Document of The World Bankdocuments.worldbank.org/curated/en/... · 12/18/2015 · (Exchange Rate Effective September 23, 2015) Currency Unit = Kuna HRK 1.00 = US$ 0.15 US$ 1.00

Document of The World Bank

Report No: ICR00003606

IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-74710)

ON A

LOAN

IN THE AMOUNT OF EURO 50 MILLION

(US$68 MILLION EQUIVALENT)

TO THE

REPUBLIC OF CROATIA

FOR A

REVENUE ADMINISTRATION MODERNIZATION PROJECT

December 16, 2015

Public Sector Management Department Croatia Europe and Central Asia Region

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

CURRENCY EQUIVALENTS

(Exchange Rate Effective September 23, 2015)

Currency Unit = Kuna HRK 1.00 = US$ 0.15 US$ 1.00 = 6.85

FISCAL YEAR

January 1st – December 31st

ABBREVIATIONS AND ACRONYMS

APIS Information Systems and Information Technologies Support Agency APL Adaptable Program Loan BCMP Business Change Management Plan CTA Croatian Tax Administration CSPM Committee for the Strategic Project Management DPC Data Processing Center DSIT Development Strategy for Tax Administration Information Technology System EMP Environmental Management Plan EU European Union GDP Gross Domestic Product GOC Government of Croatia HRM Human Resource Management ICT Information Communications Technology LMF International Monetary Fund JMBG Citizen identification number issued by Ministry of Interior LTO Large Taxpayers Office MOF Ministry of Finance OECD Organization of Economic Development and Cooperation O M Operations Manual PHARE European Union Technical Assistance Fund PIC Project Implementation Committee PIT Personal Income Tax PIU Project Implementation Unit PPF Project Preparation Facility RAMP Revenue Administration Modernization Project R&D Research and Development CBS Central Bureau of Statistics SOE Statement of Expenditures TA Technical Assistance TIN Taxpayer Identification Number VAT Value-Added Tax VIES VAT Information Exchange System

Senior Global Practice Director: James Brumby (Acting)

Practice Manager: Adrian Fozzard

Project Team Leader: Alberto Leyton

ICR Team Leader: Ismail Radwan

i

CROATIA Revenue Administration Modernization Project

TABLE OF CONTENTS

Data Sheet ....................................................................................................................................... iii

Executive Summary ............................................................................................................ 1

1. Project Context, Development Objectives and Design ............................................... 3

2. Key Factors Affecting Implementation and Outcomes .............................................. 7

3. Assessment of Outcomes .......................................................................................... 10

4. Assessment of Risk to Development Outcome ......................................................... 20

5. Assessment of Bank and Borrower Performance ..................................................... 21

6. Lessons Learned ........................................................................................................ 23

7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners ........... 24

Annex 1. Project Costs and Financing .............................................................................. 26

Annex 2. Outputs by Component...................................................................................... 27

Annex 3. Economic and Financial Analysis ..................................................................... 30

Annex 4. Bank Lending and Implementation Support/Supervision Processes ................. 34

Annex 5. Beneficiary Survey Results ............................................................................... 36

Annex 6. Stakeholder Workshop Report and Results ....................................................... 39

Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ......................... 43

Annex 8. CTA Revenues 2007-2014 ................................................................................ 48

Annex 9. List of Supporting Documents .......................................................................... 49

MAP .................................................................................................................................. 50

ii

List of Tables Table 1: CTA staff surveys 2010 and 2012 ................................................................................... 12 Table 2: Share of large taxpayers in total tax, 2012-2014 ............................................................. 12 Table 3: CTA reorganization ......................................................................................................... 12 Table 4: CTA total staff numbers in audit ..................................................................................... 13 Table 5: CTA audit staff in LTO ................................................................................................... 13 Table 6: Number of LTO audits 2013-2014 .................................................................................. 13 Table 7: Average assessments per auditor, selected years 2007-2014 .......................................... 13 Table 8: CTA training outputs ....................................................................................................... 14 Table 9: CTA training received (hours per staff) .......................................................................... 14 Table 10: Revenue collected and average cost per employees ...................................................... 14 Table 11: CTA revenue performance 2010-2012 .......................................................................... 15 Table 12: Perceptions of CTA quality (1 to 5 scale) ..................................................................... 15 Table 13: CTA website visits 2012-2015 ...................................................................................... 16 Table 14: PDO Indicator: Tax Gap 2011-2013 ............................................................................. 16 Table 15: PDO Indicator: Taxpayer compliance costs .................................................................. 17 Table 16: Doing business data ....................................................................................................... 17 Table 17: Stop-filers 2011-2014 .................................................................................................... 17 Table 18: Electronic filing 2007-2014 .......................................................................................... 18 Table 19: Spending by Component ............................................................................................... 18

iii

A. Basic Information

Country: Croatia Project Name: Revenue Administration Modernization Project

Project ID: P102778 L/C/TF Number(s): IBRD-74710

ICR Date: 12/04/2015 ICR Type: Core ICR

Lending Instrument: SIL Borrower: GOVERNMENT OF CROATIA

Original Total Commitment:

USD 68.00M Disbursed Amount: USD 15.78M

Revised Amount: USD 15.59M

Environmental Category: B

Implementing Agencies: Croatian Tax Administration

Cofinanciers and Other External Partners: None B. Key Dates

Process Date Process Original Date Revised / Actual

Date(s)

Concept Review: 12/20/2006 Effectiveness: 12/21/2007 12/21/2007

Appraisal: 05/18/2007 Restructuring(s): 09/09/2010 06/13/2013

Approval: 06/28/2007 Mid-term Review:

Closing: 06/30/2013 06/30/2015 C. Ratings Summary C.1 Performance Rating by ICR

Outcomes: Moderately Unsatisfactory

Risk to Development Outcome: Low or Negligible

Bank Performance: Moderately Unsatisfactory

Borrower Performance: Moderately Unsatisfactory

C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings

Quality at Entry: Moderately Satisfactory Government: Moderately Unsatisfactory

Quality of Supervision: Moderately Unsatisfactory

Implementing Agency/Agencies:

Moderately Unsatisfactory

Overall Bank Performance:


Overall Borrower Performance:


iv

C.3 Quality at Entry and Implementation Performance Indicators

Implementation Performance

Indicators QAG Assessments

(if any) Rating

Potential Problem Project at any time (Yes/No):

Yes Quality at Entry (QEA):

None

Problem Project at any time (Yes/No):

Yes Quality of Supervision (QSA):

None

DO rating before Closing/Inactive status:


D. Sector and Theme Codes

Original Actual

Sector Code (as % of total Bank financing)

Central government administration 100 100

Theme Code (as % of total Bank financing)

Administrative and civil service reform 25

Regional integration 25 25

Tax policy and administration 50 75 E. Bank Staff

Positions At ICR At Approval

Vice President: Cyril E Muller Shigeo Katsu

Country Director: Christian Bodewig (Acting) Anand K. Seth

Practice Manager/Manager:

Adrian Fozzard Cheryl W. Gray

Project Team Leader: Alberto Leyton Waleed Haider Malik

ICR Team Leader: Ismail Radwan

ICR Primary Author: Ismail Radwan F. Results Framework Analysis

Project Development Objectives (from Project Appraisal Document) The development objective of the Project is to achieve further improvements in efficiency, taxpayer services, and tax compliance through capacity building and systems improvement in the Croatian Tax Administration (CTA). This will be accomplished by supporting the development of a strategic plan for tax modernization and assisting in its implementation over the next five years. This investment

v

will move the tax administration closer toward harmonization of its policies and operations with EU accession expectations. Revised Project Development Objectives (as approved by original approving authority) (a) PDO Indicator(s)

Indicator Baseline Value

Original Target Values (from

approval documents)

Formally Revised Target Values

Actual Value Achieved at

Completion or Target Years

Indicator 1 : Increase in taxpayer compliance rates, based on internationally accepted method for estimating tax gap

Value quantitative or Qualitative)

2.4 At least 5% improvement

1.52

Date achieved 12/31/2008 12/31/2008 12/31/2013 Comments (incl. % achievement)

Fully Achieved. Progress was on track by 2013 but more recent data is not available.

Indicator 2 : Decrease in taxpayer compliance costs, as assessed by taxpayer services (annual tax compliance average cost for legal entities)


14,495 Kuna At least 10% improvement

11,977 Kuna


Fully achieved. Progress was on track by 2012 but there was no survey on completion as originally planned.

Indicator 3 : Perception of taxpayers regarding the quality of service provided by CTA, as measured by periodic surveys


2.66 (out of 5) At least 5% improvement

3.19


Fully achieved. Progress was on track by 2013 but there was no survey on completion as originally planned.

Indicator 4 : Perception of tax officers regarding the terms of employment and working environment, as measured by periodic surveys


3.05 At least 10% improvement

3.14


Not fully achieved. Progress was on track by 2013 but there has not been another survey since then.

vi

(b) Intermediate Outcome Indicator(s)

Indicator Baseline Value

Original Target Values (from

approval documents)

Formally Revised

Target Values

Actual Value Achieved at

Completion or Target Years

Indicator 1 : A pilot LTO in Zagreb that would report directly to the head of the CTA established

Value (quantitative or Qualitative)

No genuine national LTO exists

A fully-fledged national level LTO is in operation

A fully-fledged national level LTO is in operation


Fully achieved. The LTO operations will be enhanced once it is integrated with the CRMS in line with the BPR recommendations.

Indicator 2 : Ratio of staff in inspections functions as a percent of total staff in the LTO Value (quantitative or Qualitative)

49% 65% 51% (59% in October 2015)


Not achieved. The LTO was reorganized in January 2015. 82% of the positions are filled and they remain focused on improving services and strengthening the team.

Indicator 3 : Training per CTA staff including distance learning Value (quantitative or Qualitative)

8.00 hours 18.75 hours 29.95 (21.97 in 2014)

Date achieved 12/31/2011 12/31/2013 Comments (incl. % achievement)

Fully achieved

Indicator 4 : E-learning modules developed and implemented Value (quantitative or Qualitative)

0 (non e-learning) 10 10

Date achieved 12/31/2007 06/30/2015

Comments (incl. % achievement)

Fully achieved. The e-modules included: Anticorruption, E-Taxation, VAT, Small Value Public Procurement, Information Protection and Safety, Management Performance Strategy, Unrelated Transactions Specification, Enforced Collection and two others.

Indicator 5 : Stop filers of the medium and small taxpayer segments (VAT) Value (quantitative or Qualitative)

2.81% 2.00 3.61%

Date achieved 12/31/2011 12/31/2014 Comments (incl. % achievement)

Not achieved

vii

Indicator 6 : Annual average assessments per inspector increased Value (quantitative or Qualitative)

€ 348,075 At least 5% improvement

€ 348,707


Not achieved

Indicator 7 : Total declarations by legal entities filed electronically Value (quantitative or Qualitative)

3.0% 90.0% 99.2%


Fully achieved

G. Ratings of Project Performance in ISRs

No. Date ISR Archived

DO IP Actual

Disbursements (USD millions)

1 11/27/2007 Moderately Satisfactory Moderately Satisfactory 0.00

2 06/18/2008 Moderately

Unsatisfactory Moderately

Unsatisfactory 0.00


Unsatisfactory Moderately Satisfactory 2.13

4 06/26/2009 Moderately Satisfactory Moderately Satisfactory 7.15 5 11/22/2009 Moderately Satisfactory Moderately Satisfactory 7.15

6 03/04/2010 Moderately SatisfactoryModerately

Unsatisfactory 7.49



Unsatisfactory 7.49

8 12/28/2010 Moderately Satisfactory Moderately Satisfactory 7.97 9 07/09/2011 Moderately Satisfactory Moderately Satisfactory 7.97

10 01/25/2012 Moderately Satisfactory Moderately Satisfactory 9.89



Unsatisfactory 9.89

12 04/24/2013 Moderately Satisfactory Moderately Satisfactory 11.66 13 12/07/2013 Moderately Satisfactory Moderately Satisfactory 13.49 14 03/05/2014 Moderately Satisfactory Moderately Satisfactory 16.46

15 09/21/2014 Moderately SatisfactoryModerately

Unsatisfactory 16.46



Unsatisfactory 16.92

viii

H. Restructuring (if any)

Restructuring Date(s)

Board Approved

PDO Change

ISR Ratings at Restructuring

Amount Disbursed at

Restructuring in USD millions

Reason for Restructuring & Key Changes Made

DO IP

09/09/2010 No MU MU 7.97 Suitable site for HQ building was not available. Euro 25 million was cancelled.

06/13/2013 No MS MS 11.66

Extension of closing date and cancellation of Euro 1.5 million. Focusing of the project on institutional reform aspects.

I. Disbursement Profile

1

Executive Summary

The Croatia Revenue Administration Modernization Project (RAMP) was approved on June 4, 2007 with the objective of achieving further improvements in efficiency, taxpayer services, and tax compliance through capacity-building and systems improvement in the Croatia Tax Administration (CTA). The project supported an important part of the EU pre-accession agenda in the period up to Croatia’s accession as the 28th EU member state on July 1st 2013.

RAMP had four components: (i) Organizational consolidation and a functional realignment of the Tax Office (creation of a large taxpayer office); (ii) Knowledge and professional upgrading of tax officials and stakeholders; (iii) Technological upgrading for MIS and TIN implementation support; and (iv) Modernization management and project support. The initial project cost was € 93 million of which € 50 million was provided by an IBRD loan. The bulk of the borrower’s contribution was earmarked to finance the construction of a new headquarter building under the first component.

The project was restructured twice, first in September 2010, when it became apparent that a suitable site for the proposed HQ building could not be found. This resulted in a cancellation of € 25 million, half the original loan amount. The second restructuring took place in 2013, extending the project closing date by two years to June 30th 2015, cancelling a further € 1.5 million and focusing the project on institutional reform.

The project got off to a slow start due to inadequate preparation and lack of client ownership. The Bank and the Borrower had failed to develop stakeholder support for what was essentially a change management project. Subsequently, frequent changes in leadership at the CTA and the task team, and weak project management hampered project implementation. Project implementation proceeds fitfully: over its lifetime the project disbursed less than US$ 2 million per year.

Nonetheless, the project supported some important reforms: the creation of a fully-fledged LTO; implementation of a Taxpayer Identification Number (TIN) closely integrated into the new e-citizen initiative; successful roll-out of a series of e-training modules; implementation of e-filing, with coverage rising from 3% to 99%; and upgrading of IT systems and creation of a revamped CTA website (visited more than 26 million times in 2014). These achievements greatly improved tax-payer services, increased efficiency and reduced compliance costs, bringing significant benefits to taxpayers. The economic analysis results in an economic IRR of 259 percent and NPV of € 72.7 million in real terms. However, progress against other key indicators, such as the tax gap, number of stop-filers and average annual assessments per tax inspector did not meet expectations. The ICR concludes that the project outcome was Moderately Unsatisfactory.

Three lessons for the design and implementation of change management projects are highlighted here. Such projects should: focus on change management activities even if these account for only a small part of the overall project cost; undertake stakeholder assessments and regular consultations to build consensus within institutions and with external stakeholders; and include staff familiar with construction on the team when they include civil works to ensure that these components are properly prepared and ready for implementation.

