1

Bureaucracy and Growth

: Revisiting Evans’ Effect of Weberian Bureaucracy on Economic Growth

Yeon Ho Lee

(Yonsei University)

Yeoun Ki

(Yonsei University)

I. Introduction

II. Literature Review

III. Research Design

IV. Results

V. Conclusion

VI. Bibliography

abstract

One of key features that have been drawing countries economic growth is said to lie in

its meritocratic bureaucracy. And this is said to be the main driving force of rapid

economic growth between 1970’s to 1990’s for East Asian countries that are called the

developmental state. With State holding strong autonomy, bureaucratic structure supported

and practically carried out goals set by State as an institution and administrative procedure

of drawing economic growth. Evans and Rauch(1999) tested the effect of meritocratic

bureaucracy on economic growth empirically and suggested the significant relations

between the two. However, considering today’s economic status, it is questionable that

whether the role of bureaucracy is significantly related to the economic growth. In this

article, by testing with new and current data for quality of bureaucracy and economic

growth, it tries to examine whether the strong and significant relations between

bureaucracy and growth are still meaningful. By adopting former works of Evans and

Rauch(1999), this article empirically tested the effect of meritocratic bureaucracy in

today’s economic growth.

2

I. Introduction

Is meritocratic bureaucracy important in pushing its economic growth rate? This debate

dates back to Weber’s theory on rationality of bureaucracy. And more recently this

meritocratic feature was suggested as a significant contributing factor in leading economic

development of developing countries. The cases of East Asian countries that are classified as

developmental state are strong examples that support such argument, especially during 1970

to 1990. Evans and Rauch(1999) tested such relevance empirically and proved its

correlations.

However, there are needs to look closely into the logic of meritocratic bureaucracy and its

impact on economic growth. Especially since 1997 Asian Financial crises, so-called

developmental states undertook several reforms on its state strategy on economic growth

towards regulatory model, and the developmental model seems only to work in East Asian

areas, whereas Latin American developmental state models have shown some disappointing

outcomes. And also, the economic shape is rapidly changing after 2008 economic crisis. With

this relevance, whether meritocratic bureaucracy still holds significance needs to be tested

with new circumstances taken into account.

On given circumstances, this article tries to empirically test whether meritocratic

bureaucracy is still important in today’s economic development across the countries. And to

do so, this tries to adopt former model of Evans and Rauch(1999) and the variable suggsted

by it to argue whether their logic is still significant in evaluating today’s economic

conditions. Using statistical data with linear regression, this article tries to verify the logics

supported by Evans and Rauch(1999), and moreover former works supported by

developmental state theorists.

To test empirically, the main variables that measures meritocratic feature of bureaucracy are

government’s former recruitment system which entails strict entrance examinations and

government’s reward on bureaucratic personnel within the system. The growth of economic

change is measured between 2008 to 2013 to see the most current situations and situations

after 2008 economic crisis.

II. Literature review

With the advent of so-called four East Asian Tiger’s rapid economic growth through state’s

holding strong autonomy in leading its economic growth. And the key element of these state’s

strong autonomy lies in its bureaucratic structure. Chalmer Johnson(1982) suggested the

concept of the “capitalist developmental state” to refer to the developmental state based on

institutional arrangements common to high growth in East Asian economies.

3

Of the characteristics of such developmental states, such as holding economic growth as the

first priority and absence of any commitment to equality, the most significant feature of states

holding strong autonomy lies in its bureaucratic structure. Under strong state autonomy, the

market is guided with instruments formulated by a small-scale, elite bureaucracy, recruited

from the best managerial talent availiable in the system. Within the bureaucracy, a pilot

agency play a key role in policy formulation and implementation. Close institutionalized

links are established between the lite bureaucracy and private business for consultation and

cooperation. The organizational and institutional links between the bureaucratic elites and

major private sector firms are crucial in generating a consensus on goals, as well as in

exchanging information, both of which constitute essential components of the process of

policy formulation and implementation. It is a political system where bureaucracy is given

sufficient scope to take initiatives and operate effectively. The politicians “reign” while the

bureaucracy “rule”(Onis,1991).

