y

z z

kk k

k

x y x y

z z

, which is true for any

positive k.

Proposition 3 allows to show that if a pairwise comparison matrix A is more inconsistent than a pairwise comparison matrix B (with respect to KII), this relation is preserved by the power transform (5).

Proposition 4. Let ( )ijA a and B(b )ij be inconsistent pairwise comparison matrices of

the order n. Let KII be the Koczkodaj’s inconsistency index (3) and let ( ) ( )KII A KII B . Let ( )ijA a and B (b )ij

be transformed pairwise comparison

matrices by the transform (5), whereln 2

lnp

m . Then ( ) ( )KII A KII B .

Proof: Let the most inconsistent triad of the matrix A be , ,ik ij jka a a and the most

inconsistent triad of B , b , bik ij jkb . Then, either 1ik

ij jk

a

a a

and 1ik

ij jk

b

b b

, or 1ik

ij jk

a

a a

and 1ik

ij jk

b

b b

. Without loss of generality assume the latter.

Since ( ) ( )KII A KII B , we have 1 1ik ik

ij jk ij jk

a b

a a b b

, hence ik ik

ij jk ij jk

a b

a a b b

.

From Proposition 3 it follows that the most inconsistent triad is preserved by the transform (5). Hence, for the transformed matrices A* and B* the most inconsistent triads are , ,ik ij jka a a and , b , bik ij jkb respectively. As above, without loss of

generality we assume 1p

ik ikp p

ij jk ij jk

a a

a a a a

and 1p

ik ikp p

ij jk ij jk

b b

b b b b

.

128

Then we have: ik ik

ij jk ij jk

a b

a a b b

p p

ik ik

ij jk ij jk

a b

a a b b

p p

ik ikp p p p

ij jk ij jk

a b

a a b b

ik ik

ij jk ij jk

a b

a a b b

1 1ik ik

ij jk ij jk

a b

a a b b

( ) ( )KII A KII B .

Proposition 5: Let A be a PCM of the order n. The transformation (5) does not change ranking of all alternatives if the weights of all alternatives are determined by the geometric mean method.

Proof: Let wi, 1, 2,...,i n , be the weights of alternatives derived form a pairwise

comparison matrix ( )ijA a by the geometric mean method:

n

i

nn

jij

nn

jij

i

a

a

w

1

/1

1

/1

1 (6)

Without loss of generality it suffices to show that the relation “to be less or equal than” is preserved for an arbitrary pair of weights. Hence, let 1 2w w , and let *

1w and *2w be the transformed weights respectively, then 1 2w w should hold. When

comparing weights given by relation (6), the denominator is the same, so it can be

omitted. Also, n-th square can be omitted. Hence, 1 2w w means that 1 21 1

n n

j jj j

a a

.

After transform (5) we get:

1 1 2 21 1 1 1

k kn n n n

k k

j j j jj j j j

a a a a

, hence 1 2w w . Because the proof for any

other pair of weights is analogous, the Proposition 5 is proved.

Remark 1. It is well-known that priority vectors derived from a pairwise comparison matrix of the order n = 3 by the eigenvalue method and the geometric mean method are identical. Therefore, Proposition 5 is also valid for the eigenvalue method and a pairwise comparison matrix of the order n = 3. However, whether Proposition 5 is valid for the eigenvalue method and n > 3 remains an open question.

3 Problem solving

3.1 Numerical example

In this section the use of the proposed normalization (5) is demonstrated on an example.

Example 1. Let the pairwise comparison matrix of four objects be given as follows:

129

Tab. 2. The input PCM. 1 2 3 5

0.5 1 2 4

0.33333 0.5 1 3

0.2 0.25 0.3333 1 Source: own.

By the GM method, the vector of weights (the priority vector) is: 07.0,17.0,29.0,47.0w .

The first object has the highest weight which means it is the most preferred or important entity, the second object follows on the second place, etc. Further, there are four triads: (T1) 132312 aaa , (T2) 142412 aaa , (T3) 143413 aaa and T(4)

243423 aaa . It can be easily checked that all four triads are inconsistent, and the most inconsistent triad is (T3).

Now suppose that m is 5, 7, 9 and 20 respectively, or in other words, consider four scales: [1,5], [1,7], [1,9], and [1,20] (with reciprocals).

Then, after the transformation (5) of the PCM given in Tab. 2, normalized pairwise comparison matrices are shown in Tab. 3 along with inconsistency of all four triads and weights of alternatives (priority vectors) on the right hand side of the table. The maximal values are highlighted in blue.

As can be seen, both the maximum inconsistency and objects’ rankings are ‘invariant’ for all scales, the triad (T3) is still the most inconsistent one, and the ordering of objects remains unchanged.

Tab. 3. Four PCMs after transformation (5). scale PCM KII weights

m = 5 1 1.347866 1.6050366 2 0.11653 0.345465

0.741914 1 1.34786552 1.816741 0.183248 0.278091

0.623039 0.741914 1 1.605037 0.314114 0.222306

0.5 0.550436 0.62303875 1 0.160228 0.154139

m = 7 1 1.280056 1.4789518 1.7741 0.097399 0.328601

0.781216 1 1.28005623 1.638544 0.154155 0.274627

0.676155 0.781216 1 1.478952 0.267907 0.228203

0.563666 0.610298 0.67615456 1 0.134484 0.16857

m = 9 1 1.244413 1.41421356 1.661501 0.086757 0.319406

0.803592 1 1.24441257 1.548563 0.137801 0.272477

0.707107 0.803592 1 1.414214 0.241323 0.231264

0.601866 0.64576 0.70710678 1 0.120067 0.176853

m = 20 1 1.173956 1.28942315 1.451197 0.064396 0.300545

0.851821 1 1.1739559 1.378172 0.103044 0.26748

0.775541 0.851821 1 1.289423 0.183366 0.237166

0.689086 0.725599 0.7755406 1 0.08955 0.194809 Source: own

130

Also, it’s worth noting that with the growing upper boundary m, the weights of all objects are becoming more and more uniform (closer to each other).

4 Discussion

In the previous two sections the normalization (the power transform) of the scale for pairwise comparisons was introduced, and several natural and desirable properties of this transformations with respect to the geometrical mean (GM) method and Koczkodaj’s inconsistency index (KII) were shown and proved. The simplicity of the proposed normalization and its nice properties might provide incentive for its practical use.

Nevertheless, the use of the GM method and KII could be considered limitations of this study. The eigenvalue method for the derivation of a priority vector can be used instead of GM method, and there are many other inconsistency indices than KII proposed in the literature, such as Pelaez-Lamata PLI index, Golden-Wang GWI index, Aguaron and Moreno-Jimenez GCI index, and so on, see e.g. Brunelli and Fedrizzi (2015), which might be examined with regard to the proposed normalization. If pursued, this direction of research might prove to be interesting as well, though there is no certainty that similar results can be obtained for other inconsistency indices than KII, since KII is a maximum-based index of inconsistency unlike other, rather mean-based inconsistency indices. Yet, this research direction certainly deserves attention of experts in the field.

Conclusions

The aim of the paper was to propose a normalization of the scale for pairwise comparisons in the multiplicative AHP/ANP framework, as the AHP/ANP is scale invariant. This leads to undesired effects regarding the intensity of preference, which is, actually, dependent on the upper bound of an applied scale.

The proposed solution to the problem is a normalization in the form of a simple power transforms. The transform has several virtues, see Propositions 1-5 in section 2, namely it preserves consistency of pairwise comparisons, the most inconsistent triad, relation of inconsistency between two arbitrary inconsistent matrices, and last, but not least, it also preserves objects’ rankings when the inconsistency is expressed in terms of Koczkodaj’s inconsistency index.

The proposed approach, the scale normalization, is recommended as an additional step of AHP/ANP when different scales for pairwise comparisons are employed simultaneously, or, when results of pairwise comparisons obtained with different scales are to be compared.

Further research might focus on a scale problem in the context of additive AHP/ANP or fuzzy AHP/ANP. Also, the research of the effect of the proposed normalization on inconsistent pairwise comparison matrices with respect to other inconsistency indices would be desirable.

Acknowledgement

This paper was supported by the Grant Agency of the Czech Republic no. 18-01246S.

131

References

Alonso, J. A., Lamata M. T. (2006). Consistency in the analytic hierarchy process: a new approach, International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 14(4), pp. 445-459.

Brunelli M., Fedrizzi M. (2015). Axiomatic properties of inconsistency indices for pairwise comparisons, Journal of the Operational Research Society, 66(1), pp. 1-15. doi: 10.1057/jors.2013.135

Brunelli M. (2017). Studying a set of properties of inconsistency indices for pairwise comparisons. Annals of Operations Research, 248(1-2), pp. 143-161.

Donegan, H. A., Dodd, F. J., McMaster, T.B M. (1992). A new approach to AHP decision-making. The Statistician, pp. 295-302.

Dong, Y., Xu, Y., Li, H., Dai, M. (2008). A comparative study of the numerical scales and the prioritization methods in AHP. European Journal of Operational Research 186: pp. 229–242. doi: 10.1016/j.ejor.2007.01.044.

Elliott, M. A. (2010). Selecting numerical scales for pairwise comparisons, Reliability Engineering & System Safety, 95(7), pp. 750-763

Fülop J., Koczkodaj, W. W., Szarek, S. J. (2010). A Different Perspective on a Scale for Pairwise Comparisons. N.T. Nguyen and R. Kowalczyk (Eds.): Transactions on CCI I, LNCS 6220, pp. 71–84.

Franek, J., Kresta, K. (2014). Judgment Scales and Consistency Measure in AHP, Procedia Economics and Finance, 12, pp. 164-173.

Ishizaka, A., Labib, A. (2011). Review of the main developments in the analytic hierarchy process. Expert Syst. Appl., 38, pp. 14336-14345.

Koczkodaj, W. W. (2015). Pairwise comparison scale rating paradox, arXiv:1511.07540v2. Retreived: Dec 2015.

Koczkodaj, W. W. (1993). A new definition of consistency of pairwise comparisons, Mathematical and Computer Modeling, 18(7), pp. 79–84.

Koczkodaj, W., W., Mikhailov, L., Redlarski, G., Soltys, M., Szybowski, J., Wajch, E., Kam Fung Yuen, K., Tamazian, G. (2016). Important Facts and Observations about Pairwise Comparisons. Fundamenta Informaticae, 144(3-4), pp. 291-307.

Kramulová, J., Jablonský, J. (2016). AHP model for competitiveness analysis of selected countries. Central European Journal of Operations Research, 24(2), pp. 335-351. doi: 10.1007/s10100-015-0394-7.

Lidinska, L., Jablonský, J. (2018). AHP model for performance evaluation of employees in a Czech management consulting company. Central European Journal of Operations Research, 26(1), pp. 239-258. doi: 10.1007/s10100-017-0486-7

Lootsma, F. A. (1993) Scale sensitivity in the multiplicative AHP and SMART. Journal of Multi-Criteria Decision Analysis, 2(2), pp. 87-110.

Ma, D., Zheng, X. (1991). 9/9-9/1 scale method of AHP. In: Proceedings of the 2nd International Symposium on the AHP, vol. 1, pp. 197-202.

Salo, A. A, Hämäläinen, R. P. (1997). On the measurement of preferences in the analytic hierarchy process. Journal of Multi-Criteria Decision Analysis, 6(6), pp. 309-319.

132

Saaty T. L. (1977). A scaling method for priorities in hierarchical structures, Journal of Mathematical Psychology, 15(3), pp. 234-281.

Saaty, T. L. (1980). Analytic Hierarchy Process. McGraw-Hill, New York.

Saaty, T. L. (2004). Decision making – The analytic hierarchy and network processes (AHP/ANP), Journal of Systems Science and Systems Engineering, 13(1), pp. 1–34.

Saaty, T. L. (2008). Decision making with the analytic hierarchy process, Int. J. Services Sciences, 1, pp. 83–98.

Thurstone, L. L. (1927). A law of comparative judgments, Psychological Reviews, 34, pp. 273-286.

Triantaphyllou, E., Lootsma, F. A., Pardalos, P. M., Mann, S. H. (1994). On the Evaluation and Application of Different Scales For Quantifying Pairwise Comparisons in Fuzzy Sets, Journal of Multi-Criteria Decision Analysis, 3(3), pp. 133-155.

Starczewski, T. (2017). Remarks on the impact of the adopted scale on the priority estimation quality. Journal of Applied Mathematics and Computational Mechanics, 16(3), pp. 105-116. doi: 10.17512/jamcm.2017.3.10

Subramanian, N., Ramanathan, R. (2012). A review of applications of Analytic Hierarchy Process in operations management, International Journal of Production Economics, 138(2), pp. 215-241, doi: 10.1016/j.ijpe.2012.03.036.

Vaidya, O. S., Kumar, S. (2006). Analytic hierarchy process: An overview of applications, European Journal of Operational Research, 169, pp. 1-29.

Contact Address

Mgr. Jiří Mazurek, Ph.D. Silesian University in Opava, School of Business Administration in Karvina Univerzitní náměstí 1934/3, 733 40 Karviná, Czech Republic Email: [email protected]


133

DETERMINANTS OF INNOVATION ACTIVITIES AND SME ABSORPTION – CASE STUDY OF GERMANY

Viktor Prokop, Jan Stejskal

Abstract: SME’s are integral pillars in ladder of innovation. Due to their proximity to end users and their flexibility, they are credited with the creation of ground level product and process innovations in their local and world markets. SMEs absorption of innovations is significantly assisted by the entrepreneurial environment, globalization tendencies, rapidly changing technological issues of the environment as well as other determinants. However, depending on the industry, determinants of innovation affects each firm differently depending on the type of innovation considered. The goal of this research is therefore, to analyse what determinants influence the innovation activities of small and medium enterprises across three different German industries, namely, in the Electrical, Chemical and Pharmaceutical and the Metal Industry. Results from the SMEs were compared against themselves as well as SMEs from the three considered industries. This paper used data from the Community Innovation Survey (2010-2012) which employed stratified sampling technique with surveys. Logistic regression tool was used to analyse the impact of certain activities and expenditures, information and competitive strategies on product on process innovation. The research eventually discovered that the determinants influencing product and process innovations in selected enterprises varied according to the size of enterprise analysed. It was proposed that small enterprises should primarily focus on In-house R&D and acquisition of capital assets whilst medium size enterprises would be best served in concentrating on training for innovative activities.

Keywords: Innovation Activities, Determinant, Innovation Capacity, Germany

JEL Classification: O38, O47, H50, C33.

Introduction

Many studies present that the innovation capacity of businesses determines economic growth, both in the company itself and in the entire industry and hence in the economic growth of the country (Jorgenson, Gollop & Fraumeni, 2016; Leigh & Blakely, 2016). It has been shown that innovation is created by the involvement of various internal and external sources (production factors), but also knowledge and interaction between economic subjects (West & Bogers, 2014; Doloreux, 2015). The favorable business environment and the openness of the economy are two important components (Belás, Demjan, Habánik, Hudáková & Sipko, 2015). The open innovation concept highlights importance of the business and innovation environment. Open innovation is currently an interesting interconnected concept (Chesbrough et al., 2014). This concept primarily emphasises the role of external technological sources, which helps to create a competitive advantage (mainly by reducing the time needed to create innovation and market it; Feniser, Lungu & Bilbao, 2017). Another advantage is the orientation of the open innovation concept to small and medium- sized enterprises (SME). SME use cost sharing, experience sharing, and often the risk sharing associated with the production of an innovative products or services (West, Salter, Vanhaverbeke & Chesbrough, 2014). SME does not have sufficient resources to cover

134

increased technology or innovation costs. Some scholars add that SME occupy small positions in the space of knowledge (Feniser, Lungu & Bilbao, 2017), which is due to several factors: the first is the unavailability of investment funds. The second is the lower number of workers, especially those in research and development (again linked to the inability of SMEs to pay really high experts). The third relates to the business innovation strategy, particularly in the CEE region (Malerba et al., 2015). It is still evident that the only SME strategy is to compete with price and possibly with small innovation changes. The innovation production is not a major business strategy for SMEs in many European countries.

SMEs are a very specific group of enterprises, notwithstanding their location in advanced or developing countries, they have very similar characteristics and, above all, behavioural patterns (Ates, Garengo, Cocca & Bititci, 2013). Due to the situation in national and even more in international markets, they are forced to use their production capabilities and in-house knowledge very efficiently. The market situation forces them to be more flexible and dynamic, which often requires a change in business strategy or an increase in specialization on those products that the SME firms are able to produce and sell within the market at the moment with the available resources (Arend, 2014).

Therefore, the collaboration in cooperation-based or knowledge-based networks is much more frequent (McCann & Ortega-Argilés, 2015; Hájek & Stejskal, 2018). SMEs are part of cooperative chains and regional innovation systems in many developed European countries. It helps to increase the innovative absorption of SMEs, use spill-over effects (not only knowledge spill-over effects) and realize the technology transfer (Hajek, Henriques & Hajkova, 2014). All of these abilities strengthen their competitive advantage.

The remainder of this paper is structured as follows. In the next section, the theoretical background on the determinants of innovation activities and the goal of this paper is presented. Section 2 will provide the characteristics of the dataset and the research methodology. Section 3 will list the experimental results and in Section 4, the discussion, conclusion, limitations and future research areas are presented.

1 Theoretical background

High-quality R&D activities are an essential prerequisite for creating innovative outputs. They also significantly influence innovation activities and innovative absorption. As mentioned by Rammer, Czarnitzki & Spielkamp (2009), there are several features of R&D that are likely to result in systematic differences between small and large firms with respect to conducting in-house R&D. They include, for example: (a) R&D in SMEs is determined by a number of projects whose scope can be financed and bring at least reasonable returns on the scale. Research costs are very often fixed, thus reducing the profitability of the business; (b) similarly, R&D in SMEs is associated with high initial investment whose return-on-investment need not be positive in the early years; (c) research activities increase the business risk of the business and therefore it is possible often to encounter so-called ad hoc research, which is of a temporary nature and is usually associated with the realization of a research project. These and other characteristics of the SMEs lead to their less

135

willingness to systematically support their innovation activities and increase innovative absorption (Prokop, Stejskal & Kuvíková, 2017).

Innovative absorption is also often associated (and unfortunately sometimes severely demotivated) by determinants that affect the total competitiveness of individual outputs and the whole of the enterprise. These determinants to competitive advantage protecting are patents, utility patents, design trademarks, copyrights etc. By them, it is possible to ensure (in a globalized world for a relatively short time) the uniqueness of the product on the market. There are a number of studies that agree on the importance of patents and their impact on maintaining a competitive advantage. Bottazzi & Peri (2003) add that patents and protection features can generate the effect of research externalities across space, in generating innovation. They found that innovation is dependent on spill-over effects (and they are given by tacit knowledge in regions). Acs, Anselin & Varga (2002) have similar results. The role of geographically mediated knowledge externalities in regional innovation systems they perceived as a major issue in the research policy.

SMEs are forced to search for additional resources that can overcome the shortcomings given by the nature of SMEs. One option is to accept the public support or other R&D support activities most often from public sector organizations, in Europe, especially from the EU Structural Funds (Klímová, Žítek & Králová, 2019). This has been particularly massive in recent years and is directed, in particular, towards the acquisition of high-quality staff, the creation of innovative capacities, and often the encouragement of innovation cooperation or technology transfer. Some scholars disagree with the effectiveness of such public support, because of a high level of bureaucracy and production inefficiency. Likewise, SMEs are threatened by the effects of the so-called "low-emission fruit", i.e. the dependence on public funds. This is confirmed by a number of studies. For example: Almus & Czarnitzki (2003) analysed the effects of public R&D policy schemes on innovation activities of firms located in Eastern Germany. They also addressed the issue of whether public support stimulates R&D activities in their businesses. They found that enterprises (that did not benefit from public support) expanded innovation activities more than those who used the public support. Kang & Park (2012) examined the effects of inter-firm collaborations as well as the direct and indirect effects of government R&D support on innovation outputs of SMEs in Korea. They concluded that government support through project funding directly and indirectly affects firms' innovation by stimulating internal R&D and domestic upstream and downstream collaborations. Clausen (2009) analysed whether "research" and "development" subsidies influence private R&D activity. The results show that "research" subsidies stimulate private R&D activity, mainly by increasing research spending, while "development" subsidies in substitute private R&D activity, mainly by decreasing development expenditure. Given the results of various studies, it is necessary to state that the effectiveness of public support for innovation activities also depends on the quality of the business environment and public policies.

