Institutions and Economic Growth: Empirical Evidence fron1 a

Cross-National Analysis

Maya Vijayaraghavan* and William A.Ward**

Clemson University

Center for International Trade

Working Paper 001302

Abstract: The relationship between institutional infrastructure and economic growth rates
across 43 nations between the years 1975-90 is examined. Within the framework of the
neoclassical growth model, this study integrates a broad set of institutional variables
which together proxy for the overall institutional infrastructure of an economy. Security
of property rights, governance, political freedom and size of government are the
indicators used in the study, facilitating identification of the most important institutions
that account for the observed variations in economic growth rates among nations. Results
indicate that security of property rights and size of government are the most significant
institutions that explain the variations in economic growth rates.

* Department of Agricultural and Applied Economics, 247 Barre Hall, Clemson SC

29634-0355. Email: [email protected]. Tel: 864.656.7822
** Director, Center for International Trade, 325 Sirrine Hall, Clemson SC 29634-1394
 1

Introduction

The discovery in neo-classical economics that relative prices mattered led to an

excessive focus on them in the context of development policy. Price reforms, along with

macroeconomic stability and privatization were the reformers' rallying cries in the 1980s.

The encounter between neo-classical economics and developing societies served to reveal

the institutional underpinnings of market economies, when it became clear that incentives

would not work or generate perverse results in the absence of adequate institutions.

Some of the implications of this were recognized in discussions on rent seeking in the

trade policy context, where corruption was the main issue, and in discussions on

common-property resources, where the focus was on property rights. However, the

broader point that markets needed support from non-market institutions took a while to

gain recognition. One of the ways in which institutional capacity can affect economic

performance is through the allocation of resources. Providing public services, providing

quasi-public goods or public goods, and intervening to improve the functioning of

markets are all directly concerned with resource allocation. The simplest mechanism by

which a lack of institutional capacity might result in resource misallocation is through

inefficient investment choices by the public sector. The lack of good institutions may

also lead to inefficient investment choices by the private sector.

The process of integrating institutions into economic theory is of comparatively

recent vintage. Over a decade ago, institutional measures were first introduced into

cross-country growth equations, and there has been a recent renaissance in this literature.

Differences in institutions across countries have proven empirically to be among the most

important determinants of differences in rates of economic growth. However, the

question of which aspects of institutions matter more for long-run economic growth has
 2

proven to be more controversial than the proposition that institutions are important

overall. Diverse measures have been used, encompassing property rights, political

freedom, political instability, governance, and measures of the quality of institutions for

economic exchange. The collective classification of these measures has tended to

obscure the different channels through which institutions operate. In this study, four

measures of institutional infrastructure - security of property rights, governance, political

freedom and government consumption are used as proxies for the overall institutional

environment, and an attempt is made to identify the institutions that appear to be the most

significant source of the differences in growth rates. The results from this study will

therefore contribute to the debate in the growth and institutions literature.

The second section of this paper summarizes select literature on economic growth

and institutions. The third section details the empirical model, variables used in the

study, their likely impacts, sources of data and model estimation techniques. The results

of the regression are discussed in the fourth section, and the final section concludes the

paper.

Literature

The majority of studies investigating the economic growth-institutions nexus use

a version of the neoclassical growth model (Solow, 1956), augmented to include

measures of human capital (from Mankiw, Romer and Weil (MRW), 1992) and

institutions. Previous studies examine the relationship between one aspect of institutions

and economic growth, without controlling for the presence of other institutions that could

alter the significance of the relationships. Among the pioneers in the growth and

institutions literature, Kormendi and Meguire (1985) explored the link between the Gastil

indices of political freedom (civil liberties and political rights) and economic growth and
 3

found a marginal effect of civil liberties on growth. Results from Scully's (1988)

analysis using the same indices provide similar support for the growth-political freedom

nexus. Helliwell's (1992) study however does not find a significant net effect of

democracy on growth. Barro (1996) finds the overall effect of democracy on growth to

be weakly negative, and some indication of a nonlinear relation in which more

democracy enhances growth at low levels of political freedom but depresses growth when

a moderate level of political freedom has already been attained. The more general

conclusion of this study is that the advanced western countries would contribute more to

the welfare of poor nations by exporting their economic systems, notably property rights

and free markets, rather than their political systems, which typically developed after

reasonable standards of living have been attained.