When the project closed in June 2015, two major activities, the implementation of the Compliance Risk Management System (CRMS) and the implementation of the Disaster Recovery and Business Continuity (DRBC) site, had still not been completed. When the Borrower’s request for a project extension was rejected by the Bank, the Borrower allocated resources from the State Budget in order to complete these two important initiatives. This signals political will and commitment and suggests that the risk to development outcomes is negligible. The borrower continues to improve the CTA’s performance using financial resources from the European Commission and other partners.

2

3

1. Project Context, Development Objectives and Design 1.1 Context at Appraisal

1. At appraisal, Croatia had experienced a period of solid GDP growth (averaging 4.8 percent in 2001-06) and low inflation (contained at 2.7 percent). The unemployment rate remained high at 11.2 percent in 2006. Domestic demand and higher international energy prices caused the current account deficit to widen from 6.6 percent in 2005 to 7.4 percent of GDP in 2006.

2. The Government’s medium-term fiscal strategy aimed to continue fiscal consolidation, reduce the large current account deficit, ensure a sustainable debt path, and boost economic growth by reducing the burden of a large public sector. Government had reduced the consolidated general government deficit from 6.4 percent of GDP in 2003 to an estimated 3.1 percent in 2006. Further adjustment entailed structural reforms in the areas of pensions, health, and subsidy reduction, expenditure restraint and improvements in expenditure management and tax administration.

3. Fiscal reforms were central to the Government of Croatia’s (GOC) medium-term agenda of structural and institutional modernization that was intended to promote EU accession. Croatia joined the EU as the 28th member state on July 1st 2013. The GOC program was focused on improving the investment climate, fiscal consolidation, and strengthening governance. Government had already made improvements in its revenue administration. EU accession discussions relating to the Croatian Tax Administration (CTA) centered on three main issues: legislative harmonization; institutional capacity; and IT systems and data exchange capabilities.

4. Croatia’s overall governance and anti-corruption performance also was impressive. World Bank Business Environment and Enterprise Performance Survey (BEEPS) suggested that tax rates and tax administration improved markedly between 2002 and 2005. When benchmarked against the average of all countries in the Europe and Central Asia Region (ECA) and South Eastern European Countries (SEE), Croatia performed better than average on 20 of 21 survey dimensions. According to Transparency International’s 2005 Global Corruption Barometer, perceived corruption in tax revenue in Croatia was rated 3.3 (on an index of 5). This was better than the Central European average (3.6), but not as good as the Western European average (2.9).

1.2 Original Project Development Objectives (PDO) and Key Indicators

5. The development objective of the Project was to achieve further improvements in efficiency, taxpayer services and tax compliance through capacity building and systems improvement in the Croatian Tax Administration (CTA).

6. The reforms were to be underpinned by the development and implementation of a strategic plan for tax modernization over the next five years. The project sought to harmonize CTA policies and operations with EU accession requirements. The key project indicators to measure performance were as follows:

Increase in taxpayer compliance rates, based on internationally accepted method for estimating the tax gaps;

Decrease in taxpayer compliance costs, as assessed by taxpayer surveys;

Perception of taxpayers regarding the quality of service provided by CTA, as measured by periodic surveys; and

Perception of tax officers regarding the terms of employment and working environment, as measured by periodic surveys.

4

1.3 Revised PDO and Key Indicators, and reasons / justification

7. The first project restructuring in September 2010, left the project timeline and PDO unchanged but entailed changes to the methodologies, baselines and intermediate targets for two of the four PDO indicators to accommodate the changed project activities. The Results Framework was updated and the Intermediate Results Indicators were streamlined, reducing the original 27 intermediate indicators that appeared in the PAD and replacing them with just 8 intermediate results indicators that were more aligned with the remaining project activities.

8. A second restructuring (level 2) approved in June 2013 extended the original project closing date from June 30, 2013 to June 30, 2015, and reallocated the loan proceeds to focus on institutional reform components. The PDO indicators remained unchanged but the Intermediate Outcome Indicators were revised to remove those tracking the following activities: the consolidation of the tax administration buildings; Human Resource indicators (including the number of vacancies at CTA and the turnover rate for economists, lawyers and ICT specialists); the Physical and Legal Person ID (although the Project contributed to the establishment of the OIB, achieved in September 2010, its role was not significant). The indicators related to the newly established Large Taxpayer Office (LTO) were adjusted and indicators were added to monitor progress on capacity building through training and efficiency of tax administration for medium and small taxpayer segments.

1.4 Main Beneficiaries

9. The direct beneficiary of this project was the Croatian Tax Administration (CTA). The project sought to reduce CTA operational costs, provide for streamlined and effective data processing, reducing tax evasion and corruption, facilitate e-taxpayer services and increase tax revenues. To this end, the project supported improvements in CTA’s information exchange with different government agencies including HANFA (Croatian financial services supervisory agency) and REGOS (central registry of insured persons).

10. Croatian taxpayers would also benefit through improved customer service, reduced compliance costs, simplified forms and refund procedures, improved transparency and fair tax assessments. All Croatians could also expect to benefit from increased tax revenues raised at a lower cost.

1.5 Original Components

Component I. Organizational Consolidation and Functional Realignment of Tax Office, including Physical Facilities (Estimated Bank financing € 26 million)

11. This component financed CTA’s consolidation of the thirteen existing office locations (representing about 1,500 staff members and about 30 percent of all tax returns) into a new headquarters complex in Zagreb. The new building would include a Tax Academy and an ICT center. The component would also facilitate improved organizational and staffing arrangements, taxpayer services and enforcement processes, and human resource management policies. Through civil works, new IT systems, training and advisory services, would permit the centralized, automated, and high-volume processing of paper tax returns.

12. The component also supported CTA’s efforts to strengthen and consolidate its large taxpayers’ office (LTO) set up in 2005 in a single location, through the provision of IT, training and technical advisory services. It was estimated that approximately 1,200 large taxpayers generated about 20 percent of total tax revenue. The Zagreb LTO realignment would create separate sections for every tax administration function (e.g., taxpayer services, audit, and arrears collection) and provide targeted stakeholder outreach, risk based selection and create a single point of contact for services with specialized attention to enforcement, complex accounting and legal issues.

5

13. Key performance indicators measured: stakeholder perception of service quality, as measured by periodic surveys; taxpayers assistance broken down by service channel (phone, walk-in, writing, visit to taxpayer’s premises); tax returns filed on time and processed without error; refund claims issued within the legal deadline; disputes and complaints resolved within the legal deadline; the average amount of audit revenue assessed; and number of arrears cases closed and amount of arrears collected.

Component II Knowledge and Professional Upgrading of Tax Officials and Stakeholders (Estimated Bank financing € 7 million)

14. This component sought to raise the professionalism, competence, and integrity of tax officials and the knowledge of taxpayers and other stakeholders (e.g., accountants, tax professionals, public officials) through support to CTA to upgrade its HR management and training systems and provision of equipment to the CTA Tax Academy for distance learning in four selected locations. These capacity-building measures complemented ongoing assistance to the CTA from the EU and other donors.

15. Key performance indicators included: new HR policies developed (e.g. staffing and pay policy, including incentive system, staff performance appraisal, and recruitment); HR Division and Training Unit fully staffed and competent; long-term staffing, including recruitment, retention and training plans adopted; establishment of the Tax Academy with regional distance education capability; development of a curriculum and methodologies for training tax officials; percentage of total staff receiving training during the year and share of total staff-days of training delivered; and percentage of trainees successfully trained, as indicated by course certification.

Component III Technological Upgrading for Services, Management Information System, and TIN Implementation Support (Estimated Bank financing € 15 million)

16. This component leveraged technology to improve e-taxpayer services, enhance management decision-making, and facilitate the exchange of tax-related information among government agencies. The component supported CTA’s efforts to: modernize business processes and IT systems including e-tax; develop software applications, data security and business continuity and disaster recovery capacities; and integrate operational dataflow with resource management information. The component also provided initial support to CTA and selected government agencies to implement the TIN through the provision of technical advisory services and equipment.

17. Key performance indicators included: fully staffed ICT unit; use of e-services by different segments of taxpayers, i.e., small, medium and large taxpayers; improved CTA websites and portals as measured by internal and external surveys; implementation of business intelligence and data warehouse, an audit selection system, a risk based collection enforcement system, a case management system and management information system; establishment of a disaster-recovery, business continuity system, and up-to-date security system; and implementation of the TIN system by the CTA and the number of agencies that have adopted the new TIN as primary identifier, as well as the number of new TIN’S issued to physical and legal entities.

Component IV Modernization Management and Project Support (Estimated Bank financing € 2 million)

18. This component supported development and implementation of CTA’s Strategic Plan for CTA Modernization and project management, procurement, financial management, audits, stakeholder consultation, change management, dissemination of materials and project monitoring and evaluation through the provision of technical advisory services, training and equipment.

19. Key performance indicators included: timeliness in completion of the strategic plan; annual strategic planning exercise, including operational business plans for key areas (Human Resources,

6

ICT, Compliance, Taxpayer Services); number of staff members trained in international best practices in tax system modernization and change management, including Project Implementation Unit staff; quality assurance and change management methodologies implemented before effectiveness and used throughout the project life; annual Operational Plans prepared on time in accordance with best practices; timely reporting and fiduciary management in accordance with Bank Guidelines; annual project audit conducted in a timely manner by independent auditors; and semiannual Project Management Reports (PMR) prepared in advance of supervision missions.

1.6 Revised Components

20. By September 2010, government had still not secured the building site for the proposed new CTA headquarter building. After several years of fruitless negotiation with the Zagreb municipality, CTA decided to drop the proposed new building and focus instead on upgrading the existing CTA buildings. At this stage, it also became clear that the proposed disaster-recovery, business continuity (DRBC) facilities that had originally been expected to be part of a government-wide infrastructure upgrading program (outside the scope of the project) was not going to materialize. CTA decided upgrade facilities in Vinkovci to accommodate the DRBC facility and a design and environmental action plan were completed at the time of the restructuring. These changes led to cancellation of half the loan amount and significant changes in the allocations between components.

21. The second restructuring, requested in May 2013, reallocated loan proceeds in line with the authorities’ emerging priorities which were focused on the project’s institutional reform components and amended the Results Framework so as to facilitate measurement of these activities. In this final phase, the project introduced the idea of implementing the Compliance Risk Management System (CRMS). All components and activities of the Project were connected to this core tax administration function which would strengthen voluntary compliance and help to combat tax fraud. Project resources were reallocated to: ensure adequate IT support; strengthen CTA’s corporate governance; and strengthen human resources capacities. A two-year Action Plan for these activities was agreed upon including the reallocation of €7.55 million of the undisbursed balance of €9.05 million and cancellation of €1.5 million of the loan amount. The restructuring changed the distribution of funds with consultant services increasing from 41 to 55 percent, while funding for goods, works and other services declined from 59 to 45 percent largely because of the cancellation of the consolidation of office buildings. These changes were consistent with the shift in project focus from physical infrastructure to institutional reforms. However, a significant amount of loan proceeds remained unallocated due to a new change in CTA leadership and CTA institutional restructuring. The CTA was unable to decide how to use all of these funds, part of which were still not allocated in the project procurement plan at the end of the project.

1.7 Other significant changes

22. The second project restructuring, extended the Loan Closing Date from June 30, 2013 to June 30, 2015. This first project extension was required for the procurement and implementation of the newly introduced CRMS. It would also allow more time to strengthen CTA’s corporate governance structure, build capacity in strategy and policy design in line with EU expectations, streamline the Taxpayer Registry and improve the management of tax arrears. The implementation of the CRMS would also support LTO strengthening and improvement of tax data security through the implementation of the Disaster Recovery / Back-up Center.

23. However, the Project experienced a number of delays due to the discovery of a Roman wall at the disaster-recovery site in Vinkovci. The CTA did not address a request from the conservation services in a timely manner and after several months delay CTA financed the required archeological research and temporary protection of the site. This interrupted civil works which in turn led to delays in the procurement of IT hardware required for the DRBC. The CRMS implementation had

7

also run into contractual difficulties between the CTA and the implementing partner. So in May 2015, as the Project’s closing date approached, the CTA wrote to the Bank requesting an 18-month extension from June 2015 to December 2016 with no other changes expected. At this stage the Project was rated as moderately unsatisfactory. The CTA had experienced another change of leadership and had unilaterally down-sized the project implementation unit.

24. Given the project status, lengthy implementation schedules of other Bank-financed projects in Croatia and the fact that no agreement between the MOF and the CRMS contractor was in site, Bank management decided not to extend the Project further and instead asked government to make arrangements to close the loan on June 30th 2015. Although disappointed with the Bank’s decision, the Government decided to make provisions to continue financing the key unfinished elements of the project including the CRMS system and the disaster-recovery facility in Vinkovci. At the time of writing, both elements appears to be on track to be finalized by the end of 2016.

2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry

25. Project preparation was informed by a comprehensive diagnostic analysis of tax administration but expedited to meet the expectations of the Government of Croatia. The concept review meeting took place in December 2006 and was quickly followed by appraisal in May and Board approval in June 2007. The Ministry of Finance was under pressure to ensure that tax administration met EU standards in the context of Croatia’s expected EU accession. The Bank preparation team included a TTL with extensive operational experience and leading tax administration specialists. However, the accelerated schedule led to the project being prepared with a comprehensive diagnostic work and a dialogue that focused only on CTA and MOF. There was little stakeholder dialogue both internally and externally.

26. Following Bank approval of the Project on June 28, 2007, the project was quickly approved by Parliament on July 11, 2007 to avoid delay arising from upcoming elections in fall. However the project was far from ready for implementation. There were already warning signs of lack of commitment on the part of the beneficiary agency: the Project Preparation Facility (PPF) approved in January 2007 remained unopened for lack of specimen signatures. CTA management had not been sufficiently engaged by the Bank or by the MOF and did not show a strong commitment to the project and its institutional development objectives. CTA did not have a strategy or action plan in place. Nor was there an adequate monitoring framework to track progress. The Bank team did not verify that the proposed site for the CTA headquarter building, accounting for more than half the estimated €50 million cost, was indeed available, beyond the verbal assurances of government officials. The project was approved with six effectiveness conditions, a disbursement condition that required major legislative changes to the Tax Identification Number and five other legal covenants one of which required changes to the legislation governing civil service salaries

2.2 Implementation

27. Project implementation is rated as unsatisfactory. Some of the key factors affecting implementation are discussed below. Implementation issues are also discussed by component in section 3.

28. Inadequate preparation. More than a year was spent following Board approval on issues that should have been completed during preparation: preparing the CTA strategy; locating the site for construction; establishing the M&E framework and baselines; preparation of the strategy and operational manuals; and building internal support for the change management process. As a result, it was only18 months after approval that the project made its first disbursement. The failure to verify that the site for the proposed CTA headquarters had been secured by the CTA proved

8

particularly damaging to the project. Considerable time was spent in vain trying to secure an appropriate site until the government decided to give up on the civil works and restructure the project.