Many SMEs, which do not realize permanent in-house R&D but want to innovate, use external resources and available technologies from elsewhere. Engaging a SME enterprise into certain cooperative relationships and using network sharing is a way how to get the necessary information sources or knowledge, or access to external

136

resources needed to create innovation. The most common is to use of resources and incentives from customers, suppliers or collaborating organizations, including competitors and the public. Many studies show that these information sources are very effective and provide the necessary impetus for innovation. In addition, thanks to the willingness to share information, experience and knowledge, it is possible to get a synergistically larger amount of information for your own needs back. Some studies show that it is necessary to have certain capabilities and often internal research to use external knowledge (Nilakanta & Scamell, 1990). Balancing between internal R&D efforts and external knowledge acquisition is therefore another major concern of innovation management (Rammer, Czarnitzki & Spielkamp, 2009). Amara & Landry (2005) analysed the information sources as determinants of innovation in manufacturing firms in Canada. Innovation-based enterprises (absolute novelty for the domestic or international market) use a wider range of sources of information and use a wider range of sources of research resources to develop or improve their outputs. Varis & Littunen (2010) examined the information sourcing practices of SMEs related to the development of different types of innovation (product / process / market / organizational). They (unlike other studies) find that the introduction of innovation is associated with the use of rather free (without any consideration) accessible information resources. Analysed enterprises have introduced innovation as their business objective and therefore none of the above innovations have significantly increased the profits of the firm in first years.

It is clear that the results of many published studies differ thanks to the different business environments and the unpredictable behaviour of businesses in different industries. Therefore, it is still necessary to realize the research to help determine which determinants have a significant impact on the innovation capacity of SMEs in the industrial branches. The ideal option is to use a benchmarking approach that can be applied in economies that are linked to each other. Therefore, the goal of this paper is to analyse what determinants influence the innovation activities of small and medium enterprises across three different German industries in comparison with the results obtained for the whole industries. Industrial branches (electrical, chemical and pharmaceutical, and metal) are representing most of the manufacturing industry in Germany. Therefore, the results can be considered significant for other industries in Germany.

2 Data and methodology

To perform the empirical analysis, Eurostat collected and pre-processed the data from the Community Innovation Survey 2010-2012 (CIS). CIS uses harmonized questionnaire created for all EU Member States by Eurostat and combines stratified random sampling with exhaustive surveys. In agreement with previous related studies (e.g. see Archer & Lemeshow, 2006; Coad & Rao, 2008; Schneider & Spieth, 2013), a logistic regression model was fitted. According to previous parts and CIS dataset, we selected following three groups of independent variables (all independent variables are listed in Tab. 1.):

a) activities and expenditures for product and process innovations which include, in particular, research and development. R&D usually helps firms to create new

137

knowledge, solve scientific or technical problems and fuels firms´ innovation, knowledge stock, technology and productivity (Prokop, Odei & Stejskal, 2018);

b) sources of information and co-operation for product and process innovation. Accessing a greater number of knowledge, information and cooperation sources improve firms probability of obtaining knowledge that will lead to a valuable innovation outcomes (Leiponen & Helfat, 2010);

c) methods for maintaining or increasing the competitiveness of product and process innovations. These methods usually help firms to protect their products, processes, and brand (Brem, Maier & Wimschneider, 2016) as well as their internal knowhow and innovations.

We analyse how these variables influence firms´ product and process innovations (dependent variables) within selected German industries. The effects within small and medium enterprises as well as within the whole industries (including small, medium and large enterprises together) are compared and the comparison is presented. In total, the study analysed 474 Electrical (NACE 26-27), 473 Chemical and Pharmaceutical (NACE 19-22) and 465 Metal (NACE 24-25) firms within these industries.

To analyse whether CIS data are not correlated, Spearman's test was used. The hypothesis that the data are correlated with a level of significance at p<0.05 were rejected. Subsequently, the collinearity among the independent variables by Variance Inflation Factor (VIF) for each regression model was tested. Multicollinearity was rejected in the models (VIF<5). All calculations were made using the statistical software SPSS.

Tab. 1: Independent variables Activities and expenditures for product and process innovations

RRDIN In-house R&D

RRDEX External R&D

RMAC Acquisition of machinery, equipment, software & buildings

ROEK Acquisition of existing knowledge from other enterprises or organisations

RTR Training for innovative activities

RMAR Market introduction of innovations

RDSG Design

RPRE Other1

Sources of information and co-operation for product and process innovation

SCON Conferences, trade fairs, exhibitions

SJOU Scientific journals and trade/technical publications

SPRO Professional and industry associations

SENTG Within enterprise or enterprise group

SSUP Suppliers of equipment, materials, components, or software

1 Other in-house or contracted out activities to implement new or significantly improved products and processes such as feasibility studies, testing, tooling up, industrial engineering, etc.

138

SCLPR Clients or customers from the private sector

SCLPU Clients or customers from the public sector

SCOM Competitors or other enterprises in industry

SINS Consultants and commercial labs

SUNI Universities or other higher education institutions

SGMT Government, public or private research institutes

SCON Conferences, trade fairs, exhibitions

SJOU Scientific journals and trade/technical publications

SPRO Professional and industry associations

Methods for maintaining or increasing the competitiveness of product and process innovations

CMPAT Patents

CMUPAT Utility patents

CMRCD Design registration

CMCO Copyright

CMCTM Trademarks

CMLTAD Lead time advantages

CMCPX Complexity of goods or services

CMSEC Secrecy (include non-disclosure agreements)

Source: (CIS 2012, Eurostat)

In the following part, results of binary logistic models that analysed the influence of several independent variables – determinants of innovation activities (grouped into three groups, see above) on the product and process innovations in small and medium enterprises as well as on the product and process innovations within whole industries were presented. Generally, most of significant results were obtained during analyses of the entire industries. However, numbers of important results were also obtained from analyses within SME´s.

3 Results

Tab. 2 presents results for the firms in German Electrical industry. Factors that influenced firms´ innovation activities differ according to the size of enterprise. Research and development activities undertaken by enterprises (RRDIN) to create new knowledge or to solve scientific or technical problems significantly influenced product innovations in small enterprises and within the whole industry. Internal sources of information and co-operation for innovations (SENTG) played an important role within the whole German Electrical industry. Other in-house or contracted out activities to implement new or significantly improved products and processes such as feasibility studies, testing, tooling up, industrial engineering (RPRE), Training for innovative activities (RTR) and Lead time advantages (CMLTAD) most significantly influenced innovation activities within medium enterprises.

139

Tab. 2: German Electrical industry

Small

(under 50 employees) Medium

(50-249 employees) Industry

Product Sig. (Beta)

Process Sig. (Beta)

Product Sig. (Beta)

Process Sig. (Beta)

Product Sig. (Beta)

Process Sig. (Beta)

RRDIN .000(2.172)*** .811(-.132) .380(.772) .091(1.259)* .000(1.851)*** .541(.252)

RRDEX .630(.246) .818(.088) .076(1.330)* .778(.133) .181(.507) .507(.179)

RMAC .431(.398) .022(.949)** .504(.510) .067(1.035)* .425(.322) .000(1.266)***

ROEK .544(-.396) .272(.501) .693(-.363) .759(-.173) .680(-.207) .362(.287)

RTR .781(.139) .076(.683)* .813(.158) .040(1.005)** .703(.147) .001(.917)***

RMAR .007(1.439)*** .109(-.667) .764(-.234) .191(-.765) .017(.926)** .034(-.643)**

RDSG .139(.846) .559(.236) .037(1.696)** .776(.162) .001(1.485)*** .092(.482)*

RPRE .102(.910) .049(1.097)** .007(2.149)*** .353(.590) .006(1.133)*** .089(.642)*

SENTG .000(1.791)*** .011(1.209)** .071(1.130)* .217(.717) .000(1.598)*** .002(1.049)***

SSUP .988(-.009) .001(1.314)*** .063(1.061)* .107(.721) .151(.539) .000(.971)***

SCLPR .001(1.474)*** .453(.281) .228(.660) .889(-.068) .000(1.399)*** .082(.464)*

SCLPU .013(1.529)** .839(.085) .281(.804) .265(-.611) .001(1.451)*** .964(.013)

SCOM .851(-.091) .048(-.765)** .430(.447) .771(.138) .614(.173) .024(-.606)**

SINS .992(-.007) .816(-.151) .872(-.151) .854(.126) .846(.100) .184(.504)

SUNI .134(.773) .322(.391) .973(-.019) .449(.345) .257(.384) .147(.370)

SCON .103(.897) .864(.073) .124(.947) .011(1.305)** .033(.775)** .022(.641)**

SJOU .627(.264) .663(.182) .672(-.275) .774(-.152) .887(.054) .910(.032)

SPRO .020(-1.601)** .448(.421) .949(.047) .758(.179) .055(-.853)* .237(.385)

CMPAT .096(.851)* .405(-.338) .698(.319) .264(-.728) .006(1.066)*** .781(-.083)

CMUPAT .874(.120) .563(.293) .567(.489) .392(.546) .311(-.483) .494(.228)

CMRCD .035(-1.710)** .650(.250) .831(.286) .945(.048) .218(-.638) .364(.326)

CMCO .241(.762) .857(.079) .845(-.168) .681(.222) .742(.140) .453(.219)

CMCTM .295(.552) .797(-.107) .020(1.939)** .606(.287) .002(1.206)*** .267(.312)

CMLTAD .081(.860)* .127(.780) .008(1.755)*** .489(.400) .000(1.285)*** .054(.651)*

CMCPX .007(1.354)*** .780(.123) .381(-.669) .018(1.331)** .065(.665)* .312(.295)

CMSEC .816(.108) .055(.808)* .091(1.237)* .538(.338) .482(.244) .011(.719)*

Source: own

Tab. 3 shows that in-house and external R&D activities and expenditures for product and process innovations most significantly influenced small firms´ innovation activities in German Chemical and Pharmaceutical industries. Results also show that R&D activities play an important role in small and medium enterprises as well as in the whole industry. Similarly, to the results in Tab. 3, internal sources of information and co-operation for innovations had positive impact on the firms´ product and process innovations. Patents, trademarks and lead-time advantages most significantly supported firms´ competitiveness of product and process innovations.

Tab. 3: German Chemical and Pharmaceutical industries

Small



Product Sig. (Beta)

Process Sig. (Beta)

Product Sig. (Beta)

Process Sig. (Beta)

Product Sig. (Beta)

Process Sig. (Beta)

RRDIN .000(2.341)*** .004(1.858)*** .166(.870) 1.000(.000) .001(1.277)*** .203(.454)

RRDEX .012(-1.979)** .033(-1.287)** .614(.305) .195(.654) .452(-.289) .769(.091)

RMAC .012(1.376)** .000(2.635)*** .857(-.120) .012(1.372)** .038(.724)** .000(1.822)***

ROEK .045(1.602)** .634(.291) .772(-.172) .975(-.016) .698(.149) .138(.458)

RTR .527(-.401) .112(.818) .004(1.690)*** .007(1.307)*** .134(.523) .000(1.172)***

140

RMAR .034(1.496)** .091(-.952)* .000(2.388)*** .909(-.054) .000(2.233)*** .330(-.305)

RDSG .653(.314) .601(-.302) .381(.562) .192(.688) .128(.626) .438(-.240)

RPRE .056(1.194)* .730(.206) .466(.416) .092(.840)* .037(.730)** .073(.581)*

SENTG .002(1.507)*** .006(1.544)*** .014(1.267)** .001(1.754)*** .000(1.374)*** .000(1.587)***

SSUP .610(-.265) .193(.618) .344(.445) .637(.215) .407(.250) .111(.413)

SCLPR .093(.794)* .958(-.025) .014(1.225)** .177(.684) .000(1.175)*** .329(.283)

SCLPU .229(.812) .718(-.203) .335(.723) .652(.286) .215(.504) .965(-.014)

SCOM .985(.010) .530(-.310) .717(-.179) .818(.109) .658(-.140) .864(-.047)

SINS .959(-.039) .490(.476) .994(.004) .243(.709) .575(.226) .140(.515)

SUNI .104(.910) .855(.089) .403(-.462) .495(-.344) .079(.574)* .221(.334)

SCON .277(.609) .011(1.371)** .778(-.158) .465(.404) .966(.014) .123(.473)

SJOU .547(.329) .783(-.142) .239(.658) .403(-.455) .342(.327) .345(-.286)

SPRO .501(.521) .066(-1.197)* .497(.405) .226(-.634) .141(.645) .325(-.327)

CMPAT .035(1.298)** .007(1.466)*** .016(1.534)** .095(.846)* .008(1.022)*** .006(.880)***

CMUPAT .399(-.621) .074(-1.133)* .150(-1.086) .212(-.737) .165(-.624) .145(-.519)

CMRCD .747(.292) .276(-.759) .014(-1.976)** .898(.090) .255(-.531) .325(-.371)

CMCO .024(-1.858)** .024(-1.366)** .683(.292) .332(-.541) .079(-.745)* .084(-.563)*

CMCTM .087(1.281)* .924(.052) .146(.954) .359(.474) .003(1.202)*** .218(.390)

CMLTAD .067(1.209)* .323(.587) .015(1.497)** .050(1.098)* .000(1.488)*** .002(1.066)***

CMCPX .386(.560) .015(1.435)** .014(1.477)** .078(.873)* .028(.740)** .058(.550)*

CMSEC .001(1.760)*** .253(.559) .775(-.152) .949(.029) .071(.572)* .105(.459)

Source: own

In Metal industry (see Tab. 4), internal R&D activities, acquisition of machinery, equipment, software & buildings influenced small enterprises’ innovation activities. Internal and some external (e.g. clients and suppliers) sources of information and co-operation also led to the support of small firms innovation activities. Medium firms´ innovation activities were most significantly influenced by other in-house or contracted out activities to implement new or significantly improved products and processes such as feasibility studies, testing, tooling up, industrial engineering (RPRE), design activities (RDSG) and selected methods for maintaining or increasing the competitiveness (e.g. complexity of goods or services, trademarks, lead time advantages). Within the whole industry, number of activities and expenditures for product and process innovations and sources of information and co-operation for product and process innovation played important role. These, for example, are in-house R&D, acquisition of machinery, equipment, software & buildings, training for innovative activities, internal sources of information and co-operation for innovations, universities or other higher education institutions.

Tab. 4: German Metal industry

Small



Product Sig. (Beta)

Process Sig. (Beta)

Product Sig. (Beta)

Process Sig. (Beta)

Product Sig. (Beta)

Process Sig. (Beta)

RRDIN .000(2.172)*** .811(-.132) .380(.772) .091(1.259)* .009(1.018)*** .001(1.233)***

RRDEX .630(.246) .818(.088) .076(1.330)* .778(.133) .826(-.101) .916(-.042)

RMAC .431(.398) .022(.949)*** .504(.510) .067(1.035)* .003(1.113)*** .000(1.391)***

ROEK .544(-.396) .272(.501) .693(-.363) .759(-.173) .182(-.652) .620(-.200)

RTR .781(.139) .076(.683)* .813(.158) .040(1.005)** .948(.025) .002(.986)***

RMAR .007(1.439)*** .109(-.667) .764(-.234) .191(-.765) .000(2.109)*** .814(-.089)

RDSG .139(.846) .559(.236) .037(1.696)** .776(.162) .125(.714) .410(-.323)

141

RPRE .102(.910) .049(1.097)** .007(2.149)*** .353(.590) .001(1.156)*** .036(.709)**

SENTG .000(1.791)*** .011(1.209)** .071(1.130)* .217(.717) .002(1.171)*** .001(1.196)***

SSUP .988(-.009) .001(1.314)*** .063(1.061)* .107(.721) .346(.300) .000(1.156)***

SCLPR .001(1.474)*** .453(.281) .228(.660) .889(-.068) .001(1.211)*** .293(.375)

SCLPU .013(1.529)** .839(.085) .281(.804) .265(-.611) .030(1.064)** .415(-.359)

SCOM .851(-.091) .048(-.765)** .430(.447) .771(.138) .367(.297) .475(-.237)

SINS .992(-.007) .816(-.151) .872(-.151) .854(.126) .436(-.339) .619(.219)

SUNI .134(.773) .322(.391) .973(-.019) .449(.345) .006(.901)*** .000(1.220)***

SCON .103(.897) .864(.073) .124(.947) .011(1.305)** .447(.264) .320(.338)

SJOU .627(.264) .663(.182) .672(-.275) .774(-.152) .307(.358) .109(.552)

SPRO .020(-1.601)** .448(.421) .949(.047) .758(.179) .716(-.155) .522(.259)

CMPAT .096(.851)* .405(-.338) .698(.319) .264(-.728) .133(-.659) .184(.523)

CMUPAT .874(.120) .563(.293) .567(.489) .392(.546) .003(1.394)*** .060(.817)*

CMRCD .035(-1.710)** .650(.250) .831(.286) .945(.048) .719(-.189) .485(-.344)

CMCO .241(.762) .857(.079) .845(-.168) .681(.222) .030(-1.010)** .007(-1.139)***

CMCTM .295(.552) .797(-.107) .020(1.939)** .606(.287) .001(1.355)*** .300(.383)

CMLTAD .081(.860)* .127(.780) .008(1.755)*** .489(.400) .002(1.307)*** .008(1.062)***

CMCPX .007(1.354)*** .780(.123) .381(-.669) .018(1.331)** .065(.769)* .264(.441)

CMSEC .816(.108) .055(.808)* .091(1.237)* .538(.338) .152(.521) .031(.734)**

Source: own

4 Discussion

Results of the analyses showed that determinants influencing enterprises product and process innovations differ according to the size of enterprises. Therefore, there is a need for subsequent analyses because relationship between innovations and firms size represents an important factor (see e.g. Hwang et al., 2015). On the other hand, some factors are the same across industries and across small and medium enterprises as well as across whole industries. These factors are, mainly, research and development activities undertaken by enterprises to create new knowledge or to solve scientific or technical problems (include software development in-house that meets this requirement) – in-house R&D. These results are in agreement with previous analyses that stated the importance of in-house R&D expenditures in the process of creating innovation outputs (Catozzella & Vivarelli, 2014), also within SMEs (Newman et al., 2015). In addition, selected information and co-operation sources, specifically internal (within enterprise or enterprise group) and selected external (e.g. clients, customers, and suppliers) influenced enterprises´ innovation activities. Clients and customers and their important role are pointed, e.g., by the Lead-user theory (Franke et al., 2006). Patents, lead-time advantages, complexity of goods or services and trademarks represent most significant methods for maintaining or increasing the competitiveness of product and process innovations in this study.

The authors proposed some practical implications both for enterprises within different industries and for policy makers on general level. In Electrical industry, it is possible to recommend support of firms´ R&D activities, specifically internal activities and expenditures undertaken by enterprises to create new knowledge or to solve scientific or technical problems. Firms should properly choose their sources of information and co-operation for product and process innovation (e.g. suppliers, clients or customers) and use trademarks, complexity of goods or services and lead-

142

time activities. Similarly, the diversification of public aid and public finance schemes by enterprise size should be recommended. It is obvious that in this industry the problems vary according to the size of the enterprise. Especially in the case of medium-sized enterprises, there were unspecific problems in financing the scientific and research activities of their product and process innovations. Similarly, these businesses are unable to use their capabilities in new knowledge and information. Public incentives to co-operate with the knowledge industry, or incentives for new knowledge-based projects, can help.