Knack and Keefer (1995) pioneered the use of indicators of security of property

rights in the growth literature, with the ICRG and BER! indices as proxies for this aspect

of institutions. The results from their analysis indicate that institutions that protect

property rights are crucial to economic growth. More recently, Mauro (1995) found

corruption in countries to be growth retarding.

In the last five years, a number of studies have used the Economic Freedom Index

from the Fraser Institute to investigate the link between economic growth and

institutions. Ali (1997), and Ali and Crain (1999) find economic freedom to be a more

robust determinant of growth than political freedom and civil liberties. Ayal and Karras

(1998) find that economic growth enhances growth both via increasing total factor

productivity and via enhancing capital accumulation. In a study by Dawson (1998),

economic freedom is found to be growth enhancing. Easton and Walker (1997) find that

economic freedom is an important explanatory variable for steady-state levels of income.

 4

The addition of a variable for economic freedom is also shown to increase the

explanatory power of the neo-classical growth model. Norton's (1998) study compares

property rights to indicators of development and determines that the "well-being of the

world's poorest inhabitants [is] sensitive to the cross-national specification of property

rights." The paper shows that well-specified property rights enhance the well-being of

the world's most impoverished.

Results from empirical analyses suggest the existence of the economic growth ­

institutions nexus, but statistical support is not uniform across all indicators of

institutional quality. Depending on the institutional variables chosen, the group of

countries in the analysis, and the time period of the study, the results are mixed. This

study attempts to integrate all available indicators of institutional capital within the same

model to determine the relative importance of each of these in explaining the variations in

growth performance across nations.

Data and Methodology

Using the framework of the neoclassical growth model, this study examines the

relationship between economic growth and institutions in 43 nations for the period 1975

to 1990. All except nine nations included in the study are "developing economies" as per

the World Bank classification of 1990. Table 3 lists the countries included in the study.

The basic theoretical framework outlined in Solow (1956) and MRW (1992) is used for

the analysis. The model assumes that the economies are characterized by a production

function exhibiting the standard characteristics - constant returns to scale and

diminishing returns with respect to each of the factors of production. It is also assumed

that it is possible to derive the steady-state level of output for the model, and the
 5

dynamics of the path to such a steady-state can be described. Estimating equations are

derived similar to Dawson (1998).

The paper utilizes four measures of institutional infrastructure. A measure of

governance computed as a simple average of three indicators - corruption, rule of law

and bureaucratic quality, measured on a scale of 0-6, with higher values indicating

"better" ratings. The measure of security of property rights is computed as a simple

average of two indicators: risk of repudiation of contracts and risk of expropriation,

measured on a scale of 0-1 0, with higher values indicating "better" ratings. Political

freedom is computed as the simple average of indicators of civil liberties and political

rights, measured on a scale of 1-7, with 1 representing "most free". Government

consumption as a share of total consumption is measured on a scale of 0-10, with larger

values indicating smaller governments. Following previous studies in the literature, the

data on growth rate of GDP per worker are derived from the Penn World Tables Mark 5.6

(Summers and Heston, 1991). Initial income and the period average for investment share

are also from the same source. The data on human capital (secondary school enrolment,

%) are taken from Barro and Lee (2000), and averaged for the years 1975-90.

Descriptions of the variables used in the regression analysis are summarized in Table 1.

Expected relationship between economic growth rates and the explanatory variables

Initial level ofGDP:

Poorer countries, with low ratios of capital to labor, have high marginal products of

capital. Therefore, as an economy prospers, the return on investment declines, and the

growth rate tends accordingly to decline. On the other hand, a poor country that has a

low steady-state level of per-capita output - because, for example, it has institutions that

are inhospitable to investment - need not grow faster than a rich country. Since countries
 6

are likely to be poor or rich precisely because the underlying determinants of their steady

states are unfavorable or favorable, the model does not predict any clear pattern of simple

correlation between growth rates and initial incomes.

Labor force growth and depreciation:

There is a reduction in the capital stock per worker because of an increase in the number

of workers. If there were no new investment and no depreciation, capital per worker

would decline because of the increase in the labor force, resulting in a decline in the GDP

per worker, and therefore the growth rates.

Investment share:

The economy is closed, so that savings equals investment, and the only use of investment

in this economy is to accumulate capital. The consumers then rent this capital to firms

for use in production. As the investment share increases, more capital accumulates, and

output increases. So, the growth rate also increases.