29. Lack of attention to change management. Insufficient attention was paid to winning the hearts and minds of key CTA directors and staff during project preparation. As a result, the project suffered from a lack of champions, lack of depth in ownership, depending largely on the interest of successive CTA directors, some of whom were appointed from outside of the CTA. The project team had not foreseen the changes in leadership of the CTA that took place with changes in government. Subsequent heads of the CTA who had not been involved in the project design were often unwilling to spend money on training and technical assistance and instead focused on the civil works components at the expense of the reform. MoF showed only intermittent interest in supporting the project, usually following prompting by the Bank team during portfolio review meetings. The project governance structure did not provide a mechanism to mobilize external stakeholders who might have been able exert influence on the CTA and MoF.

30. Weak project management capacity. Early implementation relied entirely on civil servants who were not familiar with Bank procedures because the CTA leadership did not want to engage any external consultancy in the Project Implementation Unit (PIU). By the time of the mid-term review, the project had really only begun to start implementation. Project management capacity remained a concern during much of project implementation.

31. Disputes over CRMS implementation. The CRMS contract represented a large and complex IT installation. During procurement CTA raised concerns regarding qualifications and experience of the tax audit specialist presented by the lowest priced qualified bidder. Disqualifying tax audit specialist -0would have had a material effect on the award of contract and would have potentially increased the cost of the CRMS implementation by US$1 million. On the advice of Bank, the CTA reluctantly accepted the lowest cost qualified bidder. Soon after the award of contract, a dispute arose between the CTA and contractor centered on the contractor’s team and the design of the General Audit Software (GAS) module. CTA did not the in-house expertise needed to quickly resolve the issue. Protracted discussions to resolve the dispute periodically broke down. In the end, CTA agreed to remove the GAS module and reduce the value of the original contract. Along with some changes in key personnel on the supplier’s team the issue was resolved after a delay of over a year. This delay led to downgrading of the project to Unsatisfactory and influenced Bank management’s decision not to approve the second project extension.

32. Discovery of ancient archeological site. The data recovery center in Vinkovci is located in an area of archeological significance and so works for the construction of the data recovery center started with a mandatory archaeological excavation. Archeological remains were discovered during excavations, these were considered significant by the Ministry of Culture which instructed the CTA to preserve the archeological site. This required revisions to the architectural design and all the remaining documentation for the data recovery center. However, CTA was extremely slow in responding to the changed situation, resulting in a delay of over a year.

2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization

33. Little attention was paid to monitoring and evaluation, the development of the results framework and the measurement of the PDO at preparation. The PAD provides a very conceptual M&E framework and stated that, “quantitative and qualitative information on some indicators would be developed during the first year of project implementation”. It took a full year to produce an action plan and then several further years to develop baselines and methodologies to track the KPIs and intermediate indicators. The design was both ambitious in its reach (proposing to track a number of areas in detail) and yet very vague in its formulation. The M&E section of the PAD

9

proposed generic “yardsticks” that would “generally include measure of the rate of taxpayers’ compliance… taxpayer services… and staff members’ performance”.

34. There were significant delays in the formulation of a usable results framework despite repeated requests from Bank management to reduce the number and complexity of the early ISRs. The team that took over following Board approval began to streamline the 27 intermediate outcome indicators (IOIs) even before the project was declared effective. However, it was not until 2009 that the new IOIs were agreed upon following a newly appointed CTA project director. The revised IOIs and Baselines for the PDO indicators were only confirmed in 2010 after the first project restructuring. Moreover, while the PDO indicators tracked progress towards outcomes, such as closing the tax gap, they could not be used alone to monitor project implementation. This situation required more detailed and relevant intermediate outcome indicators.

35. There is very little evidence that the M&E framework was actually used as a management tool to guide project activities, resource allocations or supervision by CTA or by Project Management. CTA did not assume ownership of the monitoring arrangements and chose not invest in updating project indicators towards project closing. Information for the PDO indicators and IOIs is available only through 2013 or 2014 and in some cases only through 2012. Notably, CTA did not undertake the tax payer surveys at project closing in 2015 despite requests from the Bank team.

2.4 Safeguard and Fiduciary Compliance

36. Early in the project life in 2008-2010 the project had an MS rating for FM. This was largely due to the project’s poor performance rather than specific FM issues. The FMRs were always delivered on time and deemed satisfactory by the Bank. Given the small amount of disbursement that took place compared to initial estimates it appears that there was little pressure on the FM function.

37. Procurement performance was more problematic, often highlighted with a “Moderately Unsatisfactory” rating in consecutive ISRs. The project anticipated a full-time procurement manager but this position was not always filled, contributing to delays in procurement especially in the early years. It was not until January 2012 that a new full-time procurement specialist was brought on board at which point the pace of procurement accelerated and the rating was upgraded. Hiring the senior IT consultant also helped the procurement manager to prepare and launch the bidding documents required for the CRMS implementation but once the consultant left, CTA lacked the technical expertise to effectively deal with the contract management issues that arose during implementation.

38. The environmental assessment as per OP, BP 4.01 was satisfactory as was overall safeguard compliance throughout the project lifetime. Following the restructuring in 2010, when the building component was dropped, the associated safeguard issues also became less pronounced. When the project discovered ancient ruins at the DRBC site, all the correct national safeguards and procedures were followed although with a significant delay, allowing the CTA to protect this ancient heritage. An amendment to the contract was developed in accordance with ministry of culture requirements. There were no deviations or waivers from the Bank safeguards/ fiduciary policy procedures.

2.5 Post-completion Operation/Next Phase

39. A dispute with the CRMS supplier and the discovery of ancient ruins at the DRBC site, led to significant implementation delays which meant that the project could not be completed within the approved timeframe. The Bank rejected a request for a further extension and the CTA proceeded to put in place a plan to finance and implement the remaining project components from its own resources. The CTA secured a budget from the MoF (approximately $2.6 million) for the remainder of 2015 and 2016 (although at time of writing the 2016 budget has not been approved)

10

in order to complete the CRMS and DRBC implementations and purchase the relevant hardware. The project implementation unit was down-sized as the procurement and separate FM function were no longer required. These functions have been effectively absorbed by regular CTA staff.

40. With a committed technical head leading the CTA and strong support from MoF, the outstanding dispute with the CRMS supplier was resolved by down-sizing the contract and removing the General Audit Support (GAS) module. The latter will be delivered through in-house ICT provider APIS as a separate activity. The new contract amendment has extended the delivery date to June 2016.

41. Similarly the DRBC project in Vinkovci has also resumed with a more limited scope and extended duration to September 2016. MoF agreed to finance implementation and the required hardware, including the communications network and consolidation of the data center which could not be procured until the building refurbishment itself was complete.

42. Early in 2015, government made a decision to postpone the introduction of a property tax until after the land registration and cadaster data are fully harmonized and updated. Therefore the remaining technical assistance from the project that was originally envisioned was no longer needed.

3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation

Relevance of Objectives: High

43. The PDO – which sought improvements in efficiency, taxpayer services and tax compliance – was aligned with the GOC program that sought to create the conditions for EU accession by improving the investment climate, fiscal consolidation and strengthening governance. The PDO was relevant to the World Bank’s Country Partnership Strategies which were focused on EU accession and convergence. The PDO was aligned with two of the three objectives of the Country Assistance Strategy, (November 24th 2004) at time of loan approval, namely upgrading fiscal management systems and reducing fiscal vulnerabilities and modernizing the public sector. The PDO was aligned with the subsequent Country Partnership Strategy for FY2009-2012, extended to FY2013, which sought to support completion of Croatia’s EU accession, contributing to two of the four pillars for the Bank program; sustaining macroeconomic stability; and strengthening private sector-led growth and accelerating convergence with the EU. The CPS for FY2014-2017 coincided with Croatia’s EU membership and sought to assist Croatia's continued convergence with the EU. The Project was aligned with the pillar that sought to support Croatia’s fiscal adjustment.

Relevance of Design: Modest

44. The original design was only modestly relevant as it focused most resources on civil works, but due to a myriad of factors that played out during implementation, the project managed to focus on important reform aspects despite the poor initial design.

45. Choice of instrument. A longer preparation phase financed with an operational PPF and/or the development of the project as an Adaptable Program Loan (APL) could have been much more appropriate for such a project that required a long lead time for IT procurements, civil works and careful business process re-engineering. However, this possibility does not appear to have been adequately considered by the design team and was not supported by the GOC.

46. Selection of components. The focus on construction appears to have been misplaced since many of the key reforms were achieved without the recourse to a new building. The proposed building detracted from the focus on the reform elements until it was dropped. Only then did the

11

CTA focus on institutional capacity and IT systems, which were the two issues important for EU accession, come to the fore.

47. Change management. There was no change management element included in the design due to resistance to change management technical assistance by the CTA leadership despite the fact that the project was attempting to accomplish significant business process re-engineering and structural and organizational changes.

Project implementation: Modest

48. Changes during implementation allowed the project to become increasingly more relevant to the original objectives. The authorities postponed work on the organizational realignment of tax offices and upgrading of IT systems pending the results of a comprehensive analysis of business processes. The first restructuring removed the construction of the new facility and CTA management decided to use available loan financing for developing and implementing a Compliance Risk Management System (CRMS).

49. Other activities under component 1 (LTO establishment), component 2 (professional training for tax officials) and component 3 (development and implementation of e-audit, e-filling and upgrade of the CTA website) proceeded without major delays. The Project activities complemented technical assistance on risk analysis financed by the Netherlands and the EU-financed support on VAT.

50. Frequent changes in the CTA and PIU management and slow decision-making mechanisms in the CTA contributed to road blocks at different stages of the Project. The Bank addressed holdups in the Project by convening a workshop with CTA management in 2014. Such an event to resolve road blocks should have been convened at earlier stages of the Project.

3.2 Achievement of Project Development Objectives

51. The PDO is clearly defined and activities under each subcomponent are linked with the achievement of each objective. Annex 2 on the Assessment of Achievements by Objectives reconstructs the Results Framework in the PAD. This section reviews progress against the components of the PDO drawing on information from a variety of sources including: Bank aide-memoires and ISRs, CTA data and surveys, the borrower completion report, Doing Business Surveys, the IOTA data and a stakeholder workshop.

Improving Efficiency of Tax Administration: Modest

52. Partial achievement of the PDO indicator and two out of four intermediate results indicators determine this rating as modest even though data from external sources suggests that the CTA made progress in improving its overall efficiency. It is not possible to draw strong conclusions regarding performance of this objective because the CTA did not undertake required surveys and so lacks data on efficiency at the end of the project. A more detailed discussion of the ratings is provided below.

53. The PDO indicator to measure an improvement in operational efficiency was the perception of tax officers regarding the terms of employment and working environment as measured by periodic surveys. Since the CTA had not undertaken such surveys before and there was no methodology in place, there was no baseline for this indicator at appraisal. The CTA conducted the first and second staff surveys based on the developed methodology in 2010 and 2012. The overall score for the perception of tax officers concerning the terms of employment and working environment was 3.05 and 3.14 on a five point scale (where five is the highest rating), respectively. This represented an improvement of 3 percent, which fell short of the target of 10 percent but was measured three years prior to the end of the project. The average score for rating

12

the organizational structure and management increased, from 2.6 to 2.9. However, the share of completed responses declined more than by half, from 35.4 percent to 15.4 percent, in the second survey. CTA management note that the lack of communication of the first survey results was the main reason for the reduction in the response rate in the second survey. The CTA did not carry out the staff survey in the end of the Project and thus it is impossible to assess the most recent progress for this indicator.

Table 1: CTA staff surveys 2010 and 2012 PDO CTA Staff Survey

2010 CTA Staff Survey

2012 Perception of tax officers regarding the terms of employment and working environment

3.05

3.14

54. The CTA implemented important steps to improve the efficiency of tax administration by enhancing systems and building capacity. A fully fledged LTO was established and criteria used for the identification of large tax-payers (LTPs). The national level LTO in Zagreb started its operations in 2012. As of October 2015, the LTO monitors 643 large taxpayers that employ 30 percent of the workforce in Croatian real sector and pay nearly 40 percent of all the taxes and duties levied. The LTO’s staff includes 130 positions, of which 82 percent are currently filled. Since the LTO became operational, the CTA has collected a stable share of revenue from large taxpayers. Staff in the LTO have developed industry specific skills and can deal with risks in real time. Communication with LTPs is now almost entirely through electronic means and LTPs report that most tax related issues and queries are dealt with quickly.

Table 2: Share of large taxpayers in total tax, 2012-2014 2102 2013 2014

Share of Large Taxpayers in Total Tax/ Contribution Payments

40.9 40.4 39.6

55. In the course of Project implementation, the CTA consolidated the number of regional and local tax offices by more than 50 percent. The CTA also introduced the functional realignment of the organizational and staffing arrangements covering the whole CTA network according to six functions: registration, tax assessment and processing of tax returns, taxpayer services, tax audit, enforcement of tax collection and dispute resolution. Tax officers now perform specialized functions rather than all tasks for a certain number of taxpayers.

Table 3: CTA reorganization 2007 2015 Regional Offices 20 7 including LTO -- 1 Local Offices 122 57 including LTO -- 3

56. Following the assessment of the CTA’s business processes, the Project focused on supporting the development of a risk based approach in tax administration and the redeployment of certain number of tax inspectors to audit, as a high value added task, in regional offices and the LTO. As of July 2015, the share of the total CTA staff assigned to audit functions reached 20.4 percent which represents about a 1 percentage point improvement.

13

Table 4: CTA total staff numbers in audit Year Total CTA Staff Involved in Audit Percentage in Audit

2007 4,240 828 19.5%

2014 4,227 841 19.9%

2015 (July) 4,119 841 20.4%

57. In line with OECD experience suggesting that the primary focus of tax audit are large taxpayers, the Project put considerable effort into increasing the share of staff conducting audits in the LTO. The data for an intermediate results indicator measuring the percentage of the LTO staff in audits shows a steady increase. At the same time, the percentage of the LTO staff in the audit function still falls short of 65 percent target due to the difficulties in attracting highly qualified skills because of significantly lower pay in comparison with the private sector.

Table 5: CTA audit staff in LTO Intermediate Results Indicator

Year Percentage LTO Staff in Audit 2007 49%

2014 51%

2015 (October) 59%

58. The Project also contributed to improving the efficiency of tax audits by supporting the implementation of an e-audit tool. The Project financed preparation of a Manual for the Implementation of e-Audit, procurement of an ACL application and training of 52 auditors in how to use it. The LTO’s auditors obtained 50 percent of ACL licenses. From May 2014 to September 2015, the CTA carried out 154 e-audits. As a result of creating specialized audit staff and employing the e-audit tool, the efficiency of tax audits substantially increased, particularly those carried out by the LTO staff.

Table 6: Number of LTO audits 2013-2014

Year Number of LTO’s

Audits Value of Total LTO’s Audit Assessments (million HRK)

2013 149 463

2014 98 525

59. Similarly, the efficiency of the LTO’s tax auditors measured as an annual average assessments per auditor also improved. However, the overall efficiency of tax auditors that was an intermediate results indicator falls short of reaching the 5 percent improvement target.