In Chemical and Pharmaceutical industries, R&D activities and expenditures play an important role in the process of influencing product and process innovations – mainly, market introduction of innovations and acquisition of machinery, equipment, software & buildings. Enterprises´ internal sources (within enterprise or enterprise group) should be supported. Patent activities, copyrights and lead-time activities represent important factors supporting innovation activities. The results have shown that in the industrial sector (unlike the electrical industry), businesses (irrespective of size) can use their research and development capacities. This area appears to be fully saturated and does not require modification. On the contrary, it was found that all sizes of the enterprises were unable to use the sources of information. Again, it is probably the specificity of this industry. Authors encourage more public policy makers to work with the academic sector.

In Metal industry, similarly to other industries, enterprises should primarily focus on the activities and expenditures for product and process innovations and on proper finding of sources of information and co-operation. In this industry, the biggest differences between small and medium-sized enterprises are again seen. There is a diversification of input variables into production chains. This is due in particular to the specificities of this industry, which must also be reflected in future public policies that will aim to restructure the entire sector due to its small competitive advantage in a globalized world. Public sector attention must be focused primarily on workers and their qualifications for potential industrial change.

Conclusion

The practices of the European countries, as well as the practice of the whole EU, confirm the importance of SMEs for the local, European as well as the global markets. SMEs are an important employer, a creator of added value and GDP. Innovation makers as they are, SMEs absorption of innovation are significantly assisted by the entrepreneurial environment, globalization tendencies, rapidly changing technological issues of the environment as well as other determinants. That is why Governments offer to SMEs some subsidies or benefits to enable them to gain the basic attributes that are necessary for innovation activities. Two fundamental questions remain: (a) whether public policies are focused on the right segment of SMEs and their innovation activities and whether policies are able to help to create innovation, (b) whether firms consider public support to be beneficial and they are interested in despite the connected bureaucracy.

Generally, authors proposed implications for SMEs across industries and point out the factors that are same and influence enterprises´ innovation activities. Small

143

enterprises should primarily focus on the activities and expenditures for product and process innovations, specifically: a) in-house R&D; b) acquisition of machinery, equipment, software & buildings. Within medium size enterprises, training for innovative activities represent most important determinant of product and process innovation activities. Authors showed that different determinants have different influences on both product and process innovations. Analysed determinants have a much stronger effect on product innovations. An interesting result is that the determinants of small businesses seem to be consistent with the major determinants of the entire industry. Medium enterprises are clearly influenced by other determinants (these must be the subject of further research). This result has to be taken into account in practical implications, or taken into account in the country's industrial policy focus. The results of our study should be seen in the context of the German economy. On the other hand, benchmarking can be used as a basis for international comparison.

Finally, it should be remembered the limitations of this study. We work with obsolete data that does not allow defining implications, conclusions, and recommendations for the current present. On the other hand, analogy can be used as a scientific method that builds on the knowledge of the past in order to define more effective measures in the future. Defined conclusions, in the author's opinion, are generally valid and occur to a similar extent in other EU countries even today.

Acknowledgement

This work was supported by a grant provided by the scientific research project of the Czech Sciences Foundation Grant No: 17-11795S.

References

Acs, Z. J., Anselin, L., and Varga, A. (2002). Patents and innovation counts as measures of regional production of new knowledge. Research policy, 31(7), 1069-1085. DOI: http://dx.doi.org/10.1016/S0048-7333(01)00184-6

Almus, M., and Czarnitzki, D. (2003). The effects of public RandD subsidies on firms' innovation activities: the case of Eastern Germany. Journal of Business and Economic Statistics, 21(2), 226-236. DOI: http://dx.doi.org/10.1198/073500103288618918

Amara, N., and Landry, R. (2005). Sources of information as determinants of novelty of innovation in manufacturing firms: evidence from the 1999 statistics Canada innovation survey. Technovation, 25(3), 245-259. DOI: http://dx.doi.org/10.1016/S0166-4972(03)00113-5

Archer, K. J., and Lemeshow, S. (2006). Goodness-of-fit test for a logistic regression model fitted using survey sample data. Stata Journal, 6(1), 97-105.

Arend, R. J. (2014). Entrepreneurship and dynamic capabilities: how firm age and size affect the ‘capability enhancement–SME performance’ relationship. Small Business Economics, 42(1), 33-57. DOI: http://dx.doi.org/10.1007/s11187-012-9461-9

Ates, A., Garengo, P., Cocca, P., and Bititci, U. (2013). The development of SME managerial practice for effective performance management. Journal of Small Business and Enterprise Development, 20(1), 28-54. DOI: http://dx.doi.org/10.1108/14626001311298402

Belás, J., Demjan, V., Habánik, J., Hudáková, M., and Sipko, J. (2015). The business environment of small and medium-sized enterprises in selected regions of the Czech Republic and Slovakia. E+ M Ekonomie a Management, 18(1), 95-110. DOI: http://dx.doi.org/10.15240/tul/001/2015-1-008

144

Bottazzi, L., and Peri, G. (2003). Innovation and spillovers in regions: Evidence from European patent data. European economic review, 47(4), 687-710. DOI: http://dx.doi.org/10.1016/S0014-2921(02)00307-0

breadth, Strategic Management Journal, 31(2), 224-236.

Brem, A., Maier, M., and Wimschneider, C. (2016). Competitive advantage through innovation: the case of Nespresso. European Journal of Innovation Management, 19(1), 133-148

Catozzella, A., and Vivarelli, M. (2014). The catalysing role of in-house RandD in fostering complementarity among innovative inputs. Industry and Innovation, 21(3), 179-196. DOI: http://dx.doi.org/10.1080/13662716.2014.910890

Chesbrough, H., Vanhaverbeke, W., and West, J. (Eds.). (2014). New frontiers in open innovation. Oup Oxford.

Clausen, T. H. (2009). Do subsidies have positive impacts on RandD and innovation activities at the firm level?. Structural Change and Economic Dynamics, 20(4), 239-253. DOI: http://dx.doi.org/10.1016/j.strueco.2009.09.004

Coad, A., and Rao, R. (2008). Innovation and firm growth in high-tech sectors: A quantile regression approach. Research policy, 37(4), 633-648. DOI: http://dx.doi.org/10.1016/j.respol.2008.01.003

Doloreux, D. (2015). Use of internal and external sources of knowledge and innovation in the Canadian wine industry. Canadian Journal of Administrative Sciences/Revue Canadienne des Sciences de l'Administration, 32(2), 102-112. DOI: http://dx.doi.org/10.1002/cjas.1312

Eurostat (2016). Community Innovation Survey (CIS). Retrieved April 13, 2016, from: http://ec.europa.eu/eurostat/web/microdata/community_innovation_survey

Feniser, C., Lungu, F., and Bilbao, J. (2017). The connection between absorptive capacity and open innovation in managerial perspective. In MATEC Web of Conferences (Vol. 121, p. 07008). EDP Sciences.

Franke, N., Von Hippel, E., and Schreier, M. (2006). Finding commercially attractive user innovations: A test of lead-user theory. Journal of product innovation management, 23(4), 301-315. DOI: http://dx.doi.org/10.1111/j.1540-5885.2006.00203.x

Hájek, P., and Stejskal, J. (2018). R&D Cooperation and Knowledge Spillover Effects for Sustainable Business Innovation in the Chemical Industry. Sustainability, 10(4), 1064. DOI: https://doi.org/10.3390/su10041064

Hajek, P., Henriques, R., and Hajkova, V. (2014). Visualising components of regional innovation systems using self-organizing maps—Evidence from European regions. Technological Forecasting and Social Change, 84, 197-214. DOI: http://dx.doi.org/10.1016/j.techfore.2013.07.013

Hwang, Y. S., Hwang, M. H., and Dong, X. (2015). The relationships among firm size, innovation type, and export performance with regard to time spans. Emerging Markets Finance and Trade, 51(5), 947-962. DOI: http://dx.doi.org/10.1080/1540496X.2015.1061386

Jorgenson, D., Gollop, F. M., and Fraumeni, B. (2016). Productivity and US economic growth (Vol. 169). London: Elsevier.

Kang, K. N., and Park, H. (2012). Influence of government RandD support and inter-firm collaborations on innovation in Korean biotechnology SMEs. Technovation, 32(1), 68-78. DOI: http://dx.doi.org/10.1016/j.technovation.2011.08.004

Klímová, V., Žítek, V., and Králová, M. (2019). How Public R&D Support Affects Research Activity of Enterprises: Evidence from the Czech Republic. Journal of the Knowledge Economy, 1-20. DOI: https://doi.org/10.1007/s13132-019-0580-2

Leigh, N. G., and Blakely, E. J. (2016). Planning local economic development: Theory and practice. London: Sage Publications.

Leiponen, A., and Helfat, C. E. (2010). Innovation objectives, knowledge sources, and the benefits of

145

Malerba, F., Caloghirou, Y., McKelvey, M., and Radoševic, S. (Eds.). (2015). Dynamics of knowledge intensive entrepreneurship: Business strategy and public policy (Vol. 38). London: Routledge.

McCann, P., and Ortega-Argilés, R. (2015). Smart specialization, regional growth and applications to European Union cohesion policy. Regional Studies, 49(8), 1291-1302. DOI: http://dx.doi.org/10.1080/00343404.2013.799769

Newman, A., Gao, Y., and Zheng, J. (2015). Overcoming the innovation challenge: examining the determinants of new product innovation in Chinese SMEs. In Developments in Chinese Entrepreneurship (pp. 33-57). New York: Palgrave Macmillan US.

Nilakanta, S., and Scamell, R. W. (1990). The effect of information sources and communication channels on the diffusion of innovation in a data base development environment. Management Science, 36(1), 24-40. DOI: http://dx.doi.org/10.1287/mnsc.36.1.24

Prokop, V., Odei, S. A., and Stejskal, J. (2018). Propellants of University-Industry-Government Synergy: Comparative Study of Czech and Slovak Manufacturing Industries. Ekonomicky casopis, 66(10), 987 – 1001

Prokop, V., Stejskal, J., and Kuvíková, H. (2017). The Different Drivers of Innovation Activities in European Countries: A Comparative Study of Czech, Slovak, and Hungarian Manufacturing Firms. Ekonomicky casopis, 65(1), 31-45

Rammer, C., Czarnitzki, D., and Spielkamp, A. (2009). Innovation success of non-RandD-performers: substituting technology by management in SMEs. Small Business Economics, 33(1), 35-58. DOI: http://dx.doi.org/10.1007/s11187-009-9185-7

Schneider, S., and Spieth, P. (2013). Business model innovation: Towards an integrated future research agenda. International Journal of Innovation Management, 17(01), 1340001.

Varis, M., and Littunen, H. (2010). Types of innovation, sources of information and performance in entrepreneurial SMEs. European Journal of Innovation Management, 13(2), 128-154. DOI: http://dx.doi.org/10.1108/14601061011040221

West, J., and Bogers, M. (2014). Leveraging external sources of innovation: a review of research on open innovation. Journal of Product Innovation Management, 31(4), 814-831. DOI: http://dx.doi.org/10.1111/jpim.12125

West, J., Salter, A., Vanhaverbeke, W., and Chesbrough, H. (2014). Open innovation: The next decade. Research Policy, 43(5), 805-811. DOI: http://dx.doi.org/10.1016/j.respol.2014.03.001

Contact Address

Ing. Viktor Prokop, Ph.D. University of Pardubice, Faculty of Economics and Administration, Institute of Economic Sciences Studentska Street, 95, 53210, Pardubice, Czech Republic Email: [email protected] Phone number: +420466036067

doc. Ing. Jan Stejskal, Ph.D. University of Pardubice, Faculty of Economics and Administration, Institute of Economic Sciences Studentska Street, 95, 53210, Pardubice, Czech Republic Email: [email protected] Phone number: +420466036067


146

STAKEHOLDER POWER, SUSTAINABILITY REPORTING, AND CORPORATE GOVERNANCE : A CASE STUDY OF

MANUFACTURING INDUSTRY AT INDONESIA’S STOCK EXCHANGE

Yulita Setiawanta, Agus Purwanto

Abstract: This research explains about the way the stakeholder power can affect the implementation of sustainability reporting and the corporate governance at a manufacturing industry in Indonesia listed on Indonesia’s Stock Exchange. The results of this study are expected to provide an overview to prospective foreign investors who are interested in investing in Indonesia public companies which related to organizational behavior and governance style. However, there was less company that reported sustainability reporting continuously from 2007 up to 2016. This becomes a special and interesting concern to be investigated as the awareness phenomenon for the industries in Indonesia. With specified criteria found 60 observation data for 10 years and used multiple linear regression. The result showed that Government Control and Audit Control affected the implementation of Sustainability Reporting while the Internal control and shareholder control had no effect. The special note for shareholder control had a negative coefficient and it became a phenomenon for future research. Another finding was that the implementation of Sustainability Reporting and Government Control had an effect on good corporate governance.

Keywords: Stakeholder Power, Sustainability Reporting, Corporate Governance, Government Control, Audit Control.

JEL Classification: G30, G34, M14, M420.

Introduction

In the last three years, Indonesia has made a great achievement in economic development by changing the consumptive policy into the productive policies in the period of Jokowi-JK Working Cabinet (Schwarz, 2015). Along with the economic growth from 2012 to 2017, some problems in the manufacturing industry such as infrastructure, certainly the workforce and labor problems, even though the company continues to grow. Until the research finding was done by United Nations Industrial Development Organization (UNIDO) showed that the global manufacturing industry at quartal III in 2012 grew only 0.2 percent compared to the previous years (Lestari and WSU, 2017).

At present many companies in countries such as France, Germany, Japan, the United Kingdom, and the United States, report environmental and social issues related to the company's commercial activities to their stakeholders (Burhan and Rahmanti, 2012; Ogundare, 2013) Likewise for manufacturing companies in Indonesia (Deswanto and Siregar, 2017). Global Reporting Initiative (GRI), as it provides a common ground for sustainability reporting, Brown, de Jong, and Levy, (2009) and has been very successful in terms of the adoption rate, comprehensiveness, prestige, and visibility (Loh, Thomas, and Wang, 2017). In Indonesia, one of the reporting models is Sustainability Reporting based on the Global Reporting Initiative (GRI)

147

(Ball and Bebbington, 2008). This awareness makes the role of the company very important in every progress towards sustainability (Gray, 1996). In 2005 there were only 10 companies, but in 2016 it had risen above 120 companies. Awareness of reporting on Sustainability Reports needs to be appreciated because it is not mandatory but still voluntary disclosure (Quick, 2008). This awareness can be caused by internal pressure such as company shareholders or the company's internal control system while external pressure can also occur and be carried out by the government and external auditors (Lu and Abeysekera, 2014; Agyei-mensah, 2016).

In the last three years, Indonesia has made a great achievement in economic development by changing the consumptive policy into the productive policies in the period of Jokowi-JK Working Cabinet (Schwarz, 2015). Along with the economic growth from 2012 to 2017, some problems in manufacturing industry such as infrastructure, certainly the workforce and labor problems, even though the company continues to grow. Until the research finding done by United Nations Industrial Development Organization (UNIDO) showed that the global manufacturing industry at quartal III in 2012 grew only 0.2 percent compared to the previous years (Lestari and WSU, 2017).

The impact of this pressure, investors reacted positively to the environmental performance achieved by the company reported in the Sustainability Report as part of the implementation of good corporate governance and able to increase company value. This is an interesting question, is it possible for manufacturing companies in Indonesia, that internal and external pressure through control activities can affect the reporting process or SR that they make voluntarily and whether it also has an influence on the implementation of Good Corporate Governance in the company. That was to answer if good corporate governance for all stakeholders cannot be underestimated (Lu & Abeysekera, 2014). A good company’s system will ensure some vital points; they are openness and transparency to the shareholders and the regulator on risk management in a company (Elliott and Elliott, 2007). Openness and transparency to the shareholders will be made minimizing the asymmetry of information (Agyei-mensah, 2016).


The sustainable company’s growth is reported through the voluntary disclosure done by a company in Indonesia; year by year, it becomes a trend for a company to inform the economic, social, and environmental performances, Deswanto & Siregar, (2017) and it becomes an awareness and a compliant for all stakeholders (Ball and Bebbington, 2008). For example, a company in China informs the economic, social, and environmental performances that had been done by the public company in 2008 after the issuance of social responsibility guidance or sustainability reporting from Shanghai Stock Exchange (SSE) in 2006 (Qiu, 2017). The sustainability reporting guidance also grows gradually such as a guidance from GRI (General Reporting Index) as suggested by Loh, Thomas, & Wang, (2017) also (Li et al., 2018).

Disclosure of Sustainability Reporting in Indonesia is based on regulations relating to environmental and social responsibility in Law Number 40 of 2007 concerning Limited Liability Companies and until 2016 only 120 companies published Sustainability Reporting (Kurniawan, Sofyani and Rahmawati, 2018). In its

148

development, it has not been consistent in terms of publication of reports. Is it possible for company stakeholders in Indonesia to have an influence on disclosure of sustainability reports, as has been revealed in research conducted on companies in china by (Lu and Abeysekera, 2014).

Auditors are professionally influential in guiding their clients in initiating and promoting new accounting practices (e.g., social responsibility accounting) (Ahmad, Hassan and Mohammad, 2003). Although, Lu and Abeysekera,(2014) found that audit control could not affect the sustainability reporting disclosure. Nevertheless, according to Barrett (2005), the sustainability reporting disclosure could be well-arranged if there was an independent auditor who participated in the composition. The ownership of control in a company is the most important factor for the information of environmental disclosure (Haladu and Haliru Beri, 2016). However, the study in China revealed that the stakeholder power from the government had an impact on the sustainability reporting disclosure (Lu and Abeysekera, 2014).

Stakeholder perspective, which recognizes the ability of the government to influence corporate strategy and performance via regulations (Freeman, Wicks and Parmar, 2004). The results of the research conducted by Liu and V. Anbumozhi, (2009) found that the Chinese government had positive and significant influence on environmental disclosures. Meanwhile government control via regulations has no effect on voluntary disclosures on companies in Indonesia (Basuki and Patrioty, 2011).

Disperse corporate ownership increases pressure for management to disclose voluntary information (Ullmann, 1985). Previous studies have examined the power of shareholders to influence corporate social and environmental disclosure as part of Sustainability Reporting (Lu and Abeysekera, 2014). Government share ownership has an influence on voluntary disclosures for companies in Malaysia (Said, Zainuddin and Haron, 2009). Although voluntary disclosure for companies in Malaysia is still categorized as weak (Zain, Mustaffa and Tamoi, 2006). In Indonesia, publication of sustainability report has started to become a trend, and has become from the interests of shareholders. Other factors are driven by the annual award of sustainability reports initiated by the National Center for Sustanaibility Reporting (NCSR) (Tarigan and Semuel, 2015).

This reflects that the stakeholder power lacks intervention or lacks control for company’s internal party. It is important to be noted that corporate governance has a relation to the internal control (Agyei-mensah, 2016). Lack of good corporate governance in a country, such as the companies listed on Ghana’s Stock Exchange, has resulted in a bad performance and a failure in a company (Agyemang and Castellini, 2015). Meanwhile, (Tristiarini1, Utomo1 and Setiawanta1, 2019) have found that the corporate reputation measured by using corporate responsibility and corporate governance has a significant influence to market value.

The research on the relationship between sustainability reporting and corporate governance or corporate governance and the disclosure of sustainability reporting had been done by Mahmood, Kouser, Ali, Ahmad, & Salman (2018). They stated that the elements of corporate governance specifically were the size of the Board of directors, professionalism of the board of directors; even, the female director could affect and

149

increase the disclosure of sustainability reporting). However, Said, Zainuddin and Haron,.(2009) did not find such relationship at the company in Malaysia.