Human Capital:

There is a greater labor force role of males in most developing countries, therefore male

education attainment more important in terms of the direct effects on GDP growth.

Increased education of women leads to a sharp fall in fertility, and hence in population

growth. Therefore, the overall effect of increased human capital is to increase GDP

growth.

Institutional variables:

Size ofgovernment:

The state has a role in providing a minimum level of certain services. However, if the

government is too large, it might be taking away resources that could otherwise yield a

higher rate of return. Ifthere were a greater volume of nonproductive government

 7

consumption, it would reduce the growth rate for a given starting value of GDP. In this

sense, big government is bad for growth. Therefore, a certain "optimum" size of

government might be indicated. Growth rates tend to increase as size of government

approaches this "optimum" and then decrease beyond this point.

Governance:

This is a composite measure that reflects the quality of the bureaucracy, maintenance of

rule of law and level of corruption in government. The quality of the bureaucracy

measures mechanisms for recruitment and training, autonomy from political pressure, and

strength and expertise to govern without drastic changes in policy or interruptions in

government services when governments change. Ifrules that are in place are not "good",

and these are enforced with enthusiasm and rigor, the effect could be counterproductive.

In the presence of good rules, growth could be higher. Maintenance of rule of law

reflects the degree to which the citizens of a country are willing to accept the established

institutions to make and implement laws and adjudicate disputes. Sound political

institutions and a strong judicial system are conducive to growth. In some circumstances,

corruption maybe preferable to honest enforcement of bad rules. Outcomes maybe worse

if rules and regulations that prohibit some useful economic activity are thoroughly

enforced rather than circumvented through bribes. On the other hand, the economy

maybe hampered if few legitimate activities can be undertaken without bribes. Thus, the

overall impact of corruption maybe ambiguous. The impact of the governance variable

on economic growth is therefore uncertain.

Security ofproperty rights:

This measure is constructed from two institutional variables - risk of repudiation of

contracts by government and risk of expropriation of private investment. The risk of

 8

repudiation of contracts indicator addresses the possibility that businesses, contractors,

and consultants face the risk of a modification in a contract taking the form of a

repudiation, postponement, or scaling down" due to "an income drop, budget cutbacks,

indigenization pressure, a change in government, or a change in government economic

and social priorities." The risk of expropriation indicator evaluates the risk of "outright

confiscation and forced nationalization" of property. Lower ratings "are given to

countries where expropriation of private investment is a likely event." Security of

property rights is therefore conducive to economic growth.

Politicalfreedom:

This measure is composed of indicators of political rights and civil liberties. The

political rights indicators are based on the degree to which individuals in a state have

control over those who govern. The civil liberties indicator purports to measure the

rights of the individual, including freedom of expression, assembly, association, and

religion. Political freedom provides a check on governmental power and thereby limits

the potential of public officials to amass personal wealth and to carry out unpopular

policies. Since at least some policies that stimulate growth will also be politically

popular, more political rights tend to be growth enhancing on this count. However, the

growth retarding features of greater political freedom need to be taken into account ­

these include the enhanced role of interest groups in systems with representative

governments. Thus, the net effect of political freedom on growth is theoretically

inconclusive.

Oil exporting economies:

Exports of oil account for a significant portion of the GDP of the oil exporting

economies, a factor that cannot be explained by differences in the other explanatory

 9

variables. Increases in oil exports will therefore lead to an increase in growth rates for

these economies. The sign on this dummy variable is expected to be positive.

The MRW (1992) human-capital augmented version of the Solow model is first

estimated for the sample of 43 countries listed in Table 3. The model is of the form:

Growth = f30 + [3/nitialInmme + f32Laborforcegrowth + f3investmentshare + f34Humancapital

The model is estimated following standard procedures for ordinary least squares (OLS).

The sign and significance of the variables in the model are similar to those in MRW.

However, an analysis of the residuals (Figure 1) reveals that the oil exporting economies

of Syria (SYR) and Tunisia (TUN) are among the outliers, and their growth rates are

underpredicted by the model. These economies experience high growth merely because

of one natural resource, a fact that is not accounted for by any of the variables included in

the model. Consequently, a dummy variable is created to account for this discrepancy.

The revised version of the model attempts to identify the most important

components of the institutional infrastructure that account for the differences in growth

rates.

The revised model is of the form:

Growth = 13 0 + f3JnitialIncome + f32Labofjorcegrowth + f33Investmentshare + f3 4 Humancapital +

f3 s/nstitution j + 13 pIL

where i=1..4 andj=govemance, security of property rights, political freedom and

government consumption.