Table 7: Average assessments per auditor, selected years 2007-2014 Intermediate Results Indicator

Year Annual Average Assessments per

Auditor (Euro)

Annual Average Assessments per LTO’s Auditor

(Euro) 2007 348,075 --

2013 302,102 939,227

2014 348,707 1,050,194

14

60. The Project supported the operationalization of a set of strategies developed by the CTA: Strategy of the Tax Administration 2011-2015; Strategy for Human Resource Management in Tax Administration for 2010-2015; and the Training Strategy of the Tax Administration for 2012-2015. The Project made a substantial investment in the implementation of the training strategy which sought to enhance operational efficiency of the CTA by upgrading the knowledge and professional skills of tax officials. The strategy aimed to improve timely access to information, reduce the cost of training and develop a culture of continuous upgrading of skills to remain at current best practice. In order to meet these objectives the CTA developed the first distance and e-learning programs within the Croatian civil service. The courses were developed following a comprehensive training needs analysis. An on-line repository of opinions and guidelines was also created in order to harmonize tax decisions and treatments. An on-line forum was established and is being used regularly to share information and seek and provide guidance to tax offices throughout the country.

61. The e-learning financed by the Project included: 10 training courses with tests, three questionnaires, 184 education materials and a learning platform including 86 topics and 2,546 registered employees. Subjects completed included; anticorruption, e-Tax, VAT, value-for-money purchasing, information security, effective management, specification of unmatched payments, forced collection and enforcement. The training program comprised: 11 IT programs, 37 specialized programs and foreign language courses offered for two terms.

Table 8: CTA training outputs Staff Received IT Programs 17 IT specialists Specialized Programs 1884 tax officials Foreign Language 525 tax officials

62. The intermediate results indicators reflected good progress in achievement of targets for staff training and development of e-learning modules. The Project exceeded the target for hours of training for staff per year and was right on target for the number of e-learning modules developed.

Table 9: CTA training received (hours per staff) Intermediate Results Indicators

Year e-Learning Modules

Developed and Implemented Hours of Training per

Staff 2011 -- 8

2013 3 29.95

2014 10 21.97

63. Implementation of structural reforms, investments in IT systems and capacity building within the CTA supported by the Project contributed to a significant improvement in the overall efficiency of CTA’s operations. The CTA’s average cost per employee declined by 5.0 percent, representing administrative cost savings. Moreover, the CTA achieved a substantial result in improving its operational cost-effectiveness. Based on the CTA data which includes both tax revenue and social contributions and associated expenditures, the total expenditure of the CTA as a share of revenue collected declined by 8.3 percent.

Table 10: Revenue collected and average cost per employees

Year CTA’s Total Expenditure as

Percentage of Revenue Collected Average Cost per Employee

(Euro) 2007 0.72 25,259

2014 0.66 23,985

15

64. However, data from the Intra‑European Organization of Tax Administrations (IOTA) suggests that CTA’s efficiency has declined between 2010 and 2012. The ratio of total CTA’s expenditure to revenue collected increased by 10 percent over this period. But the IOTA methodology excludes social contributions and costs associated with its collection. The lack of IOTA data for the recent years makes it difficult to come to a definite conclusion about the performance for this indicator.

Table 11: CTA revenue performance 2010-2012

Year CTA’s Total Expenditure as

Percentage of Revenue Collected 2010 0.8

2011 0.8

2012 0.88 Source: IOTA http://www.iota-tax.org/iota-members/croatia.html

Improving Taxpayer Services: Substantial

65. The Project exceeded a target for the PDO indicator in 2012/2013. However, performance against this objective is only rated substantial because, owing to CTA’s failure to conduct the required taxpayers’ survey, there is no end of project data .

66. The PDO indicator used to monitor improvements in taxpayer services was, “the perception of taxpayers regarding the quality of services provided by the CTA and its integrity”. To measure this indicator, the CTA carried out taxpayer surveys in 2010 and 2012/2013 including 190 citizens (14 percent), 494 self-employed (36 percent) and 678 companies (50 percent). The results of the second survey showed a substantial improvement in taxpayers’ perception concerning the quality of services provided by the CTA and its integrity. An index measuring, on average, taxpayers’ views with respect to four dimensions, such as a tax system, tax offices, tax administration officials and a tax administration system, increased from 2.66 in 2010 to 3.19 in 2012/2013. This represented a 20 percent improvement significantly exceeding the 5 percent target. Unfortunately, it is not possible to assess the end of project performance for the PDO indicator because the CTA did not conduct the taxpayer survey at the end of the Project as planned.

Table 12: Perceptions of CTA quality (1 to 5 scale) PDO Taxpayer Survey

2010 Taxpayer Survey

2013 Perception of Taxpayers Regarding the CTA’s Quality of Services and Integrity

2.66 3.19

Other results indicators Perception of Taxpayers about Tax Officials 2.95 3.55 Perception of Taxpayers about a Tax System 2.09 2.38

67. Taxpayers expressed the most positive opinion about tax administration officials in the 2012/2013 survey, an average score of 3.55 compared to 2.95 in the previous survey. The tax system received a lower rating at 2.38 in 2013 as compared to 2.09 in 2010.

68. These positive results can be attributed to a number of Project activities that led to improvements in taxpayer services including: the overhaul of the CTA website that now provides regularly provides updated information in a user-friendly format; establishment of the intranet and forums on specific topics and issues that enabled effective and timely responses to taxpayers’ inquires; establishment of a dedicated phone line and email for taxpayers’ inquiries; periodically

16

updated publications for taxpayers; introduction of a legal provision that granted the CTA authority to issue legally binding opinions on tax matters where existing regulations appear ambiguous; and real-time response to inquiries for large taxpayers implemented by the LTO. In addition, the professional training for tax officials that the Project supported were important factors in strengthening taxpayer services and perceptions of the professionalism of tax officials. In January 2016, the CTA is planning to launch a new service for taxpayers—pre-filing of tax returns.

69. CTA also made a significant effort to reach out to external and internal stakeholders through its communication strategy, developed and implemented with support of the Project. On 20 June 2013, the CTA launched a new website that was completely reconstructed and adapted to the needs of users. It provides information on the rights and obligations of taxpayers by taxpayer segments (citizens, self-employed, companies, associations, non-residents), tax laws and regulations and provides an overview of the entire Croatian tax system, payment accounts and rates of taxes. The website has several functional features, including the possibility of obtaining information on a taxpayer identification number (OIB in Croatian), filing tax returns electronically (e-Tax), verification of VAT payers, verification of fiscal accounts, lists of debtors and other actualities. In addition, the CTA’s website publishes proposals for amendments to tax regulations with an explanation of the changes involved. The Charter on Cooperation of the Tax Administration and Taxpayers is also published on the website. The most important information in Croatian is also available in English. The website is regularly updated and there is a documented procedure for updating it. The number of website visits has surged dramatically since its launch.

Table 13: CTA website visits 2012-2015 Year Website Visits

2012 (old website) 3,164,106

2013 June-December 2,752,375

2014 8,717,066

2015 January - November 26,495,390

Improving Taxpayer Compliance: Modest

70. The Project achieved the two PDO indicators but because one out of two intermediate results indicator was not achieved, because of concerns about the quality of data for tax gap estimates and because of the lack of the most recent taxpayer survey for the end of the Project, the ICR rates progress against this objective as modest..

71. The PDO indicators measuring improvements in taxpayer compliance comprised the reduction in tax gap and taxpayer compliance costs. The tax gap is estimated using VAT gap as a proxy. It is based on the EU methodology employing a macroeconomic approach. Estimates generated by the Croatian Bureau of Statistics using national accounts data show that the CTA made a steady progress in decreasing the tax gap beyond the targeted 5 percent improvement. However, these estimates are not comparable with the original baseline generated using micro-economic methods and so it is not possible to arrive at definitive conclusion as regards performance.

Table 14: PDO Indicator: Tax Gap 2011-2013 PDO 2011 2012 2013 Tax (VAT) Gap Estimated Using EU Methodology (in Percent GDP)

2.05 1.81 1.52

72. Taxpayer compliance costs are measured as an average annual cost of tax compliance for legal entities and assessed using the taxpayer surveys. The results of the taxpayer surveys in 2010 and 2012/2013 showed that the average cost of compliance for legal entities declined from HRK

17

14,495 to HRK 11,977 during this period, representing a 17.4 percent reduction and significantly exceeding the end target of HRK 13,770. However, more recent data for this indicator is not available because the CTA did not carry out a taxpayer survey at the end of the Project.

Table 15: PDO Indicator: Taxpayer compliance costs PDO Taxpayer Survey

2010 Taxpayer Survey

2013 Taxpayer Compliance Costs -- Average Annual Tax Compliance Costs for Legal Entities (HRK)

14,495

11,977

73. Doing Business surveys show a similar positive trend in the number of tax payments per year and the number of hours spent over the life of the project. There was a slight deterioration in both indicators following Croatia’s EU accession owing to compliance with EU regulations concerning VAT. While the Project certainly contributed to these results they cannot be fully attributed to the Project.

Table 16: Doing business data

Year Paying Taxes

Payments (Number per Year)

Time (Hours per Year)

2007 28 232

2012 18 196

2014 19 208 Source: Doing Business Report 2008, 2013 and Doing Business/Paying Taxes Report 2015

74. The intermediate results indicator measuring stop filers for the medium and small taxpayer segment for VAT shows an upward rather than a downward trend and thus falls short of a target of 2 percent. These results need to be treated with caution. When Croatia experienced a prolonged economic recession in the aftermath of the 2008 global financial and economic crisis, the proportion of stop filers increased rather than decreased because many small and medium-size businesses filed for bankruptcy.

Table 17: Stop-filers 2011-2014 Intermediate Results Indicator

Year Percentage Stop Filers for Medium and Small

Taxpayer Segment (VAT) 2011 2.81%

2013 4.46%

2014 3.61%

75. The Project supported several important reforms that contributed to improvements in tax compliance, measured by the narrowing tax gap and reduced tax compliance costs for taxpayers as evidenced by the number of payments and financial cost involved. These included the implementation of a taxpayer identification number (TIN), which is called a physical and legal person identification number (OIB) in Croatia and e-filing and the development of a risk management information system CRMS.

76. The project financed the development and implementation of an upgraded e-filing system which was introduced together with the launch of the new CTA website. Legal entities were the first to reap benefits of the e-filing. The e-filing rates for tax declarations by legal entities increased dramatically and reached over 99 percent, well above the 90 percent target.

18

Table 18: Electronic filing 2007-2014

Intermediate Results Indicator

Year Percentage of Total Declarations by Legal

Entities Filed Electronically (VAT Returns) 2007 3%

2013 97.7%

2014 99.2%

77. The CTA introduced the OIB in January 2011. Since then it has become the only identifier for all legal and natural persons and is used by all public agencies. The OIB enabled the uniformity of tax registration and electronic integration, as well as wide application for various purposes, such as tax returns, payments, identification documents and others. It allows for tracking cash transactions and helped to combat tax evasion and the gray economy. The OIB brought about multiple benefits beyond tax administration and the public sector. It not only accelerated government business processes and generated administrative cost savings but also resulted in reduced transaction costs for citizens and the private sector.

78. As noted earlier, the CTA progress in developing the CRMS was slow. The Project financed a significant share of procurement and development of the CRMS that was intended to strengthen tax compliance and reduce tax evasion. Since the Project was not extended beyond June 2015, and given the importance of this system for effective and efficient tax administration, the Government has allocated funding to complete the CRMS in the draft 2016 budget.

3.3 Efficiency: Substantial

79. Most revenue administration projects generate substantial financial returns which are realized through increased revenues and economic returns which are realized through efficiency gains for the tax administration and taxpayers. RAMP contributed to efficiency gains such as reduced administrative costs for CTA and, much more significantly, decreased compliance costs for taxpayers. Using relatively conservative assumptions, the economic analysis indicates that RAMP’s economic efficiency was substantial: with an economic IRR of 259 percent and NPV of € 72.7 million in real terms (see Annex 3). However, financial returns are more difficult to assess because the revenue increases observed during a prolonged period of negative economic growth can also be attributed to revenue policy measures. These include: an increase in the general VAT rate from 23 to 25 percent in February 2012; abolition of the zero VAT rate and introduction of a 5 percent rate for basic food, pharmaceuticals and books in 2013; increase of the reduced VAT rate from 10 to 13 percent social and health sector contribution to 15 percent in 2014. Unfortunately, it is not possible distinguish tax administration and tax policy effects with the data available.

Table 19: Spending by Component

Components Appraisal Estimate

(US$ million)

Actual/Latest Estimate

(US$ million)

Percentage of Appraisal

A. Organizational consolidation and functional restructuring

35.36 7.41 35.75

B. Knowledge and professional development 9.52 0.57 10.29C. Technological improvement 20.40 7.69 64.26D. Modernization management and support 2.72 0.10 6.25

Total Cost 68.00 15.78 23.21

19

80. A summary of the original approved cost allocations and the actual outturn is presented below. While the project implementation period was increased by two additional years, the project only disbursed 17.44 percent of the original allocation, indicating that the allocation efficiency for the project as a whole was very low. However, the results vary by component. The low efficiency rating is largely due the performance of Component 1 where the CTA was unable to procure a suitable site for the construction of the proposed HQ building. Component 3 on technological improvement disbursed just over half the original allocation. This component was also delayed and will continue implementation using CTA funds beyond the lifetime of the project.

3.4 Justification of Overall Outcome Rating: Moderately Satisfactory

Table 20: Assessment of the Overall Project Outcome Rating

Relevance of Efficacy by Objective Efficiency Outcome

Objectives Design 1 2 3

High Modest Modest Substantial Modest Substantial Moderately Unsatisfactory

81. The PDO remains relevant in the context of the country challenges and priorities. Three of the four PDO indicators were achieved. The PDO indicators on reducing the tax gap, increasing tax compliance and reducing the costs of compliance appear to have been met, however, there are questions as to the methodology used to measure the tax gap and the lack of survey information required for compliance cost estimates at the end of the project. Although the quality of taxpayer services was increased and taxpayer perceptions on CTA quality increased, the increase did not meet the initial target in 2013 when the CTA undertook its most recent taxpayer survey.

82. Four of the seven intermediate results indicators were achieved (see Annex 2). The intermediate results indicators that were not achieved are: a reduction in the number of stop-filers, which can be attributed to the deterioration in the economic situation and consequent increase in the numbers of bankruptcies among tax payers; the average annual assessments by CTA staff which again is partly due to the economic downturn; and the number of auditors as a percentage of total CTA staff – CTA management explained that it was proving very difficult for to attract qualified auditors at government salaries. Government did not accept a proposal to de-link various positions from the broader civil service pay-scale.

83. Despite the progress against a majority of the PDO and intermediate result indicators, the Project did not complete outputs that were the focus of activities in the final two years: the DRBC facility and the CRMS system. The CRMS in particular is critical to the achievement of project objectives. Both these activities will now be completed by the authorities using their own funds.

3.5 Overarching Themes, Other Outcomes and Impacts

(a) Poverty Impacts, Gender Aspects, and Social Development

84. The principal poverty and social development impacts of tax administration reforms are indirect, realized through increases in revenue and thereby the potential to increase social spending. Unfortunately, the Project coincided with a prolonged recession in Croatia and so these impacts have been muted by an overall decline in economic activity. Nonetheless, while the value of revenues collected remained relatively constant between 2009 and 2014, improvements in tax administration – most attributable to the Project –allowed the authorities to increase revenues as a

20

share of GDP from 40.7 percent in 2007 to 43.3 percent in 20151. As a result of improved revenue mobilization, the Croatian Government was able to sustain its social spending especially its spending on health care which at 8 percent of GDP is one of the highest in the EU2.