The disclosure of company’s internal information and the company’s elements is expected to give a motivation in achieving good company’s performance for the attainment of good corporate governance (Lu and Abeysekera, 2014). By the existence of the problems occurred in manufacturing industries, the research gap found in this research, and the phenomena happened in international setting, then interested in investigating the empirical test on the stakeholder power toward the implementation of sustainability reporting and good corporate governance in the manufacturing industry listed on Indonesia’s Stock Exchange that has reported their Sustainability Reporting transparently.

2 Methods

This research was conducted in several manufacturing companies listed on Indonesia’s Stock Exchange within a period of 2007-2016. This research model used two models. The first model was aimed at testing the direct impact of Government Control, Auditor Control, Shareholder Control, and Internal Control toward Sustainability Reporting. Meanwhile, the second model was aimed at observing the implementation of Sustainability Reporting and Government Control toward corporate governance.

The control variable used here was the company size by considering that larger companies has implemented the principles of good corporate governance and disclosed the sustainability reporting. The company size in this research was measured by using natural logarithm from the company’s total asset (Lu and Abeysekera, 2014; Mahmood et al., 2018). The stakeholder theory focuses on the expectation of a certain interest group and the company that is persistent to widely serve the interest group will create a better value time by time (Harrison and Wicks, 2013)

This group has a variable characteristic and an impact level toward the company’s activity (Lu and Abeysekera, 2014). According to Ullmann (1985), the characteristic and the impact level can be manifested as the stakeholder power and they have a capacity to affect the managerial decision strategy in form of control for resources needed by the company. Therefore, the strength of stakeholders consists of government control, auditor control, shareholder control and internal control that are considered capable of influencing the implementation of sustainability reporting and good corporate governance practices.

150

Tab. 1 : Measurement for each variable No Variabel Measurement 1 Audit control*,** (

independent variable) 1 for companies audited by big four and 0 for the opposite

2 Internal control** ( independent variable)

(Actual disclosure / Total possible disclosure)

3 Government Control* (independent variable)

1 for companies owned by the government and 0 otherwise

4 Sharesholder control* (independent variable)

Precentage of shares owned by the largest sharesholder at the end of the year

5 Sustainability Reporting* (dependent/independent variable variable)

Disclosure Quantity in SR

6 Corporate Governance***(dependent variable)

Procentage of Managerial Ownerships

7 Firm Size* (control variable)

Natural logarithm of total assets

Source: Lu and Abeysekera, (2014)*; Agyei-mensah, (2016)**, Morck, Shleifer and Vishny, (1988)***

Based on the measurement of the variables in Tab. 1 above, the research model developed consisted of two models with the following equation :

ISR = β0 + β1GC + β2AC + β3SC + β4IC +ε ......................................(Model I ) (1)

CG = β0 + β5ISR + β6GC +ε ...............................................................(Model II) (2)

A research done by Freeman (2004) stating that a perspective from the stakeholders admits the government’s capability to affect the company’s strategy and the performance through a regulation and also (Lu and Abeysekera, 2014). Government control has an effect on the disclosure implementation done by the company (Liu and V Anbumozhi, 2009). So, the hypothesis as follows:

H1: Government control can affect the implementation of sustainability reporting.

As a research was done by Taylo (1992), proving that the existence of a positive relationship between the auditor and the company’s disclosure in a gas and oil industry in Australia. In a study done in Malaysia, Ahmad, Hassan and Mohammad (2003) also found that the company audited by Big-5 auditor disclosed more environmental information in their annual report. Thereby, the hypothesis as follows:

H2: Auditor control can affect the implementation of sustainability reporting.

Keim (1978) states that the distribution of the company’s ownership becomes less concentrated; if the demand is put in a company by the shareholders, it becomes more. Disperse corporate ownership increases the pressure to disclose information on social responsibility (Ullmann, 1985b). Thereby, the hypothesis as follows:

H3: Shareholder control can affect the implementation of sustainability reporting.

The agency conflict is explained in the agency theory and it is proposed by Jensen and Meckling (1976) also believe that the agency theory also explains the reason why the manager discloses the internal control information voluntarily, and reported

151

through sustainability reporting so the stakeholders will understand the internal control of a company. Therefore, the hypothesis as follows:

H4: Internal control can affect the implementation of sustainability reporting.

The previous study also stated that corporate governance had a relationship to the disclosure or the implementation of sustainability reporting Mahmood, et al (2018). Theoretical arguments alone cannot unambiguously predict the relationship between management ownership (as corporate governance driver) and market valuation (Morck, Shleifer and Vishny, 1988). So, the hypothesis as below:

H5: The implementation of sustainability reporting can affect corporate governance.

A government will provide an intervention to the company’s internal party if the company’s internal party does an action that can damage the stakeholders as a case happened in China (Lu and Abeysekera, 2014). It is not only happened in China, but it is also happened in both developed countries and developing countries. Theoretically, the hypothesis as follows:

H6: Government control can affect corporate governance.

3 Problem solving

The classical assumption test referred to a definition proposed by Ghozali (2016) can be seen in Tab. 2.

Tab. 2 : Data Quality Test (Classic assumption test)

No Variabel Model I Model II

VIF Tolerance t Glasjer VIF Tolerance t Glasjer 1 GC 1,094 0,914 -1,794* 1,022 0,979 -0,671*2 IC 1,049 0,953 0,484* 3 AC 1,063 0,941 1,087* 4 SC 1,094 0,905 -1,483* 5 ISR 1,028 0,973 0,471* 1,015 0,985 0,238*6 CG 1,014 0,986 1,047*

*. > 0,05; Source : (Author’s Result )

Based on Tab. 2, the significance value in the ANOVA table obtained a value of 0.000 that can be concluded that the two models used in this research were good. The value for the Kolmogorov-Smirnov normality test was 0.444 in the first model and 0.177 in the second model. It showed that both values exceeded a value of 0.05 and indicating that the data were normally distributed and appropriate to be used. Other classical assumption tests such as multicollinearity test If the VIF value is less than 10 and or the Tolerance value is more than 0.01 then it can be conclusively concluded that no multicollinearity problem has been proven from the VIF table column and Tolerance. Heteroscedasticity, if the significance value (Sig.)> 0.05 (glasjer collum) then there are no symptoms of heteroscedasticity, based on the table above the government control variable (GC) and shareholder control (SC) are still experiencing problems and are part of the limitations in this study.

152

Tab. 3. Model Fit of ANOVAb Model I Sig

Regression 0.0000

Model II Sig

Regression 0.0000

Source : (Author’s Result )

Based on Tab. 3, the significance value in the Model Fit of ANOVAb table obtained a value of 0,000 both for model I and model II, it can be concluded that the two models used in this study were good. To know whether an independent variable can affect the dependent variable positively or negatively, it can be seen from the t-test as follows:

Tab. 4 : Regression Results

Variabel Model I Model II

UCB t statistik UCB t statistik (Constant) 6,741 7,534 8,238 13,168 GC 0,679 2,282** 1,491 4,267*** IC 0,216 0,534 AC 2,235 5,494*** SC -0,002 -0,222 LnA 0,033 0,512 1,074 13,623*** ISR 0,571 1,677** CG 7,153 64,576 Noted : *** Significant 1%; ** Significant 5%; * Significant 10%; UCB = Unstandardized Coefficients Beta


Based on Tab. 4, it can be explained that the first hypothesis can be accepted. It had been obtained a UCB value of 0.679, the t-statistic value of 2.282, at significance value below of 5%. The second hypothesis can be accepted since it had a UCB value of 2.235, the t-statistic value of 5.494, at significance value below of 1%. The third hypothesis could not be accepted since it had a UCB value of -0.002 and the t-statistic value of -0.222 with significance value above 10%. The fourth hypothesis could not be accepted since it had a UCB value of 0.216, the t-statistical value of 0.534 at significance value above 10%. The fifth hypothesis can be accepted since it had a UCB value of 0.571, a t-statistic value of 13.623, at a significance value below of 5%. The sixth hypothesis can be accepted since it had a UCB value of 1.491, the t-statistical t value of 4.267, and significance value below of 1%.

Tab. 5. Determinant Coefficient Description Adjusted R Square

Model I (Y1) 0,343 Model II (Y2) 0,764


From the adjusted R Square at tabel 5 can be explained that the influence of Goverment Control (GC), Internal Control (IC), Auditor Control (AC) and Shareholder Control (SC) as model I, gives a value of 0.343 which can be interpreted

153

that Dependent latent variables can be explained by independent latent variables of 34.3%, while 65.7% is explained by other variables outside the study. And also at model II adjusted R Square gives a value of 0.764 which can be interpreted that Dependent latent variables can be explained by independent latent variables such as Goverment Control (GC) and Sustainability Reporting (ISR) of 76.4%, while 23.3% is explained by other variables outside the study

4 Discussion

First hypothesis of this research accepted and finding is in line with a statement by Freeman, Wicks, & Parmar (2004), stating that the perspective of the stakeholders who admit the government capability in affecting the company’s strategy and performance through a regulation. In addition, this research finding is relevant to the finding by Liu and Anbumozhi (2009), stating that China’s government can affect the disclosure of sustainability in form of accountability report toward their environment. Second hypothesis Cannot accepted, but the result is not relevant to the construction of the hypothesis as stated by Ahmad, Hassan and Mohammad (2003) which finding that a company audited by Big-5 auditor discloses mode environmental information in their annual report and the society tends to see that a bigger audit company such as Big Four can be fairer and more independent (Choi and Kim, 2010).

Third hypothesis could not be accepted, it means that the corporation pays full attention to social responsibility (Qian and Xing, 2016). Even though the company’s owners receive the information on the regulation of social responsibility disclosure, they could not perform the regulation optimally. Fourth hypothesis could not be accepted, it can be stated limitedly on the company sample that the effectiveness of internal control based on the perspective of stakeholder theory X. Li et al.,(2018) in the manufacturing companies in Indonesia was not able to be a good and optimal way in the process of sustainability reporting. Although the company’s management had more freedom in choosing the method of accounting estimation (Hogan and Wilkins, 2008).

Fifth hypothesis can be accepted, it can be concluded that the implementation of sustainability reporting strongly affects the GCG dominantly. Since there are compulsory principles to be disclosed in the sustainability reporting such as equity, comparable, accurate, sequential based on appropriate time, and accountable, an entity that discloses the sustainability reporting can reflect that the entity implements good corporate governance (Lu and Abeysekera, 2014; Janggu et al., 2014; Mahmood et al., 2018). Sixth hypothesis can be accepted, it can be concluded if a governance control increases, GCG will also increase. It is because of the performance of an entity gives an effect to the economic performance in a country (Mantari and Nuryasman, 2017). Thereby, the company’s internal party should be demanded to implement a correct and good corporate governance based on the principles of good corporate governance.

Conclusion

This study presents the current investigation towards the implementation of the issuer’s sustainability reporting in the framework of legitimacy and the stakeholders in the companies existed Indonesia. The empirical findings provided an important perception of the effect of stakeholder power toward the sustainability reporting and GCG. Audit control Choi and Kim (2010) and governance control (Liu and

154

Anbumozhi, 2009; Freeman, 2004) were the variables that could affect the implementation of sustainability reporting and governance control (Lu and Abeysekera, 2014; Mantari and Nuryasman, 2017). In addition, the implementation of sustainability reporting was the factor that could affect good corporate governance. This research finding is strongly consistent with the legitimacy theory due to the tendency of a company to comply the public supervision, such as a big company owned by the government and the company audited by the famous public accountant such as an auditor that belongs to the big four auditor.

This research also provides some unexpected yet thorough findings. As an example, the involvement of Big Four in financial audit also makes a substantial difference in the implementation of sustainability reporting. These findings provide a practical implication for the policymaker and other relevant stakeholders. When the Indonesian government and its institutions, as both a regulator or a facilitator, have not issued the regulation and the guidance yet in the process of composing the sustainability reporting, only several companies listed on Indonesia’s Stock Exchange that have a willingness to voluntarily disclose the process of company’s sustainability. Therefore, the Indonesian government needs to perform the sustainability attempt by providing guidance for the regulation and detailed guidance on the content and the level of sustainability reporting implementation. It aims to help the company in order to be able to communicate their social and environmental actions to the regulatory agency and other stakeholders.

This study provides a contribution in the literature about social and environmental accounting by enlarging the scope of the research that still exists at the social and environmental disclosure in the company’s sustainability reporting for developing countries context i.e. Indonesia. Continuous evolution and increasing importance of sustainability concepts and practices among individuals, organizations and communities around the world it seems that it needs development conceptual approach to the theory of sustainability management for management applications, especially for public companies in Indonesia (Starik and Kanashiro, 2013).

However, this research finding should be interpreted by considering some limitations in this research. First, there were few numbers of the sample used in this research are limited. It can limit the finding implementation to the company that is not listed in this research. Secondly, even though the extensive attempt on the determinant choice and the accurate proxy development for various variables had been done, subjectivity could not be avoided. Thirdly, it was due to inadequacy number of companies that disclosed the sustainability reporting periodically year by year. The future research can consider other determinant factors that have more potency and uses a greater number of samples.

Acknowledgement

This contribution is supported by the Internal Grants of the Faculty of Economics and Business, Dian Nuswantoro University, Accounting undergraduate program, academic year of 2017_2018 : " Internal Independent Research of Faculty ".

155

Contact Address

References

Agyei-mensah, B. K. (2016) ‘Internal control information disclosure and corporate governance : evidence from an emerging market’, Corporate Governance, 16(1), pp. 79–95. doi: 10.1108/CG-10-2015-0136.

Agyemang, O. S. and Castellini, M. (2015) ‘Corporate Governance in an Emergent Economy : A case of Ghana Article information ’:, (February). doi: 10.1108/CG-04-2013-0051.

Ahmad, Z., Hassan, S. and Mohammad, J. (2003) ‘Determinants of environmental reporting in Malaysia’, International Journal of Business Studies, 11(1), pp. 69–90.

Ball, A. and Bebbington, J. (2008) ‘Editorial: Accounting and reporting for sustainable development in public service organizations’, Public Money and Management, 28(6), pp. 323–326. doi: 10.1111/j.1467-9302.2008.00662.x.

Basuki and Patrioty, C. N. (2011) ‘Pengaruh Regulasi Pemerintah, Tekanan Masyarakat, Tekanan Organisasi Lingkungan, Tekanan Media Massa, Terhadap Corporate Social Disclosure’, EKUITAS (Jurnal Ekonomi dan Keuangan), 15(1), pp. 23–39. doi: 10.24034/j25485024.y2011.v15.i1.2271.

Brown, H. S., de Jong, M. and Levy, D. L. (2009) ‘Building institutions based on information disclosure: lessons from GRI’s sustainability reporting’, Journal of Cleaner Production. Elsevier Ltd, 17(6), pp. 571–580. doi: 10.1016/j.jclepro.2008.12.009.

Burhan, A. H. N. and Rahmanti, W. (2012) ‘The Impact of Sustainability reporting on Company Performance’, Journal of Economics, Business, and Accountancy Ventura, Vol 15, No(110), p. 257–272.

Choi, J. and Kim, J. (2010) ‘Audit Office Size , Audit Quality and Audit Pricing’, Auditing: A Journal of Practice and Theory. Research Collection School Of Accountancy, 19(1), p. 73.

Deswanto, R. B. and Siregar, S. V. (2017) ‘The Association between environmental disclosure with financial performance, environmental performance, and firm value’, Social Responsibility Journal. doi: 10.1108/09574090910954864.

Elliott, B. and Elliott, J. (2007) Financial Accounting and Reporting. Eleventh e. Edinburgh Gate Harlow Essex CM20 2JE England and: Prentice-Hall Inc.- Financial Times. Available at: www.pearsoned.co.uk.

Freeman, R. E. (2004) ‘The Stakeholder Approach Revisited’, Zeitschrift für Wirtschafts-und Unternehmensethik, 5(3), pp. 228–241. doi: 10.3763/jsfi.2010.0008.

Freeman, R. E., Wicks, A. C. and Parmar, B. (2004) ‘Revisited ” Stakeholder Theory and “ The Corporate Objective Revisited ”’, (January 2018). doi: 10.1287/orsc.1040.0066.

Ghozali, I. (2016) Desain Penelitian Kuantitatif dan Keualitatif untuk akuntansi , Bisnis dan Ilmu sosial lainya. 1st edn. Yoga Pratama.

Gray, R. (1996) ‘Corporate Reporting for Sustainable Development: Accounting for Sustainability AD2000’, Business and the Environment, (1), pp. 17–45.

Haladu, A. and Haliru Beri, M. (2016) ‘Corporate Characteristics and Sustainability Reporting Environmental Agencies’ Moderating Effects’, IOSR Journal of Humanities and Social Science, 21(8), pp. 19–30. doi: 10.9790/0837-2108101930.

Harrison, J. S. and Wicks, A. C. (2013) ‘Stakeholder Theory, Value, and Firm Performance’, Business Ethics Quarterly, 23(1), pp. 97–124. doi: 10.5840/beq20132314.

Hogan, C. E. and Wilkins, M. S. (2008) ‘Evidence on the audit risk model: Do auditors increase audit fees in the presence of internal control deficiencies?’, Contemporary Accounting Research, 25(1). doi: 10.1506/car.25.1.9.

Janggu, T. et al. (2014) ‘Does Good Corporate Governance Lead to Better Sustainability Reporting?

156

An Analysis Using Structural Equation Modeling’, Procedia - Social and Behavioral Sciences. Elsevier B.V., 145, pp. 138–145. doi: 10.1016/j.sbspro.2014.06.020.

Jensen, M. C. and Meckling, W. H. (1976) ‘Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure’, Journal of Financial Economics, 3(4), pp. 305–360. doi: http://dx.doi.org/10.1016/0304-405X(76)90026-X.

Keim, G. D. (1978) ‘Corporate Social Responsibility: An Assessment of the Enlightened Self-Interest Model’, The Academy of Management Review1, 3(1), pp. 32–39.

Kurniawan, T., Sofyani, H. and Rahmawati, E. (2018) ‘Pengungkapan Sustainability Report dan Nilai Perusahan: Studi Empiris di Indonesia dan Singapura’, Kompartemen: Jurnal Ilmiah Akuntansi, 16(1), pp. 1–20. doi: 10.30595/kompartemen.v16i1.2100.

Lestari, E. P. and WSU, I. (2017) ‘Analisis Kinerja Industri Manufaktur Di Indonesia’, Journal of Research in Economics and Management, 17(1), pp. 85–97.

Li, X. et al. (2018) ‘The Effectiveness of Internal Control and Corporate Social Responsibility: Evidence from Chinese Capital Market’, Sustainability-MPDI, 10(11), p. 4006. doi: 10.3390/su10114006.

Liu, X. and Anbumozhi, V. (2009) ‘Determinant factors of corporate environmental information disclosure : an empirical study of Chinese listed companies’, Journal of Cleaner Production. Elsevier Ltd, 17(6), pp. 593–600. doi: 10.1016/j.jclepro.2008.10.001.

Liu, X. and Anbumozhi, V. (2009) ‘Determinant factors of corporate environmental information disclosure: an empirical study of Chinese listed companies’, Journal of Cleaner Production. Elsevier Ltd, 17(6), pp. 593–600. doi: 10.1016/j.jclepro.2008.10.001.

Loh, L., Thomas, T. and Wang, Y. (2017) ‘Sustainability reporting and firm value: Evidence from Singapore-listed companies’, Sustainability (Switzerland), 9(11), pp. 1–12. doi: 10.3390/su9112112.