What upholds this developmental state model are one hand suggested by the term “state

autonomy” and on the other “bureaucracy” as an institutional setting. Behind this

developmental state model lies two types of human capital; elite politicians and elite

bureaucrats. The former organizes political institutions of developmental state strategy and

the latter organizes administrative and technical institutions of the developmental state. The

bureaucratic elites substantially and through policies propel the purposes set by political elites

which sets economic growth as a priority and insulates its priorities from socio-political

effects that may harm political agendas. From this perspective, it could be said that the role of

bureaucracy in accomplishing the purpose of developmental states accounts for significant

importance (Lee, 2016).

To analyze this bureaucratic structure in relationship with economic growth, Evans and

Rauch (1999) empirically tested the relationship between bureaucratic structure and

economic growth. In doing so, they established a term ‘Weberain scale’ to feature

bureaucratic proficiency. This ‘Weberian Scale’ reflects the ideas of bureaucracy suggested

by Weber that bureaucracy is a particular kind of organizational structure set in contrasts to

earlier patrimonial and prebendal forms of government administration. More precisely, the

public administrative organizations that are characterized by meritocratic recruitment and

predictable, long-term career rewards will be more effective at facilitating capitalist growth

than other forms of state organizations(Evans & Rauch, 1999).

Evans and Rauch established a new dataset concerning bureaucratic characteristics with

regards to whether each country’s bureaucracy follows the features suggested by Weberian

terms. They used data collected between 1993-1996 through expert survey questions to

illustrate 35 countries existing their bureaucratic characteristics. The dependent variable here

is the total growth of real GDP per capita during the 1970-1990 period. Results showed that

there is significant positive relations between Weberainness scale of bureaucracy and GDP

4

growth.

Figure 1

The effect of Weberianness Scale on Economic Growth(Evans&Rauch, 1999)

Evans and Rauch(1999) controlled for past average schooling years to control for the

possibility Weberianness could be a proxy for overall level of development stocks of human

capital and also controlled for initial GDP level. The relation between the Weberianness and

economic growth remained strong and significant. And the scattergram suggested another

significant feature that there is strong degree to which the regional distribution of Weberian

characteristics parallels regional differences in growth performance. Specially, the four cases

of East Asian countries which account for so-called East Asian Four Tigers epitomize, during

concerned period, both high growth and high Weberian bureaucratic traits.

However, this growth with bureaucratic structure was not without any sacrifices. Strong

state autonomy insulate outside influence that may erode its policy agenda. States, to

maintain its main driving force and attain strong autonomy, inevitably form a regime that

certain political elites dominates and that is isolated from any other social forces or pressures,

in turn it suggests the development of public goods and public wealth. Therefore, the civil

society is less likely to appear in such social conditions. And the major flaw led by this

developmental model is entailed by its ignorance towards distribution where the main focus

is paid upon development and growth. (Lee, 2009).

Political institutionalization theory, suggested by Huntington, addresses that ,at the early

5

stage of development, authorities should be centralized on the hand of modern elites in a way

that can control the mass public, which could be analyzed to pursue strong elitists

perspectives. This emphasis on central authority and institutionalization of such had

developed into a rationale that justifies authoritative regime in developing countries where

little infrastructure prevailed. And this was not the exception with East Asian developmental

states. Authoritative institutions generated backgrounds for political corruption that later

dismantled domestic political arena. And for most, this emphasis on state autonomy created

unequal access where capitals and firms are allowed to participate in the political process

whereas civil society is isolated from it(Lee, 2009).

Literatures have been published with regards to this state-led strategy that sets economic

growth as priority has generated consequences on democratic foundations.

Prezeworski(2000)suggested that the form of regime is not significantly related to its

economic growth through statistical analysis. Sirowy and Inkels(1990) suggested that there is

other variables that could contribute to economic growth other than the form of regime.