85. The Project has also contributed to improvements in the governance and efficiency of the tax system both of which bring benefits to the private sector and through economic activity, ultimately benefit the poor. The reduction in the VAT tax gap by 0.5 percent of GDP between 2011 and 2013 signals a reduction in tax evasion and efficiency of recovery. The implementation of the CRMS system once completed will also serve to combat tax evasion and tax fraud. On the production side, the reduction in the number of hours spent completing tax forms and improvements in tax services should reduce costs and increase efficiency in the private sector. An estimate of the benefits of such a change is included in the cost benefit analysis presented in Annex 3 below.

(b) Institutional Change/Strengthening

86. The project contributed to significant institutional changes, notably the creation of a fully-fledged LTO. This significantly changes the distribution of functional responsibilities within the CTA from the situation at appraisal and brings the institutional structure in line with most modern tax administrations. Even more important are the change in mindset, moving towards a model of voluntary compliance and a service-based culture promoted through the project. These changes have been supported through the functional specialization and training of CTA staff and their commitment to electronic communications, as well as the move to e-filing and upgrading of the website and interactions with taxpayers.

(c) Other Unintended Outcomes and Impacts (positive or negative)

87. There are no other unintended outcomes or impacts of the project.

3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops

88. A stakeholder workshop was held at the World Bank office in Zagreb on October 27th 2015. A summary of the proceedings is presented in Annex 6. The CTA emphasized their firm commitment to finalize the remaining ongoing project activities especially the DRBC facility and the CRMS implementation. Feedback from the private sector was extremely positive, especially as regards the new website and upgrading of the LTO, confirming the results of the stakeholder survey.

4. Assessment of Risk to Development Outcome Rating: Negligible

89. The Project’s achievements improvements in efficiency, taxpayer services and tax compliance can be sustained by the CTA’s systems and respond to demand from tax payers. The rise in electronic filing from 3 percent in 2007 to more than 99 percent in 2015 was achieved because the system provides real benefits to taxpayers: those filing taxes on-line are unlikely to go back to paper filing. The CTA website is visited more than 26 million times each year (making it one of the government’s most popular sites) and this is likely to increase in the future. The decrease in compliance costs as assessed by the taxpayer surveys from 14,495 kn in 2010 to 11,977 kn in 2015 also appears to well established. Doing business figures confirm that the number of separate tax payments has more than halved from 40 in 2007 to just 19 in 2015 while the time taken to complete a tax return has also dropped from 232 hours in 2007 to 196 in 2014. Similarly the establishment of the LTO and subsequent improvement of taxpayers’ services is entirely

1 IMF Article IV staff reports, 2007 – 2015. 2 World Bank Country Partnership Strategy 2014-2017.

21

sustainable as the institutional arrangements and dedicated and trained sector based staff are now in place and operating effectively. The routines established with support of the project have been embedded in the regulatory framework, internal operating procedures and expectations of CTA management and taxpayers.

90. CTA’s capacity to support and sustain the reforms is being further strengthened through interventions supported by other funding agencies. The EU IPA financed project for the HR management system in the tax administration, starting in late 2015, will build on the capacity building and strategy reforms established under the Project. Another EU funded initiative, starting in 2016, will upgrade the e-learning system, acquiring equipment for remote classrooms to ensure that the implementation of CTA’s inter-sectoral strategies meets all the EU requirements.

5. Assessment of Bank and Borrower Performance 5.1 Bank Performance

(a) Bank Performance in Ensuring Quality at Entry: Moderately Unsatisfactory

91. The Project was aligned with the strategic priorities of the Government and the World Bank as defined in the Country Assistance Strategy presented to the Board in November 2004 and the CAS Progress Report of May 2007. The Project focused on tax administration challenges that were critical for Croatia’s successful EU accession, integration and convergence. The project design was informed by diagnostics and recent analytical work.

92. However, there were four major shortcomings in project preparation that adversely affected project implementation. The first was the failure to secure the site for the CTA headquarter building. This should have been a pre-requisite for Bank financing as it is in major Bank-financed infrastructure projects. The proposed site was identified but it was never secured for CTA use. This caused a lot of wasted effort during the first years of implementation as the CTA searched for other locations and ultimately led to the cancellation of a large part of the project funds. Second, the Project’s logical framework was weak: the intermediate outcome indicators were not aligned with the project components or with the PDO and PDO indictors. As a result it was extremely difficult to guide the project and track progress. Third, project preparation was incomplete, with a number of critical issues left pending as effectiveness conditions and legal covenants. Again, this contributed to start-up delays. Fourth and by far the most important, project preparation had failed to secure buy in and commitment from the CTA management and staff. This was particularly true of the institutional reforms and capacity building components of the project. Failure to activate the PPF advance during preparation was an early signal of weak ownership. These problems were aggravated following the elections in November 2011 which brought in new MoF leadership and changes in CTA management.

(b) Quality of Supervision: Moderately Unsatisfactory

93. The project was supervised regularly with six monthly ISRs filed on time. ISRs documented project progress and highlighted the main implementation challenges, though ISR ratings were certainly overoptimistic in the early years: the project was rated moderately satisfactory despite disbursing a little over $1 million a year between June 2009 and June 2013. Task teams included an appropriate mix of operational and tax administration specialists. Once it was clear that the project was slow to take off, the Bank intensified its mission support to a quarterly cycle bringing in additional tax administration expertise to strengthen the dialogue and try to develop CTA ownership. The task teams received adequate support from the sector and country managers.

94. Unfortunately, the effectiveness of supervision was seriously undermined by changes in leadership in the Bank’s team at critical points in project implementation. Immediately after project

22

approval the task team that prepared the project was replaced with a new task team leader with little experience in the Bank. Subsequently, the project had five different task team leaders throughout its lifetime. The Borrower has argued that this lack of continuity on the Bank side contributed to delays in taking action to remedy problems in project design and project implementation. The ICR team concurs with this assessment. High turnover could have been mitigated by involving and empowering country office staff to a more active role during supervision and/or have a co-TTL based in the field.

95. While the Bank team responsible for implementation initiated work on revising the project results framework even before the project became effective, project restructuring was not completed until September 2010, almost three years after approval. Moreover, the decision to restructure at this stage was driven by the authorities’ cancellation of the civil works component. The Bank team was more proactive in the second restructuring, requested in May 2013. This restructuring responded to a change in CTA management and successfully reoriented the Project towards its institutional reform objectives.

96. The proactivity and effectiveness of Bank supervision improved markedly in the final two years of the project. Project ratings were more realistic at this stage. ISRs focused on the key implementation challenges. The Bank team played a constructive role in trying to resolve differences between the CTA and the CRMS contractor. Unfortunately, resolution of this disagreement came too late to allow the completion of this output before the project closed.

97. The fiduciary and safeguards supervision of the project was satisfactory throughout. The fiduciary team supported the PIU through the FM and procurement challenges. A local environmental specialist was always on hand to address safeguards issues that arose during the initial search for a suitable site as well as the assessments required for refurbishment of CTA buildings in Split and Rijeka.

(c) Rating for Overall Bank Performance: Moderately Unsatisfactory

98. Better planning and an investment in building the consensus for reform would have helped secure ownership and commitment to the project from CTA management. More work on the results framework would have helped to focus attention on the institutional reforms aspects which only came following the project restructurings. Project restructuring was slow to take place. Frequent changes in Bank team leadership hindered efforts to build a strong relationship with CTA and tackle implementation problems. The quality of supervision improved markedly in the run up to and following the second project restructuring warranting the Moderately Unsatisfactory rating.

5.2 Borrower Performance

(a) Government Performance: Moderately Unsatisfactory

99. At the time of Project preparation, MoF was the principal advocate for tax administration reform which was seen as an essential to a successful EU accession process. The challenge facing MoF was to generate support for reform within the CTA, the organization that would ultimately be responsible for implementing the Project. The new HQ building for the CTA was seen as a solution, providing an incentive for CTA management and staff to go through with the organizational reforms. MOF successfully drove the legislative reforms required for the implementation of the TIN but was not able to push through the HR reforms that would among other things decouple CTA salaries from the civil service pay scales.

100. Attention focused on the civil works component during the early stages of Project implementation at the expense of the more fundamental institutional reforms. Then, in 2009, when it appeared that the issues around securing a suitable site of the building were being resolved, MoF balked at the requirement of funding as the global financial crisis was raging and the ministry was

23

focused on cost-cutting and fiscal consolidation. MoF eventually requested a cancellation of the civil works component. From this point forward, MoF played a limited role in project implementation. Implementation delays were brought to the attention of MoF during the regular joint portfolio review (JPR) meetings. Although agreements were reached and a feasible time-bound action plan was developed in 2011, these agreements were not respected by CTA or enforced by the MoF. The Project finally picked up pace in 2013, largely due to a committed new DG at CTA rather than impetus from MoF.

(b) Implementing Agency or Agencies Performance: Moderately Unsatisfactory

101. Initially CTA leadership, equated the project with the new building and neglected the institutional reforms especially the need to create a fully-fledged LTO. CTA management had a number of competing demands and lacked the capacity and interest to implement the institutional reforms aspects in a timely manner.

102. The CTA director general changed several times following changes in Government. Several DGs did not come from a tax administration background and had limited interest in the project. This resulted in a situation where the technical cadres of CTA had to convince their political leaders of the value of the reform process. This process took two years until CTA management accepted the major project initiatives such as the LTO, the business process re-engineering and the process to launch the TIN. Still progress was slow. Negligible disbursements between 2009 and 2013 demonstrated the CTA’s limited commitment to the Project. Only in the last two years of the project did CTA management increasingly focus on the reform aspects of the project, creating the LTO and launching the procurement for the CRMS IT system and DRBC civil works.

103. Although the CTA established a functioning project unit it lacked adequate capacity for most of the project lifetime. The Bank requested a full time general procurement consultant and additional support for IT and civil works procurements. These staff were eventually hired following the Bank’s repeated requests but were subsequently let go resulting in further project delays. The CTA also unilaterally down-sized the PIU while requesting a third project restructuring and extension, despite the Bank’s request to maintain the PIU and the implementation momentum.

(c) Rating for Overall Borrower Performance: Moderately Unsatisfactory

104. At its heart, RAMP was a change management project that needed to overcome resistance to reforms from within the CTA as well as wider stakeholders. It needed a leadership with a clear strategic vision and a change management process that included all levels of CTA management. Strong ownership materialized only in the last 2-3 years of the project with committed DG’s from a tax administration background taking leadership at the CTA. Bank management, impressed with this turnaround, were persuaded to extend and restructure the project despite four years of non-performance. This decision resulted in a significant improvement in implementation and disbursements following the second restructuring in 2013. Unfortunately due to disagreements with the CRMS supplier and the discovery of ancient ruins at the DRBC site, the CTA was not able to implement the final activities within the project lifetime and Bank management could not be persuaded to extend the project yet again.

6. Lessons Learned 105. Projects that are primarily geared to institutional reform should focus on change management components and activities even if these represent only a small part of the Project cost. The civil works components of RAMP were a distraction to the more fundamental task of tax administrative reform. From the earliest stages of project preparation, attention should have focused on building ownership for the institutional reforms within CTA management and among other key

24

internal and external stakeholders. These institutional reforms should have remained the focus of the project throughout project implementation.

106. Institutional stakeholder assessments and regular stakeholder consultations should be a requirement for institutional reform projects. Change management projects are likely to face institutional resistance. Undertaking a broad stakeholder assessment would highlight where the support and resistance to the reforms would come from. Such an exercise at the institutional level would have allowed the MoF and the CTA to develop a change management plan including all the CTA management. Repeating this exercise periodically, when there were changes in institutional leadership would have helped inform and strengthen ownership among new stakeholders. This may have helped overcome resistance to use of project proceeds for training, advisory services and stakeholder engagement activities. In the event those responsible for leading and implementing the change process were not sufficiently consulted and involved and the project languished without their commitment.

107. Project results frameworks should provide a results chain linking activities and outputs to intermediate and development outcomes. RAMP demonstrates the difficulties encountered in managing projects without an adequate M&E framework. The PDO remained very high level, the intermediate indicators were far too many and only weakly linked to the Project activities and the PDO itself. Agreement on the results framework early on in project design, definition of baselines before starting implementation and close monitoring of progress towards results would have helped focus attention on RAMP’s institutional reform objectives.

108. Projects should allocate adequate time for the implementation of complex IT systems and ensure that dispute resolution arrangements are in place and used promptly. Complex IT projects are frequently subject to implementation delays, often as a result of disagreements between the purchaser and the supplier. In retrospect, the two year project extension approved for the implementation of the CRMS was based on a best-case implementation scenario. When a dispute arose between the purchaser and the supplier, there was inadequate time to resolve the dispute and then complete implementation. There were dispute resolution arrangements in place, but the parties were slow to bring these to bear. The availability of an experienced IT professional to advise the authorities could have helped them take this decision at an earlier stage.

109. Projects with substantial civil works components should ensure that all land issues are resolved prior to project approval and include staff with experience of Bank-financed civil works in the preparation team. The Bank regularly takes this approach in infrastructure projects where TTLs are used to dealing with land issues. Where the project is implemented by a Bank unit with limited experience in civil works, such as Governance, the team should include an infrastructure specialist through cross support from another practice.

110. The Bank should ensure continuity in task team leadership. Bank teams often complain of the implementation challenges resulting from changes in political and technical leadership on the client side (and this project is no exception). Maintaining continuity in task team leadership is difficult but the Bank could commit to ensuring that the task leader that prepares the project also starts the implementation. This would focus more serious thought on implementation, results and M&E. Bank management could place more emphasis on the need for a thorough handover. The Bank could also empower country office staff to play a more important role as co-TTLs where appropriate to provide for longer-term continuity.

7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners

(a) Borrower/implementing agencies

25

111. The Borrower’s Implementation Completion Report (ICR) was received on November 18th 2015. In accordance with the Minister of Finance’s request we have reproduced the ICR verbatim in Annex 7 to this report.

112. There are no issues in the Borrower’s ICR that would warrant a response here. We note the tremendous support and commitment to the ICR process that has been shown by CTA and Croatian Ministry of Finance counterparts. The information that they have provided and their leadership in the stakeholder workshop has greatly assisted the team in the completion of this report.

(b) Co-financiers

113. There were no other co-financiers to the project.

(c) Other partners and stakeholders

114. A stakeholder workshop including the private sector, taxpayers associations, chambers of commerce, IT companies and academics was convened on October 27, 2015. A summary of the proceedings of the workshop are presented in Annex 6 to this report.