Lu, Y. and Abeysekera, I. (2014) ‘Stakeholders â€TM power , corporate characteristics , and social and environmental disclosure : evidence from China’, Journal of Cleaner Production, 64(1), pp. 426–436.

Mahmood, Z. et al. (2018) ‘Does corporate governance affect sustainability disclosure? A mixed methods study’, Sustainability (Switzerland), 10(1), pp. 1–20. doi: 10.3390/su10010207.

Mantari, J. S. and Nuryasman (2017) ‘Moderation Effect of Exchange Rate to Signaling Theory Validity in Indonesia Stock Exchange’, Business and Management Studies, 3(1), p. 80. doi: 10.11114/bms.v3i1.2259.

Morck, R., Shleifer, A. and Vishny, R. W. (1988) ‘Managerial Ownerships and market Valuation * An Empeirical Analysis’, Journal of Financial Economics, 20, pp. 293–315.

Ogundare, E. A. (2013) ‘The Impact of Sustainability reporting on Organisational Performance-The Malaysia Experience’, International Journal of Accounting & Business Management, 1(2), pp. 82–100.

Qian, W. and Xing, K. (2016) ‘Linking Environmental and Financial Performance for Privately Owned Firms: Some Evidence from Australia’, Journal of Small Business Management, 56(2), pp. 330–347. doi: 10.1111/jsbm.12261.

Qiu, J. (2017) ‘Quality of CSR Reporting in China: A Comparative Analysis Between State- and Privately-Owned Real Estate Companies’, Social Impact Research Experience (SIRE).

Quick, R. (2008) ‘Voluntary sustainability reporting practices in Germany: a study on reporting quality’, Contabilidade e Gestão, pp. 7–35. Available at: http://www.otoc.pt/downloads/files/1219660131_7-35.pdf.

Said, R., Zainuddin, Y. H. and Haron, H. (2009) ‘The relationship between corporate social responsibility disclosure and corporate governance characteristics in Malaysian public listed companies’, 5(2), pp. 212–226. doi: 10.1108/17471110910964496.

Schwarz, A. (2015) ‘Joko Widodo : A New Paradigm for Indonesian Leadership ?’, A USINDO

157

Special Open Forum with Adam Schwarz Co-Founding Partner and Chief Executive Officer-Asia , Adam Schwarz ’ s RemarksExecutive Officer-Asia , Adam Schwarz ’ s Remarks, February,(The United States – Indonesia Society usindo.org).

Starik, M. and Kanashiro, P. (2013) ‘Toward a Theory of Sustainability Management: Uncovering and Integrating the Nearly Obvious’, Organization and Environment, 26(1), pp. 7–30. doi: 10.1177/1086026612474958.

Tarigan, J. and Semuel, H. (2015) ‘Pengungkapan Sustainability Report dan Kinerja Keuangan’, Jurnal Akuntansi dan Keuangan, 16(2), pp. 88–101. doi: 10.9744/jak.16.2.88-101.

Taylo, A. T. C. S. L. (1992) ‘Discretionary Disclosure of Reserves By Oil and Gas Companies : : An Economic Analysis’, ournal of Busincss Finance d Accounfing, 19(January), pp. 295–308.

Tristiarini1, N., Utomo1, S. D. and Setiawanta1, Y. (2019) ‘The Capability of Risk as a Corporate Reputation Driver to increase Market Value’, QUALITY Access to Success, 20(168), p. 54.

Ullmann, A. A. (1985) ‘Data in search of a Theory : Search RelationshIps of the A Critical Examination Social Among Performance Disclosure’, The Academy of Management Review, 10(3), pp. 540–557.

Zain, M., Mustaffa and Tamoi, J. (2006) ‘Corporate Social Disclosure (CSD) of construction companies in Malaysia’, Malaysian Accounting Review, 5(1), pp. 85–114. Available at: http://ir.uitm.edu.my/id/eprint/13791.

Contact Address

Yulita Setiawanta, SE,.M.Si,.Ak,.CA,.CPA,.CSRA.,CSP. Diponegoro University, Semarang, Indonesia Dian Nuswantoro University, Semarang, Indonesia Jl. Nakula I No. 5_11 Semarang 50131, C Building Seconds Floors Email: [email protected] Phone number: +62243567010

Dr. Agus Purwanto,. M.Si,.Akt,.CA. Diponegoro University, Semarang, Indonesia Jl. Prof Sudartho,. SH, Tembalang, Semarang 50275, Indonesia Email: [email protected]


158

FACTORS AFFECTING CHOICE OF EMPLOYMENT BY UNIVERSITY GRADUATES

Markéta Šnýdrová, Lucie Vnoučková, Ivana Šnýdrová

Abstract: The article focuses on factors influencing employer selection from the perspective of graduates of the University of Economics and Management. The aim of the article is to identify factors affecting attractiveness of organization and to determine the impact of these factors on the choice of employer in terms of the different focus of the applicants. Primary research included a questionnaire (n = 238 respondents), 4 focus groups and two individual interviews. The results of primary research have shown that the most important areas are: a prestigious and respected company, a purely Czech company, a stable established company with tradition, TOP employer, a socially responsible company, possibility of career growth, possibility of further development, the opportunity to participate in corporate goals, participation in decision making and management of the company, the possibility to apply creativity, innovation and ideas, individual work, interesting work, challenge work, ambitious project, financial remuneration, corporate benefits and geographic availability of the organization. The research also showed that preferences can be divided into factors that indicate whether an organization becomes a selected (attractive) employer based on the candidate's specific focus. Factor analysis has shown that different factors are important for identified types of candidates. The output of the article is a recommendation on what factors organization should focus on to address a particular type of candidate.

Keywords: Employer Attractiveness, Factors Influencing Employer Selection, Organization, Applicant, Employee, University Graduate.

JEL Classification J20,M51.

Introduction

The article deals with factors affecting employer selection from the perspective of graduates of the University of Economics and Management.

The data of the Czech Statistical Office show that the general unemployment rate in the Czech Republic is very low (CZSO, 2019), thus organizations must increasingly strive to become an attractive employer, attract the interest of potential job seekers while retaining existing employees (Ungerman, Myslivcova, 2018; Punjairisi, Wilson and Evanschitzky, 2008). Working with human resources is a critical factor in the success of an organization (Novotný, Mikulecký, 2011; Edwards, Edwards, 2013; Martin, Gollan, Grigg, 2011). For this reason, the importance of scientific discipline known as Personnel Marketing (Ungerman, Myslivcová, 2018) is currently growing. It is a combination of classical marketing and personnel management, or the application of marketing elements to individual personnel activities. Ungerman and Myslivcova (2018) emphasize that this approach focuses on employment as a customer, his expectations and the satisfaction of these expectations, needs and ideas from the employer. The difficult labor market situation forces organizations to respond to labor market demands, to compare themselves with competitors in response to the identified

159

needs of potential job seekers and existing employees. Thus, the employer has to build his employer brand. An organization that offers employees better benefits, greater flexibility in terms of job and time, development or career opportunities, diversity and other opportunities will ensure a competitive position. Applicants choose employers in their essence according to their ability to respond to factors that affect employer selection. The organization is thus directly dependent on whether these factors can be identified and adapted to their offer. It should also be mentioned that each candidate is different and hence his needs may differ from others. For this reason, the situation of organizations is very difficult, because what may be very important for one candidate is irrelevant to another.

The aim of the article is to identify factors affecting the attractiveness of the organization and to determine the extent of influence of these factors on the choice of employer in terms of the different focus of the applicants.

Within the introduction and the first chapter “Statement of a problem”, the researched issue is elaborated in a broader framework. Current articles and publications are used to describe the issue. The methods show how to obtain primary data and how to process it. The chapter "Problem Solving" summarizes the primary data obtained through the questionnaire, focus groups and individual interviews. Data is further investigated and analyzed. In the discussion and conclusion of the article, the key findings from primary research are summarized and compared with the already conducted research within the researched issue.


The labor market situation shows that employers, if they want to succeed, are forced to build their employer brand attractiveness (Edwards, Edwards, 2013; Martin, Gollan, Grigg, 2011).

1.1 Factors affecting employer selection

There are number of factors influencing employer selection and it is important to organizations be aware of them and focus on work in compliance with them (Twenge et al., 2010). If an organization has difficulty retaining existing workers and successfully reaching potential job seekers, it should analyze these factors (Armstrong, 2007, p.347).

Love and Singh (2011) point out that the attractiveness of an organization for potential candidates as well as for existing employees increases when the organization is directly ranked among the top employers within the area as part of the employer assessment.

The employer's attractiveness is closely linked to its reputation. If the organization's reputation is good, jobseekers offer themselves and existing employees are proud to work for such an attractive organization (Armstrong, 2007, p. 347; Horváthová, Bláha, Čopíková, 2016, p. 39; Vlková , Urbancova, 2017). Employee pride of employer prestige also has a positive impact on his work performance and behavior (Wilden, Gudergan, Lings, 2010).

Applicants consider employer quality by working conditions, salary and benefit levels (Armstrong, 2007, p.347, Maxwell, Knox, 2009; Urbancová, Šnýdrová, 2017;

160

Kumar, 2018). In their research, Vlková and Urbancová (2017) also confirm that the most important factors influencing employer selection include the amount of wages. Daly (2011) states that wages are an indicator of employees' willingness to invest in employees, coupled with a wide range of benefits. Vimrová (2016) then deals with the importance of providing various employee benefits. However, Vnoučková and Čejka (2013) emphasize that the preference of potential applicants is currently changing and shifting to free time.

Vlková and Urbancová (2017) further add that the scope of the organization close to the residence is also a significant factor. The importance of the organization's local operations is also mentioned in Rampl (2014).

Other factors influencing the employer's attractiveness include education and development opportunities (Dabirian, Kietzmann and Diba, 2017; Wilden et al., 2010). Age-centered and employer-driven research shows that job seekers, young people, employees, and graduates put emphasis on work, opportunity for education and development, a positive workplace atmosphere, while older workers emphasize job security (Šnýdrová et al., 2014; Malik, Subrabamanian, 2015). Also Aziri (2011) and Rampl (2014) mention the importance of the nature and content of the work. Sibson (2011) emphasizes the importance of making work interesting and delighting. Chalofsky and Krishna (2009) stress that it is essential that the employee or the candidate perceive the work as meaningful, which is then the strongest internal motivator.

Another important factor affecting attractiveness of the organization for candidates and existing employees is the possibility of career advancement within the organization (Wilden et al., 2010; Sibson, 2011; Snydrova, 2014; Dabirian, Kietzmann and Diba, 2017). Tüzünger and Yuksel (2009) added that the existence of an international career option is an important factor. Doherty, Richardson, and Thorn (2013) directly point out that the possibility of leaving the country and working abroad is an important factor for employer selection for a particular group of workers.

A significant influence on the attractiveness of an organization as an employer is when an organization is committed to social responsibility. Corporate social responsibility affects both the attitude and commitment of existing employees and attracts potential candidates (Jones, Willness, MacNeil, 2009; Rupp et al., 2013, Jones, Willness, Madey, 2014; Donia, Tetrault Sirsly, 2016).

Recently, job seekers have been given the opportunity to apply their own ideas and innovations to the organization (Berthon et al, 2005; Tuzunger and Yuksel, 2009), leading to a subsequent increase in employee engagement and willingness to participate in decision making (Berthon et al., 2005; Jones, Willness, MacNeil, 2009; Graeme, Gollan, Grigg, 2011; Rupp et al., 2013, Jones, Willness, Madey, 2014; Donia, Tetrault Sirsly, 2016).

Mondy, Noe and Premeaux (2002) add that flexibility is an essential factor in attracting an organization as an employer.

Tüzünger and Yuksel (2009) summarize almost all of these factors mentioned that the expectations and requirements of potential candidates for a selected employer of "employer of first choice" include both corporate data (age and location of the company) and components increasing the attractiveness of the employer (work tasks

161

that are challenging), innovation, competitive working environment, work-life balance, career development, international career opportunities, education and development, time flexibility, marketing success and organization reputation; as confirmed by Minchington (2010) and Sutherland, Torricelli and Karg ( 2002).

Based on the theoretical overview, the main factors influencing the selection of the employer were identified, including: the prestige of the organization, stability and tradition of an organization, an organization known as a top employer, a socially responsible organization, the possibility to work abroad, the possibility of a career growth, the possibility of further development, the opportunity to participate in company goals, the possibility of participation in decision-making, the possibility to apply ideas - creativity, independent work, interesting work, work that is a challenge, the amount of wages / salaries, corporate benefits, geographical proximity to the organization - easy availability.

These factors will be subject to factor analysis.

2 Methods

The data for this article were obtained as part of the content analysis of primary and secondary literary sources published in relevant databases, such as Web of Science and primary quantitative research, which focused on graduates of selected business university.

Quantitative research

Quantitative research included a questionnaire survey (238 respondents). The questionnaire survey was carried out using the CAWI method, i.e. using an electronic on-line questionnaire (Computer Assisted Web Interviewing). The questionnaire was placed on Google and respondents were approached by e-mail. By filling in the questionnaire via the link, the absolute anonymity of the respondents was ensured (the respondent did not have to provide information that could identify him / her before filling out the questionnaire) when completing the questionnaire. All graduates of the University of Economics and Management, who completed their studies in 2012-2017, were approached. In total, 870 graduates were addressed in the survey, with a questionnaire return of 283 questionnaires (i.e. 27.4% of the total number of questionnaire applications). Data collection took place from May 2017 to November 2017). The questionnaire contained a total of 31 questions (24 questions with answer options, 6 open questions and 1 question using the scale in the range from 1 - absolutely agree to 5 - absolutely disagree). The questionnaire focused on the factors that influence employer choice and included socio-demographic questions (gender, age, type of studied program). All questions enabled respondents to follow up their own answers or comments.

The obtained data were then statistically processed using software (Excel and SPSS statistics).

As stated above, a total of 238 respondents participated in the questionnaire, of which 157 were women (66%) and 81 (34%) men. This percentage is equal to the division of gender of students and graduates of the university under investigation. In terms of educational attainment, the distribution of respondents is as follows: 122

162

graduates had a bachelor's degree (51.3%), 70 graduates had a master's degree (29.4%), 32 respondents have achieved both master's and bachelor's degrees at the University (13.5%), 12 graduated in MBA program (5%), 1 respondent completed 3 month study module program (0.4%) and 1 respondent, did not complete the form of education (0.4%).

Data Processing

The first stage of processing the questionnaire data focused on the preparation of a data matrix. The data was then sorted and then it was coded and sorted according to the type of variables, qualitative or quantitative. During this phase, the data was also cleaned, and its quality was checked in order to uncover any extreme (eccentric) or deviating observations which could significantly influence the results of analyses. There were no missing values because all questions were mandatory. The last part of the data matrix involved the transformation of the variables which was necessary for several reasons. When processing a multidimensional data file, the reason for this is usually the requirement for the fulfilment of the analyses of a certain statistical method. The process of calculation and interpretation of results was used according to Hebák et al. (2014). The basic conditions of attributes to enter the analysis were fulfilled according to Hendl (2006).

The processing of the results was based on analysis of the data focused on investigating the important properties and the typical features of the statistical file. The statistical evaluation of the data was undertaken firstly by a one-dimensional analysis based on the frequency distribution, the calculation of point and interval estimates and the testing of hypotheses about the frequency of the categories of individual variable values. Secondly, a two-dimensional analysis was used based on an investigation into the dependence of two selected variables. The goal of the comprehensive analysis of several variables was to uncover any relations between data structures and to find an interpretation for these structures. The retrieval of the information in the data file was realized using the classic Pearson hypothesis testing. Pearson´s and Spearman´s correlation were applied. If the p–value calculated by means of the test was lower than the selected level of significance α = 0.05. Only resultant values of the correlation coefficient with an absolute value of more than 0.3 were selected as being significant.

Factor analysis was used to evaluate the results and specifically to identify groups of responses of students regarding their perceptions. The goal was to find groups of variables with significant appearance and consistent content and at the same time to reveal main orientation of coherent groups of graduates. The results of analyses and formed factors may help to reveal the current desired areas of orientation on labour market and preferences of graduates. The higher level of generalization of results by factor analysis helps to focus on the most important and highly recommended areas with filtering out inconsistencies, which may be studied separately as outstanding values which in turn, may also be inspiring for development.

The method of analysis used in this paper is often used in social sciences studies (Palát, 2012). Also in the area of learning and development research, this method is used quite often and favoured by researchers (Anderson, 2009). The levels of correlation coefficients were sufficient according to Anderson (2009) and Hendl (2006) to enter factor analysis. Moreover, over 85% of correlations in the correlation

163

table were statistically significant. The KMO (Kaiser-Meyer-Olkin test) value reached 0.7 which is considered as meritorious and thus adequate for factor analysis.

The number of monitored variables (factors) was reduced using the Varimax method. For the selection of substantial factors the Kaiser-Guttman rule was applied (i.e. substantial factors having a value within the range higher than 1) and subsequently Sutin test was applied. The correlation coefficients are in the interval from <-1; 1>. If the correlation coefficient is positive, it is a direct proportion (negative – indirect proportion). For the evaluation, the value of variable correlation higher than 0.3 (moderate correlation) according to Anderson (2009) was used. To evaluate the results, SPSS statistics was used. All procedures were followed in accordance with the ethical standards and Czech law relating to the use of sensitive information.

3 Problem solving

The results of primary research have shown that a significant areas affecting graduates´ choice of preferred employer are: a prestigious and respected company, a purely Czech company, a stable established company with traditions, a TOP employer, a possibility to be part of decision-making and management, the possibility to apply creativity, innovation, resp. new ideas, individual work, meaningful work, challengeable work, new ambitious project, above-standard financial proposal, corporate benefits, and geographical availability of the organization.

The obtained data were tested by multidimensional statistics, factor analysis described in Methods. The results identified 7 significant factors meeting the criteria identified down in the methodology. Tab. 1 shows the significance of the individual examined factors describing expected outcomes of students’ university studies. In total, the two identified variables explain 61% of the sample behaviour or of the possible resultant characteristics, with the specific values of factor analysis.

164

Tab. 1: Factors affecting choice of employer Factor 1 2 3 4 5 6 7

Prestigious and respected company 0.109 0.075 -0.013 -0.161 0.727 0.266 -0.111

A purely Czech company 0.067 -0.180 0.121 -0.043 0.013 0.080 0.645

Established company with tradition 0.031 -0.139 0.021 0.197 0.569 0.007 0.241

TOP employer -0.052 0.186 0.255 0.111 0.653 -0.138 -0.092

Socially responsible company 0.284 0.072 0.133 0.129 0.209 0.036 0.375

Possibility to work abroad 0.184 0.678 0.155 -0.116 0.026 -0.046 0.062

Possibility of career growth -0.039 0.096 0.070 0.100 0.122 0.840 0.061

Possibility of further development 0.239 -0.025 0.202 0.141 0.036 0.730 -0.019

Possibility to participate in company goals

0.687 0.099 0.173 -0.051 0.018 0.167 0.147

Participation in decision making 0.544 0.261 -0.274 -0.080 -0.015 0.178 0.060

Possibility to apply new ideas 0.684 -0.044 0.030 0.139 -0.024 0.132 0.026

Independent work 0.103 -0.015 -0.119 0.723 0.169 0.017 0.115

Work that I like 0.003 -0.010 0.274 0.635 -0.172 0.097 0.129

Work that is a challenge 0.352 0.215 -0.161 0.344 0.103 0.118 0.223

New ambitious project 0.571 0.068 -0.084 0.105 0.057 -0.180 -0.080

Financial rewards -0.289 0.202 0.516 0.211 0.010 0.218 -0.001

Remuneration benefits 0.029 0.076 0.643 0.043 0.265 0.163 0.181

The proximity of company 0.030 0.032 0.724 0.055 -0.042 0.037 -0.041

Name of factor Leader Expat Certaint

y Independence

Prestige Develop

ment Patriot

Source: Authors processing

The factor analysis grouped responses of surveyed graduates from selected university into seven main factors. Respondents are seeking their preferred employer in different ways, which are described by the factors listed in Tab. 1. They are either interested in being part of company decision making process, creating goals, experience challenging work and projects, and development and thus being in a leadership position (factor 1; Leader), or they are interested mainly in the possibility to work abroad (factor 2; Expat). Thirdly, group of respondents orient themselves in searching for employment mainly based on high of offered salary and benefits and geographical proximity and accessibility of job position or company. Those job applicants does not any other specifics than remuneration and thus security (factor 3; Certainty). Fourthly, group of graduates orient themselves based on their personal preferences and independence of their work – they need to like the work or job (factor 4; Independence). Fifthly, graduates are looking for sound name of their employer and wishing to work for stable, long established company with tradition, they prefer if their employer is listed within TOP companies or top employers and the name and brand is well recognized. Those job applicants expects the best from the brand ranking and looking for prestigious company (factor 5; Prestige). Sixth way of graduates’ orientation is their search for employer where they can be further developed and where they expect or have possibility for career growth (factor 6; Development). Finally, respondents when looking for preferred employer are looking for typical Czech/national company; together with socially responsible orientation of an company. Those job applicants can be thus attracted to ads focusing on development of values in the national society (factor 7; Patriot).