However, less attention has been shared with regards to Evans’s former argument with

regards to bureaucratic structure and economic growth. Libman(2012) suggested how the size

of bureaucracy is negatively associated with economic growth in the case of Russian regions.

And in doing so, Libman(2012) tested empirically with data that there is significant negative

relations between the two variables. This result suggests counter-wise as what Evans might

have suggested. But, even this lacks further development in analyzing how “Weberianness”

of bureaucratic features could contribute to economic growth after the Cold war era.

To test whether Evans’s argument is still adaptable after the cold war era, this article uses

cross-sectional data on bureaucratic professionalism to test its relevance with economic

growth. By narrowing the time span after 2008 economic crisis, this article tries to answer

whether the bureaucratic features are still relevant in state’s economic growth.

III. Research Designs

In this article, the main hypothesis to test is based upon Evans and Rauch(1999). With

regards to their work, it tries to test the hypothesis of whether bureaucratic features that could

be said to hold Weberianness have positive relation with economic growth rate.

Evans and Rauch(1999) addressed a basic equation of which standardized form is:

% Change in GDP per capita 1970-1990 = -0.317(real GDP per Capita 1965) +

0.307(average years of schooling 1965) + 0.615(Weberianness score) (1)

6

And through empirically testing this hypothetical theory that Evans and Rauch(1999)

suggested with new dataset comparable to that of what Evans and Rauch(1999) formally

used, this article tries to answer the question whether their argument is still adaptable today.

3.1 Hypothesis

According to Evans and Rauch(1999), Weberianness scale consists of mainly two parts.

First, Weberianness Scale concerns when public employee is hired, the weight is put on

education an formal entrance examination system. Second, predictable careers and long-term

employment are important for crating a working environment for bureaucracy where

meritocracy is rewarded.

Based on this two important features of Weberianness scale that Evans and Rauch(1999)

suggested, two hypotheses of this paper that relates bureaucratic characteristics to economic

growth arise.

Hypothesis 1: If the country is more likely to have meritocratic entrance examinations when

hiring its bureaucratic employees, it is more likely that it experiences higher growth

of GDP per capita.

(2)

Hypothesis 2: If the country is more likely to reward its bureaucrats with long-term

employment and predictable careers, it is more likely that it experiences higher

growth of GDP per capita.

(3)

3.2 Data

To test the two hypotheses, the source of data is QoG standard data published by QoG

Institute of University of Gothenburg is used. QoG’s cross-sectional data, published in 2017,

examines 194 countries and their quality of government. And the data for year of 2013 is

prioritized. Therefore, other control variables and dependent variables are extracted

concerning year of 2013.

For independent variable, two variables from QoG standard dataset is used for each

hypothesis 1 and hypothesis 2. Index of closedness(Closed Public Administration) and index

of professionalism(Professional Public Administration) are used respectively to test each

hypothesis. Index of closedness is relevant to whether specific government has formal

entrance examination institutionalized for hiring its personnel. Closedness index measures to

7

what extent th public administration is more closed or public-like, rather than open or private

like. On the other hand, index of professionalism is relevant to independent variable of

hypothesis 2, whether the country is more likely reward its bureaucratic personnel with

predictable careers. Both of indexes of closedness and professionalism reflect the mean value

of survey question

Table1.

Index for main independent variables and measurement

Index Measurement

Index of Closedness

1. Public sector employees are hired via formal examination system

2. Entry to the public sector is open only at the lowest level of the hierarchy

3. The terms of employment for public sector employees are regulated by special

laws that do not apply to private sector employees.

Index of

Professionalism

1. When recruiting public sector employees, the skills and merits of the applicants

decide who gets the job.

2. When recruiting public sector employees, the political connections of the

applicants decide who gets the job.