26

Annex 1. Project Costs and Financing

(a) Project Cost by Component (in USD Million equivalent)

Components Appraisal Estimate

(USD millions)

Actual/Latest Estimate (USD

millions)


A. Organizational consolidation

and functional restructuring of the Zagreb Regional Office

35.36 7.41 35.75

B. Knowledge and professional development of staff

9.52 0.57 10.29

C. Technological improvement of services of the Tax Administration, data management system and the introduction of a personal identification number - PIN

20.40 7.69 64.26

D. Modernization management and support for the project

2.72 0.10 6.25

Total Baseline Cost 68.00 15.78 23.21

Physical Contingencies

0.00

0.00

0.00

Price Contingencies

0.00

0.00

0.00 Total Project Costs 68.00 15.78 39.57

Front-end fee PPF 0.00 0.00 0.00 Front-end fee IBRD 0.00 0.00 0.00

Total Financing Required 0.00 0.00 0.00

(b) Financing

Source of Funds Type of Co-

financing

Appraisal Estimate

(USD millions)

Actual/Latest Estimate

(USD millions)


Borrower 0.00 58.10 0.00 0.00 International Bank for Reconstruction and Development

0.00 68.00 15.78 23.2

27

Annex 2. Outputs by Component

Results Indicator Baseline

Actual End of Project Target Achievement

Project development objective indicators

A. Increase in taxpayer compliance rates, based on internationally accepted method for estimating tax gap.

B. Decrease in taxpayer compliance costs, as assessed by taxpayer services.

C. Perception of taxpayers regarding the quality of service provided by CTA, as measured by periodic surveys

D. Perception of tax officers

regarding the terms of employment and working environment, as measured by periodic surveys

2.4 (2008) 14,495 kn (2010) 2.66 (2010 – score out of 5 with 5 being the highest) 3.05 (2010 – score out of 5 with 5 being the highest)

2.05 (2011) 1.81 (2012) 1.52 (2013) 11,977 kn (2012) 3.19 (2013) 3.14 (2013)

At least a 5% improvement At least a 10% improvement At least a 5% improvement At least a 10% improvement

Fully achieved. Progress was on track by 2013 but more recent data is not available. Fully achieved. Progress was on track by 2012 but there has not been another survey since then. Fully achieved. Progress was on track by 2013 but there has not been another survey since then. Not fully achieved. Progress was on track by 2013 but there has not been another survey since then.

28

Intermediate results indicator – Component 1

Organizational Consolidation and Functional Realignment of Tax Offices

A pilot LTO in Zagreb that would report directly to the head of the CTA established.

No genuine National LTO exists.

A fully-fledged national level LTO is in operation since 2012.

A fully-fledged national level LTO is in operation.

Fully achieved. The LTO operations will be enhanced once it is integrated with the CRMS in line with the BPR recommendations.

Ratio of staff in inspections functions as a percent of total staff in the LTO

49% in 2012 51% (2014) 59% (2015 October)

65% Not achieved. The LTO was reorganized in January 2015. 82% of the positions are filled and they remain focused on improving services and strengthening the team.


Knowledge and Professional Upgrading of Tax Officials and Stakeholders

Training per CTA staff including distance learning*

8.00 hours (2011) 29.95 hours (2013) 21.97 hours (2014)

18.75 hours (2015) Fully achieved.

E-learning modules developed and implemented*

0 (no e-learning) 2007 3 (2013) 9 (2014) 10 (2015)

10 Fully achieved. The e-modules developed included; anticorruption, E-Taxation, VAT, Small value Public Procurement, Information Protection and Safety, Management performance Strategy, Unrelated Transactions Specification Enforced collection, Fire

29

Protection, and Occupational Health and Safety.


Technological Upgrading, MIS and TIN implementation

Stop filers of the medium and small taxpayer segments*

VAT 2.81% (2011) CIT 4.44% PIT 3.68%

VAT 3.35% CIT 4.33% PIT 3.81% Overall 4.46 (2013) Overall 3.61 (2014)

2.00 (percent of total) VAT 2% CIT 2% PIT 2%

Not achieved.

Annual average assessments per inspector increased**

€ 348,075 (2007) € 302,102 (2013 all) € 939,227 (2013 LTO) € 229,192 (2013 other)

€ 850,000 (2009) € 500,000 (2013) At least a 5% improvement (2015)

Not achieved.

Total declarations by legal entities filed electronically**

3% (2007) 54.32% (2012) 97.66% (2013) 99.20% (2014)

80% (2009) 90% (2013)

Fully achieved.

* Indicators introduced following the 2013 restructuring. ** Indicators revised following the 2010 restructuring.

30

Annex 3. Economic and Financial Analysis

Economic Analysis

1. Economic analysis of the project considers the direct benefits and costs associated with changes in economic welfare arising from the project. These are estimated using data for 2007-2014 obtained from the Croatian Tax Administration and Statistical Office and IMF World Economic Outlook.

2. The estimated economic benefits include reduced compliance costs for legal entities and improved CTA’s operational efficiency. The decrease of tax compliance costs stems from simplified filing and payment of taxes and social contributions; reduced time dealing with tax officials and expanded e-filing; and enhanced taxpayer services, including among others updated and accurate information for taxpayers provided on the CTA’s website and automated call centers. Lower compliance costs will provide greater scope for taxpayers to engage in productive and income generating activities. Gains in operational efficiency result from the modernization of CTA’s operational functions, streamlining of business procedures and data cleansing. These measures have improved CTA’s operational efficiency and impact the overall quality of taxpayer services.

3. There are significant additional economic benefits which are difficult to quantify. These include reduced avenues for corruption, greater transparency and accountability of the CTA and increased trust in the tax administration and government. The latter is a particularly important factor for improving a taxpaying culture which is likely to generate large fiscal benefits in the long run.

4. Economic costs involve project costs as well as administrative costs associated with its implementation. Such costs encompass value of an investment and CTA staff time spent on the implementation of the project.

5. The economic analysis suggests that project is feasible. The net present value (NPV) in real terms is Euro 72.7 million at 12% discount rate, and IRR is 259%. If the discount rate is 25%, the NPV in real terms is Euro 31.3 million. Without financing, the rate of return is higher while the NPV is of a similar magnitude.

6. A more detailed discussion of this analysis and assumptions used for the period after the project completion are provided below:

Project benefits will last five years after completion. Savings from CRMS implementation in 2016 are not considered.

An EU inflation rate is 1.5%, and an inflation rate in Croatia is 2.5% per annum. The exchange rate is 7.63 Kuna to Euro.

Number of registered legal entities in the CTA’s taxpayer registry is 219,510 in 2015. It is assumed to grow at 0.5% after 2015.

CTA staff time spent on the project implementation includes three full time equivalent (FTE) staff who have spent one day per week (51 days) annually. An average net wage is Euro 23,000 (2015) prorated for inflation.

According to the surveys on compliance costs for legal entities carried out by CTA in 2010 and 2012, an average compliance cost declined by 725 Kuna (Euro 95)—from 14,495 Kuna to 13,770 Kuna. The number of legal entities in the CTA taxpayer register was 198,887 in 2012; 208,293 in 2013; 214,944 in 2014 and 219,510 in 2015. This number is assumed to grow at 0.5% annually from 2016.

31

In addition, organizational consolidation and realignment as well as the implementation of OIB, e-filing and e-Audit generated large administrative cost savings. CTA reduced administrative costs each year between 2008-2014 except in 2009 when there was an increase in administrative costs. It is assumed that the project has contributed to 3% of annual total administrative cost savings which grow by 1% from 2015.

Financial Analysis

7. Financial analysis of the project considers costs and benefits for government using a cash flow approach. Financial benefits comprise collection of tax revenues and social contributions at the level above the trend of negative economic growth experienced by Croatia since the outbreak of the 2008 global financial crisis. This occurred in 2012 and 2014. CTA managed to maintain collection of revenues and social contributions at the level above the negative GDP growth during the RAMP implementation. However, these revenue increases are mainly attributed to revenue policy measures including: an increase in the general VAT rate from 23 to 25 percent in February 2012; abolishment of the zero VAT rate and introduction of a 5 percent rate for basic food, pharmaceuticals and books in 2013; broadening of the tax base for the gambling tax and increase of the reduced VAT rate from 10 to 13 percent social and health sector contribution to 15 percent in 2014. Financial benefits also include CTA’s administrative costs savings which are attributed to the results of reform measures implemented by the RAMP. These measures included improving the accuracy of the taxpayer registry; simplification of procedures for taxpayer registration, filing and payment of taxes and social contributions; better assessment and mitigation of risks through a tax-gap analysis and risk-based audits; improved quality of taxpayer services to support voluntary compliance; expansion of e-filing; and implementation of OIB and e-Audit. Financial costs are the same as economic costs.

8. The financial analysis suggests that project is financially nonviable since significant fiscal gains for the budget were associated with revenue policy measures. The NPV in real terms is negative and thus the project generated no financial returns.

32

Economic Analysis( In million Euro, Nominal) 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020

CostsAll project costs

Project costs 0.20 0.27 1.00 1.00 1.00 2.00 3.00 3.00

Staff time costs for project implementation 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01

Loan repayments and fees

Front end fee 0.03

Commitment fee 0.13 0.12 0.12 0.02 0.02 0.02 0.01 0.01

Repayments 0.82 0.82 0.82 0.82 0.82 0.82 0.82 0.82 0.82

Total Costs 0.17 0.34 0.40 1.04 1.04 1.86 2.85 3.84 3.84 0.84 0.84 0.84 0.84 0.84

BenefitsAdministrative costs savings 0.69 -0.23 0.11 0.09 0.10 0.05 0.17 0.17 0.17 0.17 0.17 0.18 0.18Complinace costs savings for legal entities 18.90 20.29 21.46 22.46 23.14 23.84 24.55 25.29 26.05Loan disbursements 0.20 0.27 1.00 1.00 1.00 2.00 3.00 3.00

Total Benefits 0.00 0.89 0.04 1.11 1.09 20.00 22.34 24.62 25.63 23.31 24.01 24.73 25.47 26.23

Net Benefits -0.17 0.55 -0.37 0.07 0.05 18.14 19.49 20.78 21.79 22.47 23.17 23.89 24.63 25.39

( In million Euro, Real)Net Benefits -0.18 0.59 -0.39 0.07 0.06 18.70 19.88 20.99 21.79 22.25 22.71 23.19 23.67 24.16

IRR 259%NPV 12% 72.72NPV 25% 29.87

33

Financial Analysis( In million Euro, Nominal) 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020

Financial OutflowsAll project costs

Project costs 0.20 0.27 1.00 1.00 1.00 2.00 3.00 3.00

Staff time costs for project implementation 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01

Loan repayments and fees

Front end fee 0.03

Commitment fee 0.13 0.12 0.12 0.02 0.02 0.02 0.01 0.01

Repayments 0.82 0.82 0.82 0.82 0.82 0.82 0.82 0.82 0.82

Total Outflows 0.17 0.34 0.40 1.04 1.04 1.86 2.85 3.84 3.84 0.84 0.84 0.84 0.84 0.84

Financial InflowsAdministrative costs saving 0.69 -0.23 0.11 0.09 0.10 0.05 0.17 0.17 0.17 0.17 0.17 0.18 0.18

Revenue due to Project 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Loan disbursements 0.20 0.27 1.00 1.00 1.00 2.00 3.00 3.00

Total Inflows 0.00 0.89 0.04 1.11 1.09 1.10 2.05 3.17 3.17 0.17 0.17 0.17 0.18 0.18

Net Cash Flow -0.17 0.55 -0.37 0.07 0.05 -0.75 -0.80 -0.68 -0.67 -0.67 -0.67 -0.66 -0.66 -0.66

( In million Euro, Real)

Net Cash Flow -0.18 0.59 -0.39 0.07 0.06 -0.78 -0.82 -0.68 -0.67 -0.66 -0.65 -0.64 -0.64 -0.63

IRR 0%NPV 10% -2.62NPV 20% -1.27

34

Annex 4. Bank Lending and Implementation Support/Supervision Processes

(a) Task Team members

Names Title Unit Responsibility/

Specialty Lending Waleed Haider Malik Lead Public Sector Specialist GGODR TTL 2007 Andrina A. Ambrose-Gardiner Senior Operations Officer OPSPQ Zoran Anusic Senior Economist GSPDR Casandra Bischoff Consultant GGODR Jean-Charles Marie De Daruvar Senior Counsel LEGAM Arnold B. Gordon Consultant ECSPE Egli Ilic Finance Analyst WFALA Yoko Kagawa Senior Operations Officer GGODR Munawer Sultan Khwaja Lead Public Sector Specialist PRMPS Irina L. Kichigina Chief Counsel LEGLE Sanja Madzarevic-Sujster Sr Country Economist GMFDR Mirela Mart Financial Management Specialist ECADE William V. Mayville Consultant GGODR Amitabha Mukherjee Lead Public Sector Specialist GGODR Craig R. Neal Consultant GGODR TTL 07-10 Karin Shepardson Program Manager GCCIA Jiro Tominaga Senior Evaluation Officer IEGCC Antonia G. Viyachka Procurement Specialist GGODR Virginia S. Yates Program Assistant GGODR

Supervision/ICR K. Miagara De Silva Senior Economist GGODR TTL 10-11 Edgardo Mosqueira Medina Senior Public Sector Specialist GGODR TTL 11-13 Alberto Leyton Lead Public Sector Specialist GGODR TTL 13-15 Ismail Radwan Lead Public Sector Specialist GGODR ICR TTL Nataliya Biletska Public Sector Specialist GGODR ICR co-author Zoran Anusic Senior Economist GSPDR Michael Engelschalk Senior Public Sector Specialist GGODR Enrique Fanta Ivanovic Senior Trade Facilitation Specialist GTCDR Andreea Florescu Team Assistant ECCRO Dubravka Jerman Program Assistant ECCHR Mala D. Johnson Program Assistant ECSPE Mirjana Kacavenda Senior Human Resources Business HRDEM Munawer Sultan Khwaja Lead Public Sector Specialist PRMPS Matija Laco Operations Officer ECCBK Sanja Madzarevic-Sujster Sr Country Economist GMFDR Lamija Marijanovic Financial Management Specialist GGODR William V. Mayville Consultant GGODR Natasa Vetma Senior Environmental Specialist GENDR Antonia G. Viyachka Procurement Specialist GGODR Anneliese Viorela Voinea Financial Management Analyst GGODR Iwona Warzecha Sr Financial Management Specialist GGODR

35

(b) Staff Time and Cost

Stage of Project Cycle Staff Time and Cost (Bank Budget Only)

No. of staff weeks* USD Thousands (including travel and consultant costs)

Lending FY07 44.30 316.94

Total: 316.94 Supervision/ICR

FY07 0.00 0.00 FY08 17.15 88.28 FY09 21.4 141.82 FY10 25.87 125.23 FY11 21.11 135.26 FY12 8.34 119.02 FY13 27.94 240.90 FY14 14.1 110.86 FY15 13.58 110.76 FY16 11.5 78.63

Total: 160.99 1,150.76

Total Lending and Supervision 205.29 1,467.70

*Note: Staff weeks includes only Bank staff and not consultants.

36

Annex 5. Beneficiary Survey Results The establishment of a fully-fledged Large Taxpayers Office (LTO) was one of the main activities supported by the Project. With a view to soliciting feedback on the performance of the CTA and particularly the LTO, the ICR team carried out a short on-line survey of Large Taxpayers in Croatia. The survey was carried out between October 10 and 17, 2015 using a Survey Monkey and involved no additional cost to staff time. The survey was administered in Croatian and sent to all 642 registered large taxpayers. An English translation of the survey questions is presented in Box 1 below. 95 respondents participated in the Survey comprising the response rate of 14.8 percent. However, only 40 respondents answered Question 3 which represented 6.2 percent of all large taxpayers.