165

It is possible to summarize that respondent graduates are not focusing only on remuneration or top employers but also based on the possibility to be part of the decision-making process in a company, or to have a possibility for development, to work abroad or to work independently. Those are the main ways how companies can promote their job positions focusing on graduates.

4 Discussion

The attractiveness of employer is a key issue for organizations, as its efficiency is affecting further functioning of the organization because of its ability to reach and attract suitable candidates and transform them into employees. The primary data collection demonstrated the significance of all factors examined in accordance with the authors' findings in the theoretical basis of the article (e.g. Armstrong, 2007, p.347, Maxwell, Knox, 2009; Tüzünger and Yuksel, 2009; Donia, Tetrault Sirsly, 2016; Snydrova, 2017; Kumar, 2018). However, the subsequent factor analysis identified the main groups of factors with regard to the different focus of the candidates: Leader, Expat, Certainty, Independence, Prestige, Development and Patriot. Within the framework of the factor analysis, it was found that the most important factor for a group of factors called Expat is the possibility to work outside the country. This is in line with what Doherty, Richardson and Thorn (2013) who say that the most crucial aspect of employer selection is to gain experience and career opportunities for a particular type of jobseeker. Ceric and Crawford (2016) confirm the same. Security is another important factor that plays a role in finding an employer, job seekers expect the employer to provide them with the appropriate financial background and benefits, and to be within driving distance, but at the same time being able to pay them wages and not to be giving them notice without logical reason (Malik, Subramanian, 2012); this is often the primary focus for older age groups (Freese, 2007) while the other identified main factor (within group of factors) named Development is mainly preferred by younger applicants (Sibson, 20011; Šnýdrová, 2014). Xie, Bagozzi and Meland (2015) report that employer's reputation and prestige play a significant role in the recruitment process. This is in line with the results of primary research, which showed that the identified main factor Prestige focuses primarily on the employer's reputation, and employer placement in various polls.

A group of factors known as the Leader hides expectations, especially for young, ambitious candidates from the Millennium generation, i.e. the possibility of cretinality, decision-making, which is consistent with Berthon et al. (2005), Jones, Willness, MacNeil (2009), Graeme, Gollan, Grigg (2011), Rupp et al. (2013), Jones, Willness, Madey (2014), Donia, Tetrault Sirsly (2016). Another major factor has been identified as Independence, which primarily associates expectations of interest and labor intensity. This finding is in line with Aziri's research, which highlights the importance of work. The last factor in Patriot emphasizes the relationship between employment in the Czech organization and social responsibility. This area has not been adequately described in the literature yet and is a new approach that can be addressed both in terms of research and theory, but also in the practical focus of employers on this group of job seekers.

166

Conclusion

The article focused on analysis of factors influencing the choice of employer from the perspective of graduates of the University of Economics and Management. The research included graduates who completed the educational process in 2012-2017 and already have experience with finding an employer and selection process.

In line with research by Rampl (2014), the results show that if an organization wants to become competitive on the market and an attractive employer for potential candidates and existing employees, it is necessary to respond to their demands and expectations. The results of primary research show that all factors mentioned in the theoretical part of the article are important in terms of employer attractiveness. However, the significance of individual factors varies according to the applicant's focus. Thus it can be stated that if organizations want to become a preferred employer in general, they should focus on the above-mentioned factors (prestige of the organization, its stability, placement in Top Employer research, possibility to work abroad or possibility to work in a purely Czech company, possibility of career growth, possibility of further development, opportunity to participate in corporate goals, opportunity to participate in decision-making and leadership, opportunity to apply ideas, work autonomy, work environment, work as a challenge, opportunity to work in an ambitious project, wages structure, and proximity of organization.

Based on the factor analysis, 7 groups of factors were identified: Leader, Expat, Security, Independence, Prestige, Development and Patriot. It can be stated that individual groups indicate different types of applicants / employees according to their preference. A group described by factor labelled Leader refers to candidates preferring to participate in decision-making and leadership in organisation and seeking the opportunity to apply their creativity. Those applicants are also interested in participating in projects, and preferring a job that is challenging them. The factor named Expat is a candidate who prefers the opportunity to work abroad. While the Patriot factor refers to candidates preferring to work in a purely Czech company that is also socially responsible. For candidates choosing a group of factors labelled Independence the interest in work tasks and self-reliance is the most important of all respondents within this group. For candidates choosing a group of factors labelled Certainty is the amount of wage / salary, the broadness of benefits and the geographical distance of the company. Factor named Development describes a group of candidates who is looking especially for career growth and the possibility of further development. Finally, a group characterized by factor called Prestige describes type of job seeker expecting an employer to be a prestigious and respected company, a stable established company with a tradition or an organization ranked high in Top Employer ranking.

However, if an organization wants to focus on candidates preferring only one of the identified groups of factors, it is only possible to focus on those areas that have been identified as significant within the group of applicants or employees which the organization prefers.

The limit of this article is that research contained only graduates of one studies business university. The limiting factor may also be the fact that it is a university of economic focus and therefore the preferences of its graduates may be different from

167

the preferences of graduates of technical, humanities and other fields. The final limit of the article may also be the fact that the completion of the questionnaire was voluntary and therefore the results do not include the opinions of graduates who refused to participate in the research. Despite these limits, it can be assumed that the results of the research carried out can be considered as significant, as the number of respondents is representative sample with respect to the total number of graduates examined within the university for the given period.

Further research can be extended to other universities and also the difference in the preferences of factors important for the selection of the employer according to the specialization of the graduates or the gender and age of the job seekers can be compared.

Acknowledgement

This contribution was supported by project of Centre of Economic Studies of University of Economics and Management, grant no. GCES1117.

References

Anderson, V. (2009). Research Methods in Human Resource Management. London: Chartered Institute of Personnel Development.

Armstrong, M. (2007). Řízení lidských zdrojů: nejnovější trendy a postupy: 10. vydání. Praha: Grada.

Aziri, B. (2011). Job satisfaction. A Literature Review. Management Research and Practice, 3 (4), pp. 77–86.

Berthon, P., Ewing, M., Hah, L.L. (2005). Captivating company: dimensions of attractiveness in employer branding, International Journal of Advertising, 24 (2), pp. 151-172.

Ceric, A., Crawford, H. J. (2016). Attracting SIEs: Influence of SIE motivation on their location and employer decisions. Human Resource management, 26, pp. 136-148. DOI: 10.1016/j.hrmr.2015.10.001.

ČSÚ, Statistická data o zaměstnanosti. (2019). [online]. www.czso.cz. Availeble at URL https://www.czso.cz/csu/czso/cri/zamestnanost-a-nezamestnanost-podle-vysledku-vsps-4-ctvrtleti-2018. [12.02.2019].

Dabirian, A., Kietzmann, J., Diba, H. (2017). A great place to work!? Understanding crowdsourced employer branding. Business Hori-zons[online]. [cit. 2018-03-13]. Dostupné z: https://www-sciencedirect-com.proxy.k.utb.cz/science/article/pii/S0007681316301276?via%3Dihub

Daly, J. L . (2012). Human Resource management in the Public Sector. Policies and Practices. 1st edition.New York: M.E. Sharpe. Kumar, S. Dimensions of employee turnover factors in software companies in India. (2018). CLEAR International journal of research in commerce and management, 9 (4), pp.29 -33.

Doherty, N., Richardson, J., Thorn, K. (2013). Self-initiated expatriation and self-initiated expatriates. Career Development International, 18 (1), pp. 97–112. DOI: 10.1108/13620431311305971.

Donia, M. B. L., Tetrault Sirsly, C. A. (2016). Determinants and consequences of employee attributions of corporate social responsibility as substantive or symbolic. European Management Journal, 34 (3), pp. 232-242. DOI: 10.1016/j.emj.2016.02.004.

Edwards, M. R., Edwards, T. (2013). Employee responses to changing aspects of the employer brand following a multinational acquisition: a longitudinal study. Human Resource Management, Vol. 52 (1), pp. 27–54. DOI: 10.1002/hrm.21519.

Freese, C. (2007). Organizational Change and the Dynamics of Psychological Contracts: A longitudinal study. Ridderkerk: Ridderprint Offsetdrukkerij B.V.

168

Graeme, M., Gollan, P.J., Grigg, K. (2011). Is there a bigger and better future for employer branding? Facing up to innovation, corporate reputations and wicked problems in SHRM. International Journal of Human Resource Management, 22 (17), pp.3618-3637. DOI: 10.1080/09585192.2011.560880.

Hebák, P. et al. (2014). Statistické myšlení a nástroje analýzy dat. Praha: Informatorium.

Hendl, J. (2006). Přehled statistických metod zpracování dat: analýza a metaanalýza dat. Praha: Portál.

Horváthová, P., Bláha, J., Čopíková, A. (2016). Řízení lidských zdrojů: nové trendy. Praha: Management Press.

Chalofsky, N., Krishna, V. (2009).Meaningfulness, commitment, and engagement: Theintersection of a deeper level of intrinsic motivation. Advances in Developing Human Resources, 11 (2), 189–203. DOI: 10. 1177/1523422309333147.

Jones, D. A., Willness, C. R., MacNeil, S. (2009). Corporate social responsibility and recruitment: testing person-organization fit and signaling mechanisms. In G. T. Solomon (Ed.), Best paper proceedings of the 69th annual meeting of the academy of management, Chicago, IL. DOI: 10.5465/AMBPP.2009.44265576.

Jones, D. A., Willness, C. R., Madey, S. (2014). Why are job seekers attracted by corporate social performance? experimental and field tests of three signalbased mechanisms. Academy of Management Journal, 57, pp. 383-04. DOI:10.5465/amj.2011.0848.

Love, L. F., Singh, P. (2011). Workplace Branding: Leveraging Human Resources Management Practices for competitive Advantage Through Best Employer Surveys. Journal of Business and Psychology, 26 (2), pp.175–181. DOI: 10.1007/s10869-011-9226-5.

Malik, S., Subramanian, V. (2015). Job Expectations and Perception Variations across Career Stages:

An Empirical Study in Indian Tractor Industry. Paradigm, 19 (2), pp. 212–231. DOI: 10.1177/0971890715610017.

Maxwell, R., Knox, S. (2009). Motivating employees to “live the brand”: A comparative case study of employer brand attractiveness within the firm. Journal of Marketing, 25 (9-10), pp.893-907. DOI: 10.1362/026725709X479282.

Martin, G., Gollan, P. J., Grigg, K. (2011). Is there a bigger and better future for employer branding? Facing up to innovation,. corporate reputations and wicked problems in SHRM. International Journal of Human Resource Management, 22 (17), pp.3618–3637. DOI: 10.1080/09585192.2011.560880.

Minchington, B. (2010). The employer brand manager’s handbook.Torrensville: Collective Learning.

Mondy W. R., Noe R.M., and Premeaux S. R. (2002). Human Resource Management, 8 /Ed: Upper Saddle River, New Jersey: Prentice Hall

Novotný, J., Mikulecký, P. (2011). Znalostní management a jeho uplatnění v menších podnicích. Scientific papers of the university of Pardubice, Vol. 10(2), pp. 102–113.

Palát, M. (2012). Statistické zpracování dat. [online]. 2012 [cit. 2017-03-03]. Available at: http://www.vuchs.cz/OPVpK/dokumenty/Palat-Statistika-2.pdf

Punjaisri, K., Wilson, A., Evanschitzky, H., 2008. Exploring the influences of internal branding on employees´ brand promise delivery: Implications for strengthening customer-brand relationships. Journal of Relationship Marketing, 7 (4), pp. 407-424.

Rupp, D. E. et al. (2013). Applicants' and employees' reactions to corporate social responsibility: the moderating effects of first-party justice perceptions and moral identity. Personnel Psychology, 66, pp. 895-933. DOI: 10.1111/peps.12030.

Rampl, V. L. (2014). How to become an employer of choice:transforming employer brand associations into employer first-choice brands. Journal of Marketing Management, 30 (13–14), pp.1486–1504. DOI: 10.1080/0267257X.2014.934923.

Sutherland, M., Torricelli, D. G., Karg, R. (2002). Employer of choice branding for knowledge workers. South African Journal of Business Management, 33 (4), pp. 13–20.

169

Sibson, R. (2011).Career choice perceptions of undergraduate event, sport and recreationmanagement students: An Australian case study. Journal of Hospitality, Leisure, Sport& Tourism Education, 10 (2), pp. 50–60.

Šnýdrová, M. (2014). Změny struktury pracovního trhu v souvisosti s nástupem generace Y a dopady těchto změn na získávání zaměstnanců. Ekonomické Listy, 5 (3), pp.18-34.

Twenge, J. M. et al(2010). Generational differences in work values: Leisure and extrinsic values increasing, social and intrinsic values decreasing. Journal of Management, 36, pp.1117–1142. DOI: 10.1177/0149206309352246.

Tüzünger V. L., Yuksel, C. A. (2009). Segmenting potential employees according to firms’ employer attractiveness dimensions in the employer branding concept. Journal of Academic Research in Economics, 1 (1), pp.46–61.

Ungerman, O., Myslivcová , S. (2018). Current use of personnel marketing in Czech companies. Journal of Interdisciplinary Research, 8 (1), pp. 232-239.

Urbancová, H. Šnýdrová, M. Remuneration and Employee benefits in Organizations in the Czech Republic. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 65 (1), pp 357-368.

Vimrová, H. (2016). Relevance vykazování zaměstnaneckých požitků v souladu s ifrs v české podnikové praxi. Acta Oeconomica Pragensia, 24 (6), pp. 38-55.

Vlková, M., Urbancová, H. (2017). Proces získávání pracovníků s důrazem na externí personální marketing: Případová studie. Ekonomické Listy, 8 (6), pp.17-30.

Vnoučková, L., Čejka, R. (2013). Důsledky mobility zaměstnanců. Trendy ekonomie a managementu, 7 (15), pp. 92-104.

Wilden, R., Gudergan, S., Lings, I. (2010). Employer branding: Strategic implications for staff recruitment. Journal of Marketing Management, 26 (1-2), pp. 56-73. DOI: 10.1080/02672570903577091.

Xie, C., Bagozzi, R. P., Meland, K. V. (2015). The impact of reputation and identity congruence on employer brand attractiveness. Marketing Intelligence and Planning, 33 (2), pp 124-146. DOI:10.1108/mip-03-2014-0051.

Contact Address

PhDr. Markéta Šnýdrová, PhD. University of Economics and Management, Nárožní 2600/9a, Praha 5, 158 00, Czech Republic Email: [email protected]

doc. Ing. Lucie Vnoučková, PhD. University of Economics and Management, Nárožní 2600/9a, Praha 5, 158 00, Czech Republic Email: [email protected]

PhDr. Ivana Šnýdrová, CSc. University of Economics and Management, Nárožní 2600/9a, Praha 5, 158 00, Czech Republic Email: [email protected]


170

DOPADY SYSTÉMU OBCHODOVÁNÍ S EMISNÍMI POVOLENKAMI V ZEMÍCH EU28

IMPACTS OF EMISSION TRADING SYSTEM IN EU28 COUNTRIES

Jarmila Zimmermannová, Vít Pászto, Ondřej Vícha

Abstract: European Union Emission Trading Scheme (EU ETS) was introduced in EU countries in the year 2005. Currently, the EU ETS is in operation for more than a decade; moreover, the European Commission adopted rules for the next 4th trading period in March 2018. The main goal of this article is to evaluate possible environmental impacts of EU ETS, mainly the impact of emission allowance price on the amount of greenhouse gas emissions in the EU28 countries as a whole. For this goal, (geo)visual analysis of spatial data, correlation and regression analysis are used. The results of the analysis show that the EU ETS contributes to the reduction of the amount of greenhouse gas emissions in the EU-28 countries as a whole, due to emission allowances price and period. Nevertheless, the influence of the way of emission allowances distribution is crucial for the desirable environmental impact. The extension of auctioning represents our key recommendation for the EU ETS.

Keywords: EU ETS, Environmental Impact, Carbon Emissions, Economic Indicators, Geovisual Analysis.

JEL Classification: H23, K32, Q58, R12.

Úvod

Evropský systém obchodování s povolenkami na emise skleníkových plynů (European Union Emissions Trading Scheme, dále jen „EU ETS“) je hlavním nástrojem k plnění cílů EU v oblasti ochrany klimatu s cílem podpořit snižování emisí skleníkových plynů nákladově efektivním a ekonomicky účinným způsobem.

Systém EU ETS byl zřízen směrnicí Evropského parlamentu a Rady 2003/87/ES ze dne 13. října 2003 o vytvoření systému pro obchodování s povolenkami na emise skleníkových plynů ve Společenství a o změně směrnice Rady 96/61/ES. V současnosti se jedná o největší obchodní systém pro emise skleníkových plynů na světě. Setkáme se rovněž s označením systému jako „cap-and-trade“, který obecně funguje tak, že stanoví stropy pro celkové emise z odvětví, která produkují velké objemy emisí. V systému EU ETS strop omezuje emise více než 11 000 energeticky náročných podniků v celé EU a zahrnuje zhruba polovinu všech emisí skleníkových plynů. Na evropské úrovni není cílem systému EU ETS pouze snížit emise skleníkových plynů na stanovený strop, ale také nastavit tržní cenu uhlíku a finančně podpořit každou tunu ušetřeného CO2ekv. Producenti emisí by tak měli být motivováni k uplatňování nákladově efektivních opatření na snížení emisí, přičemž tržní cena povolenky by měla podporovat investice do nízkouhlíkových technologií.

Klíčovými otázkami v souvislosti se systémem EU ETS a zejména jeho účinností jsou: „Jaký je ve skutečnosti dopad celého systému? Má EU ETS opravdu vliv na emise CO2?“ Na uvedené otázky odpovídá tento článek.

171

Hlavním cílem článku je provést vyhodnocení vlivu systému emisních obchodovatelných povolenek EU ETS na množství emisí skleníkových plynů v zemích EU28. Nejprve bude představen systém EU ETS a jeho specifika, včetně legislativního vývoje a vědeckých studií. Jelikož je EU ETS nástroj vztažený ke konkrétnímu území, tj. ohraničuje prostor pro tzv. cap, bude následně provedena geovizuální analýza s využitím dat Eurostatu. Klíčovou částí je provedení regresní analýzy, která nám umožní nastínit možné souvislosti mezi vývojem emisí CO2 a vybranými indikátory. Zjištěné výsledky budou diskutovány s ohledem na širší souvislosti a aktuální legislativní změny v oblasti EU ETS.