3. The top political leadership hires and fires senior public officials.

4. Senior public officials are recruited from within the ranks of the public sector.

Main dependent variable is, as suggested by Evans and Rauch(1999), the percentage of

change of GDP per capita in each country. To control the effects of 2008 economic crisis and

to reflect the most current status of economic growth, the dependent variable is extracted

from percentage of change in GDP per capita from 2008 to 2013. The source of data is World

Bank database.

% Change of GDP per capita (2008-2013)= [{(GDP per capita in 2013) – (GDP per capita

in 2008)}/(GDP per capita in 2008)] *100 (4)

Evans and Rauch(1999) controlled for former GDP per capita level to control the initial per

capita levels and initial human capital levels separately. The logic behind this is firstly

suggested by the fact that more highly developed bureaucracies are more likely to be found

among the developed countries. And also to avoid the criticism that Weberianness is simply a

proxy for overall level of development or existing stocks of human capital, they controlled

8

for the initial average level of schooling for each countries, which accounts for average

schooling years in the total population over 25. And both control variables are the score in

1965.

In this article, however, as an index of preexisting conditions of each economic background,

only GDP per capita of 2000 is controlled, of which source of data is World Bank Database.

Because of existing strong correlations between GDP per capita of 2000 and average years of

education in 2000 for each country, only GDP per capita is controlled for initial conditions.

Instead, to control for development of human capital alternative variable of average schooling

years, which accounts for the averages years schooling spent by populations over age of 25 in

2010 is controlled. Considering the time-scope of dependent variable from 2008-2013 and the

strong correlations between each initial conditions, average year of schooling in 2010 is more

adequate to control for the effects of development of human capital and change of GDP per

capita. And the data is extracted from Barro-Lee Data set which provide educational

attainment data for 146 countries.

And last control variable is the income level of each country. Instead of classifying countries

into regional distribution, by using cross-sectional data, this article tries to examine how

economic classes where countries are part of could affect the outcome of relationship

between bureaucracy and economic growth. Whereas Evans and Rauch(1999) only focused

on developing countries, the variable of high level income countries, upper middle level

income countries, lower middle level income and low income level countries are added as

dummy variable which are extracted from World Bank Database.

IV. Results

Through linear regression test, this article tries to analyze whether the Weberianness of

bureaucracy still is in significant and positive relations with economic growth, as Evans and

Rauch(1999) suggested. Through testing two hypotheses with empirical data, it tries to solve

whether the argument suggested by Evans and Rauch(1999) is still adoptable. The result of

the regression model is summarized in Table 2 and Table 3.

The first hypothesis that tries to verify the causal relationship between formal bureaucratic

entrance examination with economic growth is shown in model 1 to model 5. The variable

that showed constantly significant was the former economic conditions. This suggests that

countries where former GDP per capita level is high are associated with slow rate of

economic growth rate. As is expected by economic theories, where economic conditions are

high, it is highly likely that those countries are associated with slower growth rate.

The main independent variable, bureaucratic closedness, which indicates the level of

institutions that is prone to have formal recruitment system, is shown significantly negatively

9

correlated with the economic growth. This is partially different from what Evans and

Rauch(1999) suggested. Evans and Rauch(1999) suggested in their paper that their

Weberianness scale is conducted of importance of exams in recruiting civil servants to the

core economic agencies and more generally. The result of their works indicated that the

Weberianness in positive correlation with economic growth rate between 1970-1990.

Figure 2

Effect of bureaucratic closedness on % change of economic growth(2008-2013)

However, the result suggested opposite correlation from Evans and Rauch(1999) in this

paper. The higher the degree to which government depend on formal examination system to

recruit its personnel, the more likely it is faced with slower rate of economic growth. When

controlled for the level of income of country classified by World Bank dataset, to be more

specifically, this tendency showed constant significant relationship in ‘High income

countries’ and ‘Lower middle income countries’. The best example, as drawn in attached

scattergram, could be suggested by the case of France. In short, hypothesis 1 tuned out

negatively significant and this was true of countries that are classified as high income and

lower middle income.