The survey responses are summarized in Figures 1 and 2 below. The results of the Survey show that the majority of respondents perceive that the CTA has achieved improvements in the quality of its services (50.5%). The other areas in which the respondents think that the CTA has demonstrated progress include: the quantity of services (36.8%), timeliness of responses to inquiries (32.6%) and timeliness of VAT refunds (31.6%). Only 12.6 percent of respondents thought that CTA had made no improvements in dealing with large tax-payers. This confirms the strong endorsement of the CTA that tax-payers voiced during the stakeholder workshop (see Annex 6).

Box 1: World Bank survey of Large Taxpayers in Croatia The World Bank has been working together with CTA on improving services to large taxpayers. We have recently closed our project and are documenting its achievements and shortcomings. In this regard we would very much appreciate five minutes of your time to complete the attached survey. The survey is being administered by an independent party and all responses will be non‐attributable and totally confidential. 1. In your opinion, what improvements have the CTA made in dealing with large taxpayers?

Please select relevant answers. Timeliness of responses to inquiries by large taxpayers Timeliness of VAT refunds Quality of CTA services Quantity of CTA services None

2. In your opinion, what areas still require improvements? Please select relevant answers. Timeliness of responses to inquiries by large taxpayers Timeliness of VAT refunds Quality of CTA services Quantity of CTA services None

3. Please provide any other feedback, positive or negative, on CTA services and / or suggestions for improvements in the future.

37

Figure A1. CTA’s Improvements in Dealing with Large Taxpayers Responses (In Percent)

On the other hand, the majority of the respondents believe that the timeliness of responses to inquiries of large taxpayers still needs to be improved. The other two areas that a significant percentage of the respondents feel still requires improvements are the quantity of services (31.6%) and the quality of services (28.4%). Once again only 12.6 percent of the respondents think that the CTA has made no improvements or does not require any improvements.

Figure A2. Areas Still Requiring CTA’s Improvements Responses (In Percent)

The third question asked respondents to provide feedback, positive or negative, on the CTA services and/or suggestions for improvements in the future. In their responses, the respondents noted the CTA’s progress in replying to inquiries and resolving issues experienced by large

12.63%

36.84%

50.53%

31.58%

32.63%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

None

Quantity of CTA services

Quality of CTA services

Timeliness of VAT refunds

Timeliness of responses to inquiries by largetaxpayers

12.63%

31.58%

28.42%

5.28%

54.74%

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

None

Quantity of CTA services

Quality of CTA services

Timeliness of VAT refunds

Timeliness of responses to inquiries by largetaxpayers

38

taxpayers in a timely manner but noted the lack of qualified staff to provide clear and unambiguous replies as well as the tendency of some tax inspectors to seek reasons to impose penalties on taxpayers rather than providing them guidance interpreting tax regulations. A few respondents also indicated that the CTA lacked flexibility and imposed strict measures that were sometimes not commensurate with the materiality of a matter in dispute. About half the respondents to the third question made suggestions for improvements. These included: increased timeliness of replies to inquiries; enhancement of professionalism and knowledge of tax inspectors as well as IT skills; rationalization of requirements for compliance with tax obligations; improved structure of tax forms (i.e. inclusion of check summary lines); standardization of answers to inquires through a specialized service to avoid different interpretations by different inspectors; issuance of official opinions free of charge; improved dispute resolution procedures; and enable electronic submission of all the necessary documents in the online regime.

39

Annex 6. Stakeholder Workshop Report and Results The World Bank in collaboration with the Croatian Tax Administration organized a stakeholder workshop on October 27, 2015. The main objective of the workshop was to discuss the achievements of the Revenue Administration Modernization Project. The ICR team solicited feedback from stakeholders on how the collaboration of the CTA and the Bank on the Project could have been improved and what areas in tax administration the CTA needs to strengthen further. The list of participants is included in Table A1. The workshop program started with opening remarks by the CTA and the Bank followed by CTA presentations on the main components and results of the Project and concluded with an open discussion. This report includes the main highlights of the CTA’s presentations and discussion with the stakeholders. The CTA emphasized that the Project sought to support the CTA’s modernization and reform in many areas but the key outputs of the Project that had influenced or would affect improvements in the CTA’s performance the most were: the design, development and implementation of a Compliance Risk Management System (CRMS); implementation of a Taxpayer Identification Number (TIN or OIB in Croatian); overhaul of the website; and training and e-learning for the CTA staff. Presentation 1 : Compliance Risk Management System (CRMS) The CRMS is the main output financed by the Project and its purpose is to increase the efficiency of tax collection by strengthening the fight against tax evasion. The CRMS includes four sub-systems: risk analysis (RMS), data warehouse (DWH), general audit support (GAS), and an integration solution for GAS. IBM Croatia is the lead contractor for the RMS and DWH while, following a dispute with IBM the GAS has been removed from the original CRMS contract and has been tasked to government’s in-house IT agency APIS. Hence, the integration solution will link the GAS with the other sub-systems. At the Project’s closure, the CRMS was still under implementation and thus the Government allocated funding from the state budget to complete it. The CRMS will be an important tool for the CTA to identify, assess and manage risks in tax administration. The CTA noted that Croatia’s loss of revenue due to tax evasion amounted to about 10-15 percent of total revenue collected, mainly through non-filing, underreporting and underpayment. In the process of developing the CRMS, the CTA identified over 400 risks using 30 data sources, including from other government agencies and third party information, and involved staff across the different CTA functions. The resulting risk catalogue will become an integral part of the CRMS with appropriate measures to address these risks. Benefits of the CRMS implementation include: automation of risk management in one place; using limited resources to focus on high-risk taxpayers and specific risk areas, classification of taxpayers in risk categories based on their behavior; transparency in selecting taxpayers for audit; and automation of VAT refund processing. Presentation 2 : e-Audit The Project supported the establishment of e-Audit with a view to increasing the efficiency and effectiveness of tax audits carried out by the CTA. The Project financed the procurement of 52 licenses for ACL, a software allowing to analyze large data in a short time; training in the use of ACL; development of e-audit procedures; and, in conjunction with the EU IPA 2009 Twining

40

Program, preparation of an e-Audit Manual following EU best practice. The implementation of e-Audit allowed CTA to reduce the duration of tax audits which generated time savings for tax auditors and taxpayers undergoing audit, as well as increase the number of tax audits per year. Presentation 3 : Taxpayer Identification Number (OIB) The Project supported the introduction of TIN (OIB in Croatian) that is now the single government identifier for all legal and natural persons. The OIB was a turning point in enabling the integration of all state administration and the creation of an e-Citizens platform which allowed government to simplify and fast-track administrative processes and reduce administrative costs. It has a wide application including tax returns, payment transactions, registration of identity documents and many others. The OIB also generated large spillover benefits to the private sector by reducing transaction costs for businesses and citizens. The CTA demonstrated three examples of such benefits: registration of a newborn child is now followed by government automatically issuing an OIB and registration of residence without additional paperwork with the CTA and the Ministry of Interior; registration of new legal entity allows to automatically issue an OIB without a visit to the CTA; and control over cash transactions to combat the gray economy. Presentation 4 : New CTA Website The Project supported the development of a Strategy for Internal and External Communication, an important element of which was the design of a new CTA website to enable open communication and partnership with taxpayers. The objective of the upgraded website was to facilitate the reduction of compliance costs for taxpayers, simplification of procedures and increased tax collection. The new website was adapted to the needs of modern users and includes information on the rights and obligations of taxpayers by taxpayer segment, namely citizens, artisans and freelancers, companies, nonprofit organizations and non-residents. The functional features of the website comprise: current information on over 1000 laws and regulations governing tax issues; a brief overview of the tax system; tax forms and CTA publications; links to the e-filing (e-Tax) system and OIB; verification of VAT payments and others. Presentation 5 : Training and e-Learning The Project financed the CTA training and e-learning activities aimed to support the implementation of the CTA strategies: a Tax Administration Strategy for 2011-2015, a Human Resource Management Strategy for 2010-2015 and a Strategy for Professional Training for 2012-2015. The main objectives of the training and e-learning were to enable easier and faster access to information and continuous learning and knowledge sharing, provide reliable and just-in-time information for tax officials, and reduce costs of training. The training comprised IT programs, special topical and issue program and foreign language courses. The e-learning consisted of e-courses with tests, e-questionnaires, educational materials, a forum for tax officials, and a repository of the CTA opinions and guidelines. The CTA was the first government organization in Croatia to provide e-learning to its staff. Key Highlights of Discussion Overall, private sector stakeholders noted the significant progress of the CTA in creating a customer-oriented culture and the tremendous value of the upgraded CTA website and e-filing, financed by the Project. Both of these achievements decreased tax compliances costs for businesses.

41

Several stakeholders suggested that the CTA should be more proactive in tax policy to ensure stability of the tax system, as frequent changes in the tax legislation and regulation present difficulties for taxpayer compliance in fulfilling their tax obligations. The workshop participants also noted that the Project would have benefitted from less frequent changes of the Project’s TTLs on the Bank side and the PIU Heads on the CTA side, as every time a newly appointed person had to learn about the Project all over again. In addition, the CTA noted that the Bank rules are different from the rules of other development partners and it involved additional time for new PIU Heads to learn them.

42

Table A1. List of Stakeholder Workshop Participants.

Name Title

Ms. Sunčana Sović-Digna Head of Unit, CTA Ms. Ivana Predović Independent Clerk - General Services, CTA Ms. Marijana Strugar Senior Administrative Advisor, CTA Ms. Emilia Ozeke EU Department, CTA Mr. Zoran Jančiev Assistant Director, CTA Mr. Tomislav Balun Key Accounts Manager, IBM Croatia Mr. Tomislav Majerić Global Business Services Sales Leader, IBM Croatia Mr. Željko Pađen Senior Expert Advisor, Croatian Chamber of Commerce Ms. Mila Vuluga Head of UVP, CTA Ms. Nada Pavlinić Head of Branch Office Zagreb, CTA Mr. Josip Benčić Head of Branch Office Central Croatia, CTA Mr. Darjan Dragičević AMCHAM, Croatia Mr. Krešimir Musa Head of Consulting, CROZ Ms. Marta Vidaković Expert Associate, MoF Ms. Sanja Flegar Senior Expert Specialist, MoF Ms. Jasna Vujnović Executive Director, IN2 d.o.o. Zagreb Ms. Martina Ciglević Assistant Director, CTA Mr. Igor Borošak Assistant Director, CTA Ms. Nada Petrović Assistant Director, CTA Ms. Lidija Grgić Head of Branch Office Slavonia and Baranja, CTA Mr. Šimun Buzov Expert Associate, Croatian Chamber of Commerce Ms. Ljiljana Jagatić Project Coordinator, CTA Mr. Stjepan Bobinac INFODOM Zagreb Mr. Danijel Šimić Head of CRMS Project, IBM Croatia Ms. Gordana Marić Assistant Director, CTA Ms. Marijana Vuraić-Kudeljan Deputy Director, CTA Mr. Neven Završki Croatian Chamber of Trades and Crafts

Mr. Josip Komes Head of Branch Office North Croatia, CTA Mr. Hrvoje Somun President of the Management Board, APIS IT d.o.o. Mr. Hrvoje Sagrak INFODOM Zagreb Mr. Marko Nenadić PR, CTA

43

Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR The following information was submitted to the World Bank by the Croatian Ministry of Finance on November 18th 2015. It is included verbatim including their highlights. GENERAL PROJECT INFORMATION Revenue Administration Modernization Project, which was prepared in 2006 and the first half of 2007 and described in the Project Appraisal Document which was compiled by the World Bank on May 21, 2007, has been designed as a development project in order to achieve further improvements in efficiency, service to taxpayers and consolidation of the tax system by strengthening capacity and improving the Tax Administration system. The Loan Agreement for financing was signed on July 3, 2007 and came into effect on December 21, 2007. In order to achieve the goal the project supported the development of a strategic plan for the modernization of the tax system and assisted in the implementation of it during the initially contracted five years of the loan duration (until June 30, 2013), which with the subsequent restructuring was extended by another two years (until June 30, 2015.). Investment of loan funds in the modernization was aimed at facilitating harmonization of tax policies and business operations with the requirements of accession to the EU.

The project was designed as a big ambitious project worth EUR 92.74 million (of which the loan was EUR 50 million), which included four inter-related modernization components. The first three components were aimed at achieving greater operational efficiency, strengthening services to taxpayers and increasing revenue collection of the state budget, of which the Tax Administration is in charge, while operational support for the project was provided within the last component with the aim of providing as simple and higher quality implementation of project activities as possible. The project components were:

1. Organizational strengthening and functional restructuring of the tax office, including physical facilities,

2. Improving the knowledge and expertise of tax officials and stakeholders, 3. Technological improvement of services, the MIS system and resources for implementing

the PB systems 4. Support to the modernization and management.

As stated in the document on the evaluation of the project, in the years preceding the year of contracting the loan for the project, Croatia recorded solid growth and low inflation as GDP growth averaged at about 4.8 percent in the 2001-2006 period, and maintained inflation at 2.7 percent. We state the data to understand the conditions in which the modernization of the Tax Administration was planned. After the very first activities on the project when the first consulting analyses and assessments of the situation were prepared with proposals for changes and improvements, the economic crisis was nearing, which affected Croatia as well. With it the implementation of key objectives for improvement became difficult. In the period of the implementation of the project, Croatia has harmonized regulations with the EU acquis, has been working intensively on the negotiations on individual chapters, and finally on 1 July, 2013 became a full member of the EU. The Tax Administration has used the possibility of using the EU pre-accession funds to finance major development projects that went hand in hand with projects financed by the World Bank loan.

Effective implementation of the project started in early 2008 while as part of the project, by June 30, 2015, 70 contracts and amendments were concluded. It is important to emphasize that the first component was central to the whole project, given that it included the construction of a new

44

building to consolidate and functionally reorganize local offices of the Zagreb Regional Office. The new building would have integrated functions that took place at several locations in Zagreb, would have enabled functional restructuring of the Tax Administration, and would have provided space for IT and communications equipment of the Tax Administration with the aim of reducing the use of IT services of external partners, i.e., greater independence in the field of data processing. It was also planned to equip the space for the training of tax officials (Tax Academy). Regardless of the efforts of the Tax Administration to provide adequate space for the construction of the new building we did not succeed.

Abandoning the construction of the building meant that the project needed to be focused on those areas where we could achieve significant improvements, thus we launched the project restructuring procedure. The Bank approved the restructuring on September 24, 2010 and that’s when half of the loan was cancelled with further financing of projects within RAMP coming only from the loan without participation of budgetary funds. The Tax Administration has initially mainly independently done all the tasks related to the preparation of tender documents for certain projects, with limited assistance from outside consultants with respect to the government's recommendation on the implementation of projects based on their own resources available for civil servants. In this way the necessary pace could not have been achieved, thus in January 2012, on the recommendation of the World Bank experts, we hired one full-time consultant for procurement, and we also hired another one in August. After that, the procedure of preparing the documentation and communication with the World Bank was significantly accelerated, with the high quality of documentation submitted to the Bank, and so was the timely electronic communication, which was necessary for project success.

Given the long-term preparation of major projects, such as the Compliance Risk Management System-CRMS), and the Reconstruction Project, upgrading and equipping of the Tax Administration building in Vinkovci and the establishment of the Disaster Recovery Centre, as well as a lengthy bidding process for the selection of suppliers and other objective circumstances, contractual deliveries could not have been completed by the agreed closing date of the loan (June 30, 2013), and the Administration had once again requested the restructuring of the loan with the extension of the duration of the loan by 2 years and cancellation of 1.5 million Euros. The restructuring was carried out on May 8, 2013 and since then the amount of the loan has been 23.5 million Euros with the duration until June 30, 2015.