1 Formulace problematiky

1.1 Historie a současnost systému EU ETS

Systém EU ETS se vyvíjel postupně. První fáze (2005–2007) byla tříletým pilotním obdobím, které mělo připravit druhou fázi (2008–2012). Emisní povolenky byly poprvé rozděleny prostřednictvím národních alokačních plánů na základě historických emisí. Povolenky z prvního období nebyly převoditelné do druhé fáze. Cílem první fáze bylo zřídit fungující trh, určit tržní cenu uhlíku a vybudovat nezbytnou infrastrukturu pro monitorování, vykazování a ověřování skutečných emisí. Údaje generované z první fáze následně zaplnily informační mezeru a pomohly stanovit vnitrostátní stropy pro druhou fázi (European Comission, 2015).

Druhá fáze (2008–2012) byla souběžná s obdobím Kjótského protokolu. Evropská unie se zavázala v období 2008–2012 dosáhnout celkového snížení emisí CO2 o 8 % oproti úrovním v roce 1990. Tuto cílovou hodnotu sdílelo 15 členských států EU na základě právně závazné dohody – rozhodnutí Rady 2002/358/ES ze dne 25. dubna 2002 o schválení Kjótského protokolu k Rámcové úmluvě OSN o změně klimatu jménem Evropského společenství a o společném plnění závazků z něj vyplývajících. Na základě ověřených emisí vykázaných v první fázi byl objem emisních povolenek povolených ve druhé fázi snížen o 6,5 % ve srovnání s úrovní roku 2005. Z důvodu ekonomické krize ovšem došlo k významnému přebytku přidělených povolenek, které byly převedeny do třetího obchodovacího období (2013-2020) (European Comission, 2015).

Ve třetí fázi (2013–2020) se podmínky fungování EU ETS zásadně změnily v návaznosti na změnu směrnice 2003/87/ES provedenou v rámci tzv. klimaticko-energetického balíčku směrnicí 2009/29/ES. Evropská komise též představila cíle, kterých chce dosáhnout EU v oblasti energetiky a změn klimatu do roku 2020 (známé jako cíle „20-20-20“). Jedním z těchto cílů bylo i snížení emise skleníkových plynů EU o 20 % v porovnání s úrovní roku 1990. Revizí systému EU ETS došlo k nahrazení systému národních alokačních plánů, strop pro emise je nyní stanoven na úrovni EU. Dražba byla stanovena jako nejtransparentnější metoda alokace, která uvádí do praxe zásadu „znečišťovatel platí“ (k této zásadě blíže Vícha, 2014). Následně v roce 2015 EU přijala rozhodnutí o vytvoření rezervy tržní stability (Market Stability Reserve -MSR) s cílem odstranit velký přebytek emisních povolenek, který se v systému EU ETS vytvořil. Rezerva tržní stability měla být vytvořena v roce 2018 a od 1. ledna 2019 se uplatňuje umisťování povolenek do této rezervy. Jedná se o změnu harmonogramu dražeb, tzv. „backloading“, čili odklad dražby celkem 900 milionů povolenek z let 2014 až 2016 na roky 2019 až 2020.

172

Dne 15. července 2015 předložila Evropská komise návrh směrnice představující čtvrtou fázi EU ETS (2021–2028), v anglickém znění označovanou „Post-2020 reform of the EU ETS“. Dne 19. března 2018 byl v Úředním věstníku publikován výsledný text směrnice Evropského parlamentu a Rady 2018/410/EU ze dne 14. března 2018, kterou se mění směrnice 2003/87/ES za účelem posílení nákladově efektivních způsobů snižování emisí a investic do nízkouhlíkových technologií.

1.2 Vědecké studie v oblasti EU ETS

U vědeckých studií analyzujících systém EU ETS je důležité rozlišit jednotlivá obchodovací období EU ETS, zejména z důvodu odlišných institucionálních podmínek a pravidel v jednotlivých obdobích. Odlišná pravidla mohou mimo jiné ovlivňovat také environmentální dopad EU ETS (Segura et al., 2018). Studie analyzující první pilotní obchodovací období (2005-2007) a druhé obchodovací období, tzv. Kjótskou fázi EU ETS (2008–2012) se zaměřovaly zejména na analýzu tržních cen obchodovatelných emisních povolenek, jejich dopady na chování stakeholderů na trhu s emisními povolenkami i na chování samotných znečišťovatelů. Problematice modelování a predikcí cen obchodovatelných emisních povolenek se věnovali například studie Li et al.(2011), Garcia-Martos et al. (2013), Lecuyer a Quirion (2013), další studie analyzovaly incidenci variantních cen uhlíku (Grainger a Kolstad, 2010), zkoumaly faktory ovlivňující cenu povolenky (Aatola et al., 2013; Lutz et al., 2013), mezní náklady energeticky náročných odvětví (Chernyavska a Gulli, 2008), vliv obchodování s emisemi na výrobce elektřiny (Falbo et al., 2013) nebo inovační dopady systému EU ETS (Rogge et al., 2011; Rentizelas et al., 2012).

Autoři vědeckých studií používali celé spektrum pokročilých metod, například Li et al. (2011) používá fuzzy modelování pro plánování obchodování s emisemi CO2 v průmyslových podnicích za podmínek nejistoty. Aatola et al. (2013) vytvořil rovnovážný model trhu s emisními povolenkami za účelem určení ceny obchodovatelné emisní povolenky EUA. Falbo et al. (2013) vytvořil model založený na funkci zisku za účelem sledování dopadů EUA (EU Allowances) na optimalizační politiku výrobce elektřiny. Garcia – Martos et al. (2013) použil ARIMA a VARIMA modely pro vytvoření multivariantního modelu pro určení cen povolenky. Lecuyer a Quirion (2013) vytvořil analytický a numerický model energetického a uhlíkového trhu v EU a sledoval dopady nulové ceny uhlíku na výběr optimálního nástroje hospodářské politiky. Lutz et al. (2013) použil Markov regime-switching GARCH model pro určení nelineárních vztahů mezi cenou EUA a ekonomickými indikátory.

Z hlediska dopadů EU ETS na environmentální parametry a chování podniků jsou zajímavé empirické studie analyzující druhé obchodovací období založené na oficiálních datech, dotazníkových šetřeních a interview. Studie se zaměřují na dopady EU ETS na environmetální inovace a investice v Německu (Rogge et al., 2011), ve Švédsku (Lofgren et al., 2014) nebo v Evropské unii jako celku (Feng et al., 2012; Lovell et al., 2013; Martin et al., 2014).

Vědecké studie zkoumající třetí obchodovací fázi se vyznačují v daleko větší míře analýzou vlivů zásahů regulátora na fungování systému EU ETS a jeho parametry. Studie francouzských vědců (Cretí a Jöets, 2017) se zaměřuje na cenové bubliny, spekulace a analyzuje vývoj ceny emisní povolenky EUA na burze. Autoři dospěli k závěru, že největší výkyvy v ceně jsou důsledkem oznámení změn v politice ochrany klimatu nebo energetické politice. Obdobným tématem se zabývá i studie Fan et al.

173

(2017), která prezentuje podrobný přehled dopadů politických opatření regulátora v rámci systému EU ETS na cenu povolenky EUA za období zahrnující všechny tři obchodovací fáze (2005–2015). Autoři se zaměřují na soulad očekávaných dopadů založených na teoretických předpokladech se skutečným dopadem politických opatření regulátora a dospívají k závěru, že ne všechny politiky (události) vedou požadovaným směrem. Studie Skovgaarda (2017) se zaměřuje na analýzu rozhodování jednotlivých ministerstev financí v Nizozemsku, Dánsku a Německu v souvislosti s intervencemi a regulací systému EU ETS. Analýza ukazuje, že ministerstva byla součástí rozhodovacích procesů, podporovala intervence v rámci EU ETS, ale současně měla v jednotlivých zemích rozdílné cíle v oblasti fiskální i makroekonomické, navíc finální rozhodnutí bylo dáno politickou orientací vlády (zejména v Německu a Nizozemsku).

V České republice se problematikou EU ETS dosud zabývaly zejména studie Chmelíka a Zámyslického (2007), kteří provedli komparaci systémů emisních povolenek na základě statistických dat a platné legislativy. Zámyslický (2013) dále zkoumal vliv obchodování s emisemi skleníkových plynů na ekonomiku obnovitelných zdrojů energie. Studie Solilové a Nerudové (2014) se zaměřila na emisní obchodování a uhlíkové zdanění v Evropské unii a Zimmermannová (Zimmermannová, 2015; Zimmermannová et al., 2015) provedla ex post analýzu počátku třetího obchodovacího období EU ETS.

Pokud se zaměříme na environmentální účinnost EU ETS, můžeme najít převážně ex post analýzy chování firem a změn jejich environmentálních parametrů v důsledku EU ETS, ať už jsou to výše zmíněné studie zaměřené na environmentální inovace a investice (Rogge et al., 2011; Lofgren et al., 2014; Feng et al., 2012; Lovell et al., 2013; Martin et al., 2014; Zimmermannová, 2015) nebo studie zkoumající vztah mezi environmentálními a ekonomickými parametry firem v jednotlivých obchodovacích obdobích EU ETS (Marin et al., 2018; Segura et al., 2018). V oblasti ex post hodnocení dopadů systému EU ETS na celkové množství emisí skleníkových plynů v zemích EU28 se autorům nepodařilo v době psaní článku najít relevantní studie, zahrnující všechna obchodovací období. To je také hlavním důvodem pro provedení následující analýzy.

2 Metody

2.1 Data

Za účelem provedení geovizuální analýzy byla použita data z databáze Eurostat (Eurostat, 2018), konkrétně vývoj emisí CO2 celkem. Jedná se o roční data dostupná za země EU28, a to časové řady za období 1995–2016. Z databáze Eurostatu, respektive z její podřízené složky pro správu geografických dat (Geographic Information System of the COmmission, zkráceně GISCO), byla také použita geografická data, konkrétně se jednalo o referenční prostorová data administrativních jednotek států EU k roku 2014.

Za účelem provedení regresní analýzy byla dále použita data z burzy EEX (2019), konkrétně vývoj ceny obchodovatelných emisních povolenek EUA prodaných v aukci, kdy aukční ceny jsou k dispozici třikrát týdně (EEX, 2019). Z těchto cen byl vypočítán pro účely analýzy roční průměr; v rámci analýzy byly nejprve použity i průměrné ceny povolenek za první a druhé obchodovací období, nicméně vzhledem k velké volatilitě nebyly tyto ceny vhodné pro regresní analýzu. Pro účely analýzy byla použita rovněž data od českého Operátora trhu s elektřinou (OTE, 018), konkrétně průměrná roční

174

cena emisní povolenky ze všech uskutečněných obchodů, tj. nejen výsledná cena z aukcí, ale i z obchodů na spotovém trhu. Byla použita data za období 2005–2016.

2.2 Metody

Nejprve byla provedena geovizuální analýza prostorových dat za účelem zjištění vývoje emisí skleníkových plynů v zemích EU28 v období 2005 (rok zavedení systému EU ETS) až 2015 (období 10 let po zavedení systému EU ETS). Geovizuální analýza je termín pro oblast zkoumání geografických dat za účelem analytického myšlení a rozhodování v dané problematice, a to za využití (geo)vizualizací, výpočetních metod či konstrukce znalostí, reprezentací a manažerských strategií (Andrienko et al., 2007). Geovizuální analýza byla provedena nad zobrazením indikátorů v prostoru pomocí areálové metody (obrázek 1). U map s využitím areálové metody je u každého státu EU číslem znázorněn procentuální rozdíl mezi lety 2005 a 2015, kdy od hodnot z roku 2015 byly odečteny hodnoty z roku 2005. Proto záporné hodnoty znamenají pokles a kladné nárůst.

Následně byl prostřednictvím regresní analýzy zjišťován možný vliv systému EU ETS na snížení emisí skleníkových plynů ve státech EU28 jako celku. Za tímto účelem byla použita metoda lineární regrese a vytvořeny samostatné regresní modely. Rovnice obecného regresního modelu má tvar:

𝑌 = 𝛽0 + 𝛽1𝑋1 + 𝛽2𝑋2 + 𝛽3𝑋3+ 𝛽4𝑋4 + 𝛽5𝑋5 + 𝑢 (1)

V této rovnici jsou parametry β0...β5 regresní koeficienty, které vyjadřují vliv nezávislé proměnné na závislou proměnnou. Závislou proměnnou Y jsou emise skleníkových plynů v zemích EU28 vyjádřené v milionech tun ekvivalentu CO2. Parametr u představuje náhodnou složku modelu. Nezávislé proměnné v regresní rovnici jsou následující: X1 – průměrná roční cena povolenky EUA v aukci, X2 – roční HDP v zemích EU28 v běžných cenách, X3 – roční objem investic v zemích EU28 v běžných cenách, X4 – průměrná roční cena povolenky EUA na všech trzích (OTE, 2018), X5 – čas. Jednotlivé proměnné jsou popsány v následující Tab. 1.

Tab. 1: Proměnné v regresním modelu Proměnná Jednotky Typ proměnné Teoretický/

očekávaný dopad

Emise skleníkových plynů t/CO2 ekv. Závislá -

Průměrná roční cena EUA v aukci, běžné ceny

EUR/povolenka Klíčová vysvětlující

Negativní

Roční HDP v zemích EU28, běžné ceny

mil. EUR Kontrolní vysvětlující

Pozitivní

Roční objem investic v zemích EU28, běžné ceny

mil. EUR Kontrolní vysvětlující

Negativní

Průměrná roční cena EUA na všech trzích, běžné ceny

EUR/povolenka Kontrolní vysvětlující

Negativní

Čas 1 rok Kontrolní vysvětlující

Negativní

Zdroj: vlastní zpracování

175

Jednotlivé vysvětlující proměnné byly zvoleny z důvodu teoretického/očekávaného dopadu na množství emisí skleníkových plynů. Průměrná roční cena emisní povolenky EUA v aukci (klíčová vysvětlující proměnná) byla zvolena na základě teoretického konceptu environmentálních ekonomů (více např. Kolstad, 2010), kde zvyšující se cena obchodovatelné emisní povolenky působí adekvátní snížení emisí skleníkových plynů. Jako kontrolní vysvětlující proměnná potom byla zvolena průměrná roční cena emisní povolenky na všech trzích, která vychází ze stejného teoretického konceptu a očekává se u ní rovněž negativní vliv. S ohledem na přidělování emisních povolenek v prvních fázích obchodování zdarma (tzv. grandfathering) lze očekávat slabší motivační efekt než v případě, kdy si firmy povolenky kupují v aukci.

Proměnná roční HDP v zemích EU28 vyjadřuje ekonomický vývoj zemí EU28 a období ekonomické expanze, stagnace či recese. Ekonomický růst/stagnace/pokles je na základě teorie provázen odpovídajícím růstem/stagnací/poklesem produkce a souvisejícím množstvím emisí znečišťujících látek do ovzduší (Samuelson a Nordhaus, 2010; Kolstad, 2011).

Proměnná roční objem investic v zemích EU28 zahrnuje všechny korporátní investice ve sledovaném období. Bohužel se autorům nepodařilo získat díky nedostatku dat samostatný indikátor tzv. environmentálních investic – korporátních investic do nových, environmentálně šetrných technologií a procesů na úrovni zemí EU28. Teoretická očekávání ohledně použité proměnné „korporátních investic“ jsou v souladu s tzv. Porterovou hypotézou a výsledky obdobných studií (např. Liu et al., 2017; Jin et al., 2017), kde nárůst korporátních investic přináší adekvátní snížení emisí skleníkových plynů. Předpokládáme, že současné „nové“ technologie jsou více environmentálně šetrné než původní, tj. modernizace kapitálu vede ke zlepšení environmentálních parametrů produkce.

Poslední proměnná – čas – byla zvolena s ohledem na výsledky geovizuální analýzy, prezentované v tomto článku v kapitole 3.1. Z výsledků analýzy vyplývá, že ve sledovaném období 2005–2015 došlo až na výjimku (Lotyšsko) ke snížení emisí skleníkových plynů ve všech členských státech EU28.

V souvislosti s hlavním cílem výzkumu, uvedeným výše, je stanovena výzkumná otázka „Systém EU ETS (prostřednictvím ceny povolenky EUA v aukci) má vliv na množství emisí skleníkových plynů v EU28, tj. tento ekonomický nástroj je environmentálně účinný v EU28 jako celku“.

3 Rozbor problému

3.1 Geovizuální analýza prostorových dat

Do systému EU ETS jsou zapojeny všechny členské země EU28 a dále Norsko, Lichtenštejnsko a Island. Nejprve byla provedena geovizuální analýza (více např. Pászto et al., 2017) za účelem vyhodnocení prostorových dat.

176

Obr. 1: Emise CO2 (rozdíl 2005 a 2015) celkem

Zdroj: Eurostat, 2018; vlastní zpracování

Obrázek 1 ukazuje rozdíly v množství emisí skleníkových plynů v jednotlivých zemích zapojených do systému EU ETS mezi rokem 2005 (rok zavedení EU ETS) a rokem 2015. Je zřetelné, že ve všech zemích s výjimkou Turecka, Islandu a Lotyšska došlo ke snížení emisí. Snížení se pohybovalo od 29 (Řecko) po 1,1 procenta (Norsko). Česká republika s hodnotou snížení emisí CO2 celkem o 13,3 procenta patří spíše k těm zemím s menším poklesem. Nicméně sousední Polsko a Německo vykázalo snížení ještě méně výrazné. Na druhou stranu Turecko signifikantně (o 42,8 procent) zvýšilo ve sledovaném období celkovou produkci CO2. Je zajímavé, že Island, který je charakteristický šetrností k životnímu prostředí, vykázal ve sledovaném období nárůst celkové produkce emisí CO2 o 22,4 procenta. Nutno podotknout, že v případě Islandu jsou absolutní hodnoty emisí v řádech jednotek (oproti stovkám u Německa či Polska).

Otázkou tedy je, zda pozorovaný pokles emisí CO2 souvisí se snižováním dopadu ekonomiky na životní prostředí, zaváděním šetrnějších technologií a energetickými úsporami či zda je spíše způsoben poklesem investiční aktivity a snižováním produkce ve sledovaných zemích. K alespoň částečnému objasnění tohoto problému nám může pomoci regresní analýza a regresní model pro země EU28.

3.2 Korelační a regresní analýza EU28

Tab. 2 obsahuje popisnou statistiku k regresním modelům. U proměnné EUA aukce byla v období, kdy aukce ještě neprobíhaly, použita nula vyjadřující nulovou cenu v aukci. Rovněž u proměnné EUA průměr byla v období, kdy ještě nebyl zaveden systém EU ETS, použita nula vyjadřující nulovou cenu tuny uhlíku na burze.

177

Tab. 2: Popisná statistika Proměnná Max Min Průměr Medián

Emise CO2 (tuny) 5096987,48 3970581,22 4642231,33 4853406,98

EUA aukce (EUR) 7,61 4,40 1,05 0

HDP (mil. EUR) 14855223 7346248,80 11427945,30 11923366,20

Investice (mil. EUR) 3023653,40 1586410,20 2410834,67 2464648,85

EUA průměr (Kč) 613,46 41,64 156,48 80,14

Čas (roky) 2016 1995 2005,5 2005,5 Zdroj: Eurostat, 2018; EEX, 2019; OTE, 2018; vlastní zpracování

Nejprve byla provedena korelační analýza s využitím dat za období 1995–2016.