Table 2

10

Linear regression result of bureaucratic closedness on % change of economic growth(2008-

2013)

model1 model2 model3 model4 model5

bureaucratic closedness

-1.938*

(0.0602)

-1.950*

(0.059)

-1.326

(0.1932)

-1.890*

(0.5515)

-1.465

(0.151)

bureaucratic

professionalism

Average year spent at

schooling(25+)

0.0937

(0.3549)

0.967

(0.340)

0.978

(0.335)

0.951

(0.348)

1.032

(0.308)

Former GDP per

capita(2007,$)

-4.778*

(0.000)

-2.553**

(0.015)

-5.430***

(0.000)

-3.698***

(0.000)

-4.414***

(0.000)

High income countries -0.904

(0.372)

Upper middle income

countries

-2.248**

(0.0308)

Lower middle income

countries

1.152

(0.137)

Low income countries 1.732*

(0.091)

Number of countries 38 37 37 37 37

Adjusted R-squared 0.3585 0.3553 0.4219 0.3805 0.3914

(* p<0.1, **p<0.05, *** p<0.001)

Hypothesis 2, which tries to clarify the existing correlation between ‘ If the country is more

likely to reward its bureaucrats with long-term employment and predictable careers, it is

more likely that it experiences higher growth of GDP per capita’ , is tested through separate

regression model, models from model 6 to model 10. The test on overall countries showed

statistically non-significant correlations between bureaucratic professionalism and change of

economic growth. Only former GDP per Capita level, as suggested by testing hypothesis1,

proved to be significant.

However, when controlled for the classified income level provided by World bank, one

group showed statistical significance that is relevant to Evans and Rauch(1999). Countries

that are classified as lower income group showed positive correlation between bureaucratic

professionalism and the degree of economic growth.

11

Table 2

Linear regression result of bureaucratic professionalism on % change of economic

growth(2008-2013)

(* p<0.1, **p<0.05, *** p<0.001)

Whereas Evans and Rauch(1999) only used data from so-called developing countries, this

article used cross-sectional data regardless of the stage of development. Additionally, World

Bank no longer classifies countries according to the stage of development that countries are

undergoing. To generalize Evans and Rauch(1999) and clarify wether it could be adopted in

today’s cross-sectional areas, this article tested two hypotheses that was suggested in Evans

and Rauch(1999) by using different data.

The results are partially contradicting and partially supporting Evans and Rauch(1999). For

model6 model7 model8 model9 model10

bureaucratic closedness

bureaucratic professionalism 1.601

(0.113)

1.979*

(0.051)

1.498

(0.138)

1.665**

(0.0998)

1.590

(0.1157)

Average year spent at

schooling(25+)

0.697

(0.487)

0.832

(0.4077)

0.675

(0.502)

0.718

(0.474)

0.708

(0.4812)

Former GDP per capita(2007,$) -4.084***

(0.000)

-1.366

(0.176)

-4.235***

(0.000)

-3.373***

(0.001)

-

3.773**

*

(0.000)

High income countries -2.541**

(0.013)

Upper middle income countries -1.194

(0.236)

Lower middle income countries 2.198**

(0.030)

Low income countries 1.065

(0.2901)

Number of countries 82 81 81 81 81

Adjusted R-squared 0.1707 0.223 0.175 0.2081 0.172

12

meritocratic recruitment of bureaucratic personnel, there exists negative correlations between

meritocratic recruitment and % of change in GDP per capita. This is proving hypothesis 1 in a

negative direction. And for hypothesis 2, there is no statistically significance between

bureaucratic professionalism and economic growth. Only for the case of lower income groups

are positively suggesting that bureaucratic professionalism accounts for growth rate. If it

could be argued that lower middle income countries are those countries that traditionally

classified as developing countries, Evans and Rauch(1999)’s argument could partially be

supported due to the findings.