In early 2015, the situation in compliance risk management system (CRMS) projects and construction of the building in Vinkovci suggested that the project activities, due to a variety of objective circumstances, will not be completed by June 30, 2015. Therefore, the Tax Administration, in cooperation with experts from the World Bank, in April 2015 submitted a new request to the World Bank for the extension of the loan period by 18 months, i.e., until December 31, 2016. The Bank did not approve the new request, thus June 30, 2015 is the final closing date of the loan. In the end, the Tax Administration has used 11,466,035.00 Euros from the loan which makes 48.79%.

We wish to emphasize that during the years of the cooperation on the project the Tax Administration has received exceptional professional and selfless assistance from colleagues from the World Bank. We thank you for the assistance and support at a time when our project was not proceeding according to plan. We particularly appreciate the approval of the extension of the loan in 2013 and we thank the Bank team that supported us because this extension allowed us to continue with important projects and achieve significant improvements.

With thanks and praise for the involvement of experts from the World Bank we need to mention

45

the situations that did not contribute to the speed of implementation of project activities. First of all we mean the frequent changes in the project manager by the World Bank (we had 6 managers). Each new manager needed some time to get acquainted with the project, which sometimes resulted in delays in the delivery of responses to requests of the Tax Administration. In addition, there were changes of project managers by the Tax Administration, several changes of directors and a number of changes in the composition of the Committee for Strategic Management of the Project. All this affected the pace of the execution of project activities, thus we conclude that steady management needs to be provided to ensure success of the project, as well as an Implementation Unit, which has sufficient capacity and is committed to the project.

Improvements Which Have Been Achieved

Regardless of the fact that the project underwent significant restructuring, and that only a portion of the funds that were available were spent, I wish to point out that in the years of project implementation and cooperation with experts from the World Bank significant results were achieved, and that now that the loan is closing the Administration is more modern, better organized, and more efficient as it provides better service to taxpayers. With the contribution of staff of the World Bank and the IMF, the Tax Administration adopted the Strategy of Tax Administration for the period from 2008-2013, which was updated in 2011, and a new strategy of the Tax Administration for the period from 2011-2015 was developed. Within the project these strategic documents were developed as well:

- The Strategy of Human Resources Management for the period from 2010-2015, - The Strategy of Professional Training of Tax Administration Staff for the period from

2012-2015, - The Strategy of External and Internal Communication of the Tax Administration for the

period from 2012 - 2015 - Report on Business Process Reengineering

Given the period covered by the said strategies, the Tax Administration has started the preparation for updating documents, i.e., determining strategic goals for the next period. The establishment of the PIN system, or a permanent identification mark for every Croatian citizen and legal entity based in the Republic of Croatia, launched a general computerization of the public administration, which increased efficiency of state institutions and reduced the administrative burden on citizens, automated data exchange between government bodies and other state institutions, which allowed citizens faster and easier exercising of their rights; we have a better view of the assets of citizens and legal entities, as well as income and cash flow, which is a key prerequisite for a transparent economy and systematic combating of corruption. It ultimately harmonized the Croatian legislation with the EU legislation. The establishment of the PIN system was a turning point in the networking of the state administration. The project The Analysis and Re-engineering of Core Business Processes and Functions of the Tax Administration Central Office was aimed at analysing and documenting business processes involved in the core operational functions and functions of the Tax Administration Central Office, and at consideration of different ways to present current problems in business processes to remove or overcome the possible re-engineering of processes (or some other suitable change), and at preparation of a proposed plan for the implementation of the necessary changes. Documents created are very comprehensive and in summary we can say that the proposed changes, which include the experience of consultants, particularly from other tax administrations, as well as the best practices from other tax administrations and related projects, were grouped in 23 proposals that were later

46

used in the work of the Tax Administration. Proposals were related to improving existing business processes, both internal and external communication, management activities, document management, functional organization of business processes, reorganization of the organizational structure of the CTA, and changing legislation. Most proposals were taken into account and their implementation led to significant improvements in the work of the Tax Administration. We particularly emphasize that the functional organization of business processes and new organization of the Tax Administration have been in force since early 2015. With the modernization of the IT equipment and information systems, the Tax Administration today communicates with taxpayers electronically, almost all reports are submitted in electronic form, and thus taxpayers have significant savings in the time required for the fulfilment of tax obligations. Since implementing the project External and Internal Communication of the Tax Administration, the strategy for the external and internal communications has been made, the website of the Tax Administration has been modernized, which provides updated information needed for daily operations of taxpayers, a new intranet page has been created, and so has the new visual identity of the Tax Administration. Information about the rights and obligations of taxpayers are segmented by categories of taxpayers as follows: Citizens, Craftsmen and Freelancers, Companies, Associations, and Non-residents. The website presented in mid-2013 is much visited and is the only place where taxpayers can find consolidated texts of regulations. The increase in the number of visits to the website from 2,750,000 in 2013 to 8,170,000 in 2014 best shows user satisfaction with the changes that have been made. We deal with large taxpayers with particular care. With great effort of experts from the World Bank we established the Large Taxpayers Office where these taxpayers, who pay about 45% of tax revenues into the budget, have special services and individual approach. We developed horizontal monitoring in order to increase the satisfaction of taxpayers, reduce the administrative costs of taxation, and ultimately increase the level of voluntary compliance with tax obligations. Our inspectors are equipped with laptops and use tools for electronic audit. Within the project financed through the loan, 52 licenses were procured, and officers were trained who are now able to check the accounting information of the taxpayer much faster. The efficiency of inspectors has been increased collegial collaboration through teams that share information and experience based on audits that are carried out. This has also led to the achievement of equalization of treatment in the control on the entire territory of the Republic of Croatia. RAMP has brought significant improvements in the area of training of our officers. An e-learning system has been developed that has 10 different programs, and has been very well accepted by all employees. That is also the only e-learning system within the state administration. The use of the module outside regular business hours has also been provided so that employees could choose the time for education that best suits them. A series of trainings have been conducted that significantly raised the level of professional training of officers. The training was completed by 17 employees in IT programs, 1,884 officers completed specialized financial seminars (accounting), and 525 employees completed 2 semesters of foreign language education (mostly English). We particularly want to emphasize that the most important improvement and change in all aspects of the Tax Administration will follow the completion of the aforementioned project Compliance Risk Management System-CRMS of the execution of tax liabilities. The project was created with great help from the World Bank experts in all phases, including a tender for the selection of suppliers, and subsequent negotiations with the selected bidder to achieve a high quality final result of the project. The project is in progress with a deadline on June 30, 2016, and part of the delivery, which could not have been completed until the closing of the loan (30 June, 2015) will be financed

47

from own resources, or the budget of the Tax Administration; the total value of the project is around 4.54 million Euros. With the realization of the project we expect significant reduction of tax evasion, i.e., improving tax collection because the Tax Administration will be able to analyse over 400 identified risks and use over 30 different data sources, all with the aim of detecting risks at the very start and preventing the risks to be actualized. The benefits that the Tax Administration will have from the implementation of this IT tool are as follows: Automation of the process of VAT refund Automated support to the risk processing in one place Recognition of taxpayers who are prone to risks Directing limited audit resources towards the riskiest tax payers Classification of taxpayers in the predefined risk categories based on their behaviour Focusing monitoring on specific risk areas Transparency in selecting taxpayers for audit Decreasing of internal risks A substantial increase in awareness and knowledge of Tax Administration employees on

risk analysis. The realization of the modernization and upgrading of the Vinkovci office will enable the formation of the Disaster Recovery Centre. The delay in the implementation of the construction was caused by a discovery of archaeological findings that we as investors had to protect at the site. Archaeological findings reduced the gross usable area, and it was necessary to change the architectural layout of the building. The new architectural design has been created, and the World Bank has supported the continuation of the project (construction of the building), but the continuation of construction works will be financed from the budget. In conclusion, we would like to point out that the Tax Administration, in cooperation with the World Bank, using the loan as well as expert advice which helped us, has achieved significant improvements that we have mentioned in the report. We will continue to work on improvements, primarily on the completion of projects that are in progress, and for which we can no longer use the funds from the loan, as well as on other projects. For the previous cooperation we sincerely wish to thank you and hope that the World Bank will support us in the future in our efforts to achieve visions according to which the Tax Administration is a professional organization that provides quality service to taxpayers and performs efficient collection of budget revenues.

48

Annex 8. CTA Revenues 2007-2014

In HRK

Serial No.

NAME OF REVENUE - BUDGET Realized

payment 2007 Realized




payment 2011. Realized



payment 2014.

1 2 3 5 8 3 3 5 8 8

I Own tax revenues of the state budget (1 to 6) 47.747.639.498 53.171.748.679 48.736.633.251 47.279.257.593 46.353.695.129 49.707.834.333 48.035.466.199 47.985.058.245

1 Value-added tax 1 37.747.987.242 41.308.035.643 37.050.353.547 37.688.520.292 37.718.153.978 40.652.023.144 40.252.579.097 40.903.625.709

2 Corporate income tax 8.816.375.273 10.564.702.627 9.439.857.956 6.407.083.665 7.288.029.753 7.697.342.287 6.365.443.106 5.657.765.291

3 Sales tax 168.521.668 166.530.596 123.479.327 123.904.648 129.672.348 126.841.366 135.318.519 153.166.253

4 Other taxes on goods and services 509.636.132 543.867.968 532.699.866 524.077.915 514.816.671 513.833.268 508.432.360 423.014.797

5 Taxes and fees from games of chance 505.119.184 588.611.845 532.833.762 671.681.682 668.868.424 709.722.565 769.804.484 845.680.769

6 Special tax on salaries, pensions and other income 0 0 1.057.408.794 1.863.989.391 34.153.955 8.071.702 3.888.634 1.805.426

II Own county taxes (a to d)2 244.280.000 262.618.000 266.763.000 258.982.000 242.570.000 240.065.000 220.663.000 216.069.000

a) Inheritance and gifts tax 2.039.000 2.507.000 3.992.000 2.985.000 6.793.000 6.774.000 5.226.000 8.922.000

b) Road motor vehicle tax 227.900.000 245.122.000 248.663.000 242.891.000 223.783.000 222.039.000 205.295.000 198.025.000

c) Boat tax 1.636.000 1.905.000 1.943.000 2.010.000 2.422.000 2.411.000 2.626.000 2.768.000

d) Slot machine tax 12.705.000 13.084.000 12.165.000 11.096.000 9.572.000 8.841.000 7.516.000 6.354.000

III Own city/municipal taxes (a to h)2 172.178.000 176.681.000 163.434.000 154.017.000 153.355.000 168.839.000 201.796.000 187.631.000

a) Consumption tax 69.085.000 70.049.000 63.265.000 57.409.000 57.973.000 61.785.000 93.494.000 90.166.000

b) Second home tax 34.673.000 37.008.000 36.945.000 38.164.000 40.928.000 44.675.000 44.638.000 42.420.000

c) Advertisements tax 241.000 192.000 47.000 38.000 40.000 64.000 27.000 -5.000

d) Trading name or corporate name ta 63.035.000 65.160.000 58.037.000 55.175.000 52.338.000 59.604.000 59.364.000 50.293.000

e) Public land use tax 4.353.000 4.690.000 5.074.000 3.638.000 3.425.000 3.432.000 4.252.000 4.218.000

f) Tax on uncultivated agricultural land 3.000 -7.000 0 0 4.000 0 -4.000 0

g) Tax on unused entrepreneurial property 781.000 20.000 26.000 -200.000 -1.244.000 -738.000 4.000 539.000

h) Tax on undeveloped building land 7.000 -431.000 40.000 -207.000 -109.000 17.000 21.000 0

IV Total own tax revenues for the state, county and city/municipal budget ( I + II + III )

48.164.097.498 53.611.047.679 49.166.830.251 47.692.256.593 46.749.620.129 50.116.738.333 48.457.925.199 48.388.758.245

V Joint tax revenues (1 to 2) 13.810.220.584 14.902.378.526 14.190.135.956 12.573.251.636 12.540.342.635 13.139.586.210 13.845.860.213 13.883.688.561

1 Personal income tax and surtax 12.385.046.239 13.336.047.273 12.879.058.831 11.479.697.845 11.435.689.267 12.159.901.011 12.706.922.751 12.932.996.001

2 Real property transaction tax 1.425.174.346 1.566.331.253 1.311.077.125 1.093.553.791 1.104.653.368 979.685.199 1.138.937.462 950.692.560

VI Total tax revenue ( IV + V ) 61.974.318.083 68.513.426.205 63.356.966.206 60.265.508.229 59.289.962.765 63.256.324.543 62.303.785.412 62.272.446.806

VII Total contributions ( 1 to 3 ) 37.203.485.795 40.703.483.985 39.994.738.956 38.712.382.061 38.605.066.634 37.845.870.621 37.147.345.429 41.686.069.297

1 Health insurance 16.940.153.497 18.652.621.817 18.287.811.222 17.751.622.660 17.722.809.921 16.714.500.262 15.888.678.107 17.302.733.243

2 Retirement insurance 18.543.082.518 20.154.787.768 19.838.374.844 19.153.377.842 19.075.125.033 19.290.518.027 19.406.174.630 22.451.695.866

3 Employment 1.720.249.780 1.896.074.400 1.868.552.890 1.807.381.559 1.807.131.680 1.840.852.332 1.852.492.692 1.931.640.188

TOTAL TAX AND CONTRIBUTION REVENUES OF THE STATE (VI + VII )

99.177.803.878 109.216.910.190 103.351.705.162 98.977.890.290 97.895.029.399 101.102.195.164 99.451.130.841 103.958.516.103

Source: P -1 and P-2 forms and the Information Technology System of the CTA

49

Annex 9. List of Supporting Documents 1. Project Appraisal Document, Report No. 39219-HR dated June 4, 2007.

2. Loan Agreement, Loan Number 7471 HR dated July 3, 2007.

3. Project Information Document No. AB2696, dated November 27, 2006.

4. Environment Management Plan No. E1622, dated April 2007.

5. Restructuring Paper, Report No. 56518 v1 dated September 7, 2010.

6. Restructuring Paper dated June 5, 2013.

7. Amendments to the Loan Agreement dated September 8, 2010.

8. Amendments to the Loan Agreement and Partial Cancelation of Loan Proceeds dated June 2013.

9. Progress Report for Revenue Administration Modernization Project for 1st quarter 2015 prepared by the Republic of Croatia, Ministry of Finance.

10. Implementation Status and Results Reports No. 1 to 16 dated between November 27th 2007 and May 13th 2015.

11. Croatia Country Assistance Strategy, November 24, 2004.

12. Croatia CPS Progress Report, May 2007.

13. Aide Memoirs of Bank missions from 2007 to 2015.

14. Establishing and calculating key performance indicators (KPIs) for the RAMP Project. Deloitte Consulting July 1st 2010.

15. Investment Project Financing Economic Analysis Guidance Note, OPSPQ, October 2014.

16. ICR Guidelines, May, 2011.

17. IEG Guidelines for reviewing World Bank ICRs, November 2013.

50

MAP