Tab. 3: Korelační analýza Emise Aukce HDP Investice EUA vše Čas

Emise 1

Aukce -0,78 1

HDP -0,83 0,59 1

Investice -0,66 0,44 0,96 1

EUA vše -0,29 0,01 0,53 0,57 1

Čas -0,91 0,65 0,98 0,89 0,48 1 Zdroj: vlastní zpracování

Lze pozorovat statisticky významné negativní korelační koeficienty mezi emisemi skleníkových plynů, cenou emisních povolenek v aukci, HDP a časem. Za účelem vyhodnocení možných závislostí bude provedena regresní analýza, s využitím třech různých modelů. Model MOD1 – VŠE zahrnuje všechny nezávislé proměnné použité v rámci korelační analýzy. Druhý model MOD2 – aukce zahrnuje pouze jednu nezávislou proměnnou – cenu emisní povolenky v aukci. Nakonec MOD3 - EKONOM zahrnuje nezávislé proměnné, u kterých vyšly statisticky významné negativní korelační koeficienty v rámci korelační analýzy.

Tab. 4: Regresní analýza EU28 MOD1–VŠE MOD2–AUKCE MOD3-EKONOM

Parametr Význ. Koef. Význ. Koef. Význ. Koef.

Aukce 0,08 -22848,92 0,01 -64454,79 0,04 -24748,44

HDP 0,01 0,55 X 0,00 0,28

Investice 0,17 -0,58 X x

EUA vše 0,56 73,86 X x

Čas 0,00 -207047,59 X 0,00 -145217,24

Konstanta 0,00 414949650,80 0,00 4404171,39 0,00 292705036,30

R2 0,98 0,83 0,98

N 22 9 22

Význam.F 0,00 0,01 0,00

DW 1,69 1,51 1,13

dL; dU 0,67;1,69 0,55;0,99 0,83;1,40 Zdroj: vlastní zpracování; hladina významnosti 0,05; DW – hodnota Durbin-Watsonova testu; dL,dU

hranice u Durbin-Watsonova testu

178

Z hlediska statistické významnosti jsou všechny testované modely statisticky významné, u prvního MOD1 – VŠE ovšem nejsou statisticky významné všechny proměnné. Hodnota Durbin-Watsonova testu na autokorelaci 1,69 odpovídá hraniční hodnotě pro zamítnutí hypotézy o autokorelaci. Model MOD3 – EKONOM vychází jako statisticky významný včetně jednotlivých proměnných a jejich indikátorů. Problematický se zde ovšem jeví test autokorelace (Durbin-Watson), kde hodnota 1,13 ještě spadá do oblasti s nejednoznačným rozhodnutím (interval 0,83 – 1,40). Poslední z modelů se zaměřuje pouze na samotné emisní povolenky prodávané v aukci a jejich vliv na emise CO2. Tento model pracuje s kratším časovým obdobím zahrnujícím druhou a třetí fázi EU ETS (od roku 2008). Je zřejmé, že model je statisticky významný a hodnota Durbin-Watsonova testu prokázala, že zde není autokorelace.

Na základě výsledků regresní analýzy lze konstatovat, že v zemích EU28 systém EU ETS přispívá ke snižování emisí skleníkových plynů. Z analýzy ovšem vyplývá, že vliv různých způsobů obchodování povolenek má různý vliv na snižování emisí. U emisních povolenek, které byly nejprve přidělovány zdarma a následně obchodovány, není možné potvrdit jejich environmentální dopad. Naopak u emisních povolenek prodávaných v aukci je jejich vliv na znečišťovatele daleko výraznější, a to jak samostatně (model MOD2 - VŠE), tak v kombinaci s vývojem ekonomiky a časem (MOD3 - EKONOM).

4 Diskuze

Pokud se zaměříme na vyhodnocení dosažených výsledků s ohledem na výzkumnou otázku „Systém EU ETS (prostřednictvím ceny povolenky EUA v aukci) má vliv na množství emisí skleníkových plynů v EU28, tj. tento ekonomický nástroj je environmentálně účinný v EU28 jako celku“, můžeme konstatovat, že cena emisní povolenky EUA v aukci je environmentálně účinná. Množství emisí skleníkových plynů v zemích EU28 ve zvoleném regresním modelu ovlivňují negativně konkrétně dvě proměnné - cena obchodovatelné emisní povolenky EUA v aukci a čas (zahrnující období EU ETS). Proměnná HDP ovlivňuje sledované množství emisí pozitivně. Ostatní ekonomické indikátory nejsou v rámci regresního modelu statisticky významné.

Ohledně environmentálního dopadu ceny emisní povolenky EUA je však potřeba zdůraznit, že se jedná pouze o cenu emisní povolenky v aukci (systémová změna platná od třetí obchodovacího období 2013-2020). Pokud se zaměříme na vliv ceny EUA na všech trzích po celou dobu účinnosti systému EU ETS od roku 2005, je tato proměnná z pohledu modelu statisticky nevýznamná. Získané výsledky částečně korespondují s výsledky studie Segura et al. (2018), kde autoři rovněž zdůrazňují rozdílný vliv jednotlivých obchodovacích období na environmentální parametry firem, přičemž nejvýznamnější redukce emisí CO2 byly zjištěny právě ve třetím obchodovacím období. Je zřejmé, že systém aukcí má na environmentální účinnost EU ETS výrazný vliv. S ohledem na poslední vývoj ceny EUA v aukci v průběhu období 2018-2019, kdy se cena jedné povolenky dostala až na úroveň 27 EUR/t (EEX, 2019), lze do budoucna očekávat ještě výraznější dopad evropského systému emisního obchodování na snížení emisí skleníkových plynů u jednotlivých znečišťovatelů.

V regresních modelech se také projevil jako významný faktor čas, což může souviset s celkovým tlakem společnosti na environmentální parametry firem a environmentální inovace; z pohledu nástrojů hospodářské politiky je systém EU ETS nejen ekonomickým nástrojem spojeným s emisním obchodováním, ale rovněž

179

představuje přísnou legislativu – administrativní nástroj, který se v čase rovněž „přiostřuje“, což vyplývá i z nově schválené legislativy ke čtvrtému obchodovacímu období systému EU ETS. Směrnice 2018/410/EU pro čtvrtou fázi EU ETS (2021-2030) se soustředí zejména na zvýšení podílu povolenek dražených v aukci (57 %) a přísnější pravidla pro přidělování bezplatných povolenek.

S ohledem na výsledky analýzy prezentované v tomto článku doporučují autoři tvůrcům hospodářské politiky prosazovat další zvýšení množství emisních povolenek prodávaných v aukci, které je žádoucí, pokud má mít systém EU ETS výraznější environmentální dopad.

Závěr

Hlavním cílem tohoto článku bylo provést analýzu vlivu systému emisních obchodovatelných povolenek EU ETS na množství emisí skleníkových plynů v zemích EU28 jako celku. Z geovizuální analýzy prostorových dat je zřejmé, že téměř ve všech sledovaných zemích došlo v období 2005–2015 ke snížení emisí skleníkových plynů. Na základě výsledků regresní analýzy můžeme dále konstatovat, že ke snižování emisí skleníkových plynů v zemích EU28 jako celku přispívá i systém EU ETS, zejména po změně systému přidělování povolenek. V rámci regresních modelů hraje významnou roli také čas, kde můžeme pozorovat negativní vztah s množstvím emisí skleníkových plynů. Významný vliv času může souviset s celkovým tlakem hospodářské politiky na environmentální parametry firem a environmentální inovace.

V souvislosti s dalším obchodovacím obdobím EU ETS bude potřeba v průběhu roku 2021 vyhodnotit třetí obchodovací období 2013–2020 a jeho ekonomickou a environmentální účinnost. Předmětem dalšího výzkumu v oblasti obchodovatelných emisních povolenek by proto, s ohledem na prostorový charakter tohoto ekonomického nástroje, měla být komplexnější prostorová analýza zaměřená na možné rozdílné působení EU ETS v jednotlivých členských zemích a související prostorová analýza nákladů na zamezení.

Poděkování

Tento článek vznikl díky podpoře projektu Spationomy (2016-1-CZ01-KA203-024040) spolufinancováno z programu Evropské unie Erasmus+.

Reference

Aatola, P., Ollikainen, M., Toppinen, A. (2013). Price determination in the EU ETS market: Theory and econometric analysis with market fundamentals, Energy Economics, 36, March 2013, s. 380–395.

Andrienko, G., Andrienko, N., Jankowski, P., Keim, D., Kraak, M. J., MacEachren, A., Wrobel, S. (2007). Geovisual analytics for spatial decision support: Setting the research agenda. International journal of geographical information science, 21(8), s. 839-857.

Cretí, A., Joëts, M. (2017). Multiple bubbles in the European Union Emission Trading Scheme. Energy Policy, 107(April), s. 119–130, DOI: 10.1016/j.enpol.2017.04.018.

EEX. (2019). EU Emission Allowances – Prices and Trading Volumes. [online] Dostupné na: https://www.eex.com [13.5.2019].

European Commission. (2015). EU ETS Handbook. [online] Dostupné na: https://ec.europa.eu/clima/sites/clima/files/docs/ets_handbook_en.pdf [20.9.2018].

Eurostat. (2018). Greenhouse gas emissions. GDP and its components. [online] Dostupné na: http://ec.europa.eu/eurostat/ [20.9.2018].

180

Falbo, P., Felletti, D., Stefani, S. (2013). Free EUAs and fuel switching, Energy Economics, 2013, 35, January 2013, s. 14–21.

Fan, Y., Jia, J. J., Wang, X., Xu, J. H. (2017). What policy adjustments in the EU ETS truly affected the carbon prices? Energy Policy, 103 (July 2016), s. 145–164, DOI: 10.1016/j.enpol.2017.01.008.

Feng, Z. H., Wei, Y. M., Wang, K. (2012). Estimating risk for the carbon market via extreme value theory: An empirical analysis of the EU ETS, Applied Energy, 2014, 99, s. 97–108.

García-Martos, C., Rodríguez, J., Sánchez, M. J. (2013). Modelling and Forecasting Fossil Fuels, CO2 and Electricity Prices and their Volatilities, Applied Energy, 101, January 2013, s. 363–375.

Grainger, C. A., Kolstad, C. D. (2010). Who Pays a Price on Carbon?, Environmental and Resource Economics, 46 (3), s. 359-376.

Chernyavska, L., Gulli, F. (2008). Marginal CO2 cost pass-through under imperfect competition in power markets, Ecological Economics, 68, December 2008, s. 408-421.

Chmelík, T., Zámyslický, P. (2007). EU Emissions Trading Scheme as Implemented in the Czech Republic. Acta Polytechnica, 47 (6), s. 43 - 45.

Jin, L., Duan, K., Shi, C., Ju, X. (2017). The Impact of Technological Progress in the Energy Sector on Carbon Emissions: An Empirical Analysis from China. International Journal of Environmental Research and Public Health, 14(12), s. 1505.

Kolstad, C. D. (2011). Environmental Economics, 2nd Edition. New York: Oxford University Press.

Lecuyer, O., Quirion, P. (2013). Can Uncertainty Justify Overlapping Policy Instruments to Mitigate Emissions?, Ecological Economics, 93, September 2013, s. 177–191.

Li, M. W., Li, Y. P., Huang, G. H. (2011). An Interval-Fuzzy Two-Stage Stochastic Programming Model for Planning Carbon Dioxide Trading under Uncertainty, Energy, 2011, 36 (9), s. 5677-5689.

Liu, Y., Hao, Y., Gao, Y. (2017). The environmental consequences of domestic and foreign investment: evidence from China. Energy Policy, 108, s. 271-280.

Lofgren, A., Wrake, M., Hagberg, T., Roth, S. (2014). Why the EU ETS needs reforming: an empirical analysis of the impact on company investments, Climate Policy, 2014, 14 (5), s. 537–558.

Lovell, H., Bebbington, J., Larrinaga, C., De Aguiar, T.R.S. (2013). Putting carbon markets into practice: a case study of financial accounting in Europe, Environment and Planning C-Government and Policy, 31(4), s. 741–757.

Lutz, B. J., Pigorsch, U., Rotfuß, W. (2013). Nonlinearity in cap-and-trade systems: The EUA price and its fundamentals, Energy Economics, 40, November 2013, s. 222–232.

Marin, G., Marino, M., Pellegrin, C. (2018). The impact of the European Emission Trading Scheme on multiple measures of economic performance. Environmental and Resource Economics, 71(2), s. 551-582.

Martin, R., Muuls, M., De Preux, L. B., Wagner, U. J. (2014). On the empirical content of carbon leakage criteria in the EU Emissions Trading Scheme, Ecological Economics, 2014, 105, s. 78–88.

OTE (2018). Průměrná cena emisní povolenky. [online] Dostupné na: https://www.povolenky.cz/ [20.9.2018].

Pászto, V., Burian, J., Panek, J., Macku, K. (2017). Mapping the economic data - Case studies and best practices, International Multidisciplinary Scientific GeoConference Surveying Geology and Mining Ecology Management, SGEM, 17 (21), s. 975-982, DOI: 10.5593/sgem2017/21/S07.123

Rentizelas, A. A., Tolis, A.I.,Tatsiopoulos, I.P. (2012). Investment planning in electricity production under CO2 price uncertainty, International Journal of Production Economics, 140(2), s. 622-629.

Rogge, K. S., Schneider, M., Hoffmann, V. H. (2011). The innovation impact of the EU Emission Trading System — Findings of company case studies in the German power sector, Ecological Economics, 70, January 2011, s. 513-523.

Samuelson, P. A., Nordhaus, W. D. (2010). Economics. 19th ed. The McGraw-Hill series economics.

181

Segura, S., Ferruz, L., Gargallo, P., Salvador, M. (2018). Environmental versus economic performance in the EU ETS from the point of view of policy makers: A statistical analysis based on copulas. Journal of Cleaner Production, 176, s. 1111-1132, DOI: 10.1016/j.jclepro.2017.11.218.

Skovgaard, J. (2017). The Role of Finance Ministries in Environmental Policy Making: The case of European Union Emissions Trading System reform in Denmark, Germany and the Netherlands. Environmental Policy and Governance, 27(4), s. 351–364, DOI: 10.1002/eet.1767.

Solilova, V., Nerudova, D. (2015). Evaluation of Greenhouse Gas Emissions and Related Aspects: Case of the Czech Republic, Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 63(1), s. 281–292.

Vícha, O. (2014). Princip „znečišťovatel platí“ z právního pohledu. 1. vydání, Praha: Linde Praha a.s.

Zámyslický, P. (2013). Vliv obchodování s emisemi skleníkových plynů na ekonomiku obnovitelných zdrojů energie. Disertační práce. České vysoké učení technické v Praze, Fakulta elektrotechnická.

Zimmermannová, J. (2015). Pilot Analysis of the Behaviour of Companies Within the 3rd Trading Period of the EU ETS in the Czech Republic. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 63(6), s. 2213–2220, DOI: 10.11118/actaun201563062213.

Zimmermannova, J., Cermak, P., Novak, P. (2015). Ex-post analysis of the EU emission trading in year 2013 in the Czech Republic. Economics and Sociology, 8(2), DOI: 10.14254/2071-789X.2015/8-2/13

Kontaktní adresa

doc. Ing. Jarmila Zimmermannová, Ph.D. Moravská vysoká škola Olomouc, Ústav ekonomie Tř. Kosmonautů 1, 779 00 Olomouc, Česká republika E-mail: [email protected] Tel. číslo: + 420587332318

Mgr. Vít Pászto, Ph.D. Univerzita Palackého v Olomouci, Přírodovědecká fakulta, Katedra geoinformatiky Moravská vysoká škola Olomouc, Ústav informatiky a aplikované matematiky Tř. Kosmonautů 1, 779 00 Olomouc, Česká republika E-mail: [email protected] Tel. číslo: + 420587332320

JUDr. Ondřej Vícha, Ph.D. Univerzita Palackého v Olomouci, Právnická fakulta, Katedra správního práva a finančního práva Tř. 17. listopadu 8, 771 11 Olomouc, Česká republika E-mail: [email protected] Tel. číslo: + 420585637666


182

REVIEW PROCEDURE At first, each submission is assessed by executive editor and by The Editorial Board, who check whether the submission satisfies the journal requirements. If it does, they send the papers to the 2 referees. Referees elaborate evaluation reports in the form of a standard blank form (sent to referees by the executive editor). Referees evaluate the papers from content and formal point of view, a summary of comments for the author(s) is included. The referee recommendation to the editor should be one of the following: Accept for publication, accept with minor revisions, accept with major revisions, reject. The Editorial Board reserves the right to return to the authors any manuscript that in their opinion is not suitable for publication in the journal, without expressly stipulating the reasons for doing so. Accepted articles for publication do not undergo linguistic proofreading. Executive editor informs the author about the result of the review procedure.

EDITORIAL BOARD

Senior editor doc. Ing. Romana Provazníková, Ph.D.

Executive editor Ing. Filip Gyenes

Members prof. Ing. Jan Čapek, CSc. doc. Ing. Ivana Kraftová, CSc. doc. RNDr. Bohdan Linda, CSc. doc. Ing. et Ing. Renáta Myšková, Ph.D. doc. Ing. Jolana Volejníková, Ph.D. Ing. Karel Šatera, PhD. MBA

SCIENTIFIC BOARD

prof. Dr. John R. Anchor, University of Huddersfield, England doc. Ing. Josef Brčák, CSc., Czech University of Life Sciences Prague, Czech Republic prof. Ing. Jiří Dvořák, DrSc., University of Technology Brno, Czech Republic doc. Ing. Peter Fabian, CSc., The University of Žilina, Slovak Republic prof. Ing. Mojmír Helísek, CSc., The University of Finance and Administration Prague, Czech Republic prof. Dr. Roberto Henriques, Universidade Nova de Lisboa, Portugal prof. Lazaros S. Iliadis, Democritus University of Thrace, Greece prof. Ing. Ivan Jáč, CSc., Technical university of Liberec, Czech Republic prof. Ing. Christiana Kliková, CSc., Masaryk University, Czech Republic prof. Sergey V. Kostarev, PhD., Omsk Institute of Plekhanov Russian University of Economics, Russia prof. JUDr. Jozef Králik, CSc., Pan-European University, Slovak Republic

183

prof. PhDr. Miroslav Krč, CSc., University of Defence Brno, Czech Republic prof. PhDr. Karel Lacina, DrSc., The University of Finance and Administration Prague, Czech Republic doc. Ing. Ivana Linkeová, Ph.D., Czech Technical University in Prague, Czech Republic prof. JUDr. Eleonóra Marišová, PhD., Slovak University of Agricultural in Nitra, Slovak Republic prof. dr. Ing. Miroslav Pokorný, Technical University of Ostrava, Czech Republic prof. Ing. Jiří Polách, CSc., Business School Ostrava, Czech Republic prof. univ. dr. Angela Repanovici, Universitatea Transylvania Brasov, Romania prof. dr. habil. Aleksander Sladkowski, Silesian University of Technology, Poland prof. dr. Ligita Šimanskienė, Klaipėda University, Lithuania prof. dr. Teodoras Tamošiūnas, Šiauilai University, Lithuania doc. Ing. Miloš Vítek, CSc., Emeritus academic staff University of Hradec Králové, Czech Republic

CONTACT US University of Pardubice Faculty of Economics and Administration Studentská 95, 532 10 Pardubice, Czech Republic Email: [email protected] www.uni-pardubice.cz

Journal web page: http://www.upce.cz/en/fes/veda-vyzkum/fakultni-casopisy/scipap.html

184

Title Scientific Papers of the University of Pardubice, Series D Publisher University of Pardubice

Faculty of Economics and Administration Studentská 95, 532 10 Pardubice, Czech Republic

Registration number IČ 00216275 Edited by Ing. Filip Gyenes Periodicity 3 per year Volume number 27 Issue number 46 (2/2019) Place Pardubice Publication Date 30. 8. 2019 Number of Pages 184 Number of Copies 50 Edition First Printed by Printing Centre of the University of Pardubice

MK ČR E 19548 ISSN 1211-555X (Print) ISSN 1804-8048 (Online)

1804-8048 (online)