V. Conclusion

Unlike former analysis on the relationship between economic growth and meritocratic

bureaucracy provided by Evans and Rauch(1999), this article finds that Evans and

Rauch(1999) is no longer in present. The myth that has supported the economic growth of

developmental states in 1970s and 1980s is shown to be falsified in present economic

conditions. Meritocratic bureaucracy does not hold significant positive effect on economic

growth after 2008. Even, it is statistically found that meritocratic bureaucratic feature has

negative effect on economic growth. And further case study is needed to verify why such

phenomena take place in presence.

13

VI. Bibliography

Charron, N., Dahlström, C., & Lapuente, V. (2015). Measuring Meritocracy in the Public

Sector in Europe. QoG Working Paper Series, 8, 8.

Chibber, V. (2002). Bureaucratic Rationality and the Developmental State 1. American

Journal of Sociology, 107(4), 951-989.

Cho, W., & Porumbescu, G. (2011). Weberian Bureaucracy and Performance of Civil Service:

Cross-County Analysis. Seoul National University, 1-24.

Clements, B. J., Gupta, S., Karpowicz, I., & Tareq, S. (2010). Evaluating government

employment and compensation. International Monetary Fund, Fiscal Affairs Department.

Dahlström, C., Lapuente, V., & Teorell, J. (2011). Dimensions of bureaucracy II: A cross-

national dataset on the structure and behavior of public administration. QoG Working

Paper Series, 6.

Evans, P. B. (1989, December). Predatory, developmental, and other apparatuses: a

comparative political economy perspective on the third world state. In Sociological Forum

(Vol. 4, No. 4, pp. 561-587). Kluwer Academic Publishers-Plenum Publishers.

Evans, P., & Rauch, J. E. (1999). Bureaucracy and growth: A cross-national analysis of the

effects of" Weberian" state structures on economic growth. American sociological review,

748-765.

Evans, P. B. (2012). Embedded autonomy: States and industrial transformation. Princeton

University Press.

Ho, S. P. (1987). Economics, economic bureaucracy, and Taiwan's economic development.

Pacific Affairs, 226-247.

Huskey, E. (2010). Elite recruitment and state-society relations in technocratic authoritarian

regimes: The Russian case. Communist and post-communist studies, 43(4), 363-372.

Leff, N. H. (1964). Economic development through bureaucratic corruption. American

behavioral scientist, 8(3), 8-14.

Lee,Y.H.(2009).The Institutionalist Political Economic Developmnetal State Theory and

Democracy. 21st centry Political Science Review, 19(2),99-120

Lee,Y.H. (2016). Regulatory Reform and the Reinforcement of the Bureaucratic Power in South

14

Korea. Journal of East and West Studies, 28(1), 233-258.

Libman, A. (2012). Democracy, size of bureaucracy, and economic growth: evidence from

Russian regions. Empirical Economics, 43(3), 1321-1352.

Olsen, J. P. (2006). Maybe it is time to rediscover bureaucracy. Journal of public

administration research and theory, 16(1), 1-24.

Öniş, Z., Amsden, A. H., Deyo, F. C., Johnson, C., & Wade, R. (1991). The logic of the

developmental state.

Pollock III, P. H. (2015). The essentials of political analysis. Cq Press.

Rauch, J. E., & Evans, P. B. (2000). Bureaucratic structure and bureaucratic performance in

less developed countries. Journal of public economics, 75(1), 49-71.

Seligman, L. G. (1964). Elite recruitment and political development. The Journal of Politics,

26(3), 612-626.

Silberman, B. S. (2005). The bureaucratic state in Japan: the problem of authority and

legitimacy. CORNELL EAST ASIA SERIES, 123, 226.

Van Rijckeghem, C., & Weder, B. (2001). Bureaucratic corruption and the rate of temptation:

do wages in the civil service affect corruption, and by how much?. Journal of development

economics, 65(2), 307-331.

Woo-Cumings, M. (1999). The developmental state. Cornell University Press.