The standard assumptions and estimation procedures for OLS apply to this revised

version of the model.

 10

Regression Results

Results of the models explaining the differences in growth rates across countries

are given in Table 2. Modell reproduces MRW's (1992) human-capital augmented

version of the Solow model for the 43 countries under consideration in this study. The

sign and significance of all of the variables are qualitatively similar to the results obtained

by MRW, although the proxy for human capital is not significant. The measure of human

capital is highly correlated with initial income in this sample of 43 countries (Table 6),

which could be one reason for the lack of significance of this variable in Modell. Model

2 is an extended version of Modell, with a dummy variable included for the oil

exporters, to account for higher growth rates merely because of the presence of one

natural resource, exports of which account for a significant fraction of GDP in these

economies. As expected, the dummy variable for oil exporters is positive and significant,

indicating that other things being equal, oil exporters would be expected to have higher

economic growth rates. A plot of the residuals from this model (Figure 2) reveals a good

prediction of the growth rates for the oil exporters. The sign and significance of the other

variables in Model 2 is as expected.

Model 3 includes all of the institutional variables for the 43 countries included in

the study. All four institutional variables affect economic growth rates positively. The

negative sign on the political freedom variable is merely a reflection of the "reverse

scale" - i.e. a country with a value of 1 is politically the "most free". However,

government consumption is the only institutional variable that is statistically significant

in Model 3, indicating that smaller governments are "better". Over the last century, the

size and scope of government have expanded enormously, particularly in the industrial

economies. The post-World War-II confidence in the government bred demands for it to
 11

do more. Industrial economies expanded the welfare state, and much of the developing

world embraced state-dominated development strategies. The result was a tremendous

expansion in the size and scope of government worldwide. State spending now

constitutes almost half of total income in the established industrial countries, and around

a quarter in the developing economies (World Development Report, 1997). The positive

effect of government consumption on economic growth rates in this sample of countries

is a reflection of this shift in government consumption.

In an attempt to determine the relative significance of the four component

measures of institutional capital, a step-wise regression is employed, where we start with

a model containing all of the component institutional indices (Model 3) and then

individually drop those that are statistically insignificant. This approach results in

governance and political freedom dropping out of the model, leaving us with Model 4, in

which the security of property rights and size of government are the insti111tional

variables that explain to a significant extent the differential growth performance across

nations. A plot of the residuals from this model (Figure 3) reveals no outliers, and the

model explains upto 50% of the variability in growth rates. More secure property rights

lead to a high level of GDP per capita through their effect on allocative efficiency. When

property rights are not well defined, resources maybe directed towards unproductive

activities. Transaction costs also tend to be high, and may prevent mutually beneficial

transactions. With well-defined property rights, growth will occur either through an

increase in the quantity of factors of production or through a more efficient use of

available factors of production. The first direct effect of security of property rights on

growth arises through a more efficient use of capital. Capital devoted to productive

activities will enhance the productive capacity of the economy. The structure of property
 12

rights is also expected to affect the allocation of capital. The presence of a secure system

of property rights promotes innovation, since rewards can be reaped from new products

or processes. In the absence of property rights, human capital may be used for rent­

seeking and other redistributive activities. Other things being equal, growth rates are

higher with a more secure system of property rights. For the sample of countries

included in this study, the high growth rates of the East Asian countries and the poor

growth performance of Sub-Saharan Africa is consistent with differences in their security

of property rights. However, the lack of statistical significance for the institutions of

governance in the cross-country regression should not be construed to imply that this

institution is unimportant for the process of economic growth. Rather, the high

correlation (0.73, from Table 6) between governance structures and property rights is one

possible reason why the results of the regression are as obtained. In the context of New

Institutional Economics, Williamson (2000) distinguishes four levels of social analysis.

The top level is the social embeddedness. This is where the norms, customs, mores,

traditions etc. are located. Religion plays a role at this level. Although an analysis of this

level is undertaken by some economic historians, it is taken as given by most institutional

economists. The New Institutional Economics has been concerned principally with levels

2 and 3. The second level is referred to as the institutional environment. Much of the

economics of property rights are of a Level 2 kind. The third level is where the

institutions of governance are located. Within this theoretical framework, there is a

virtuous cycle of feedback between the governance structures and security of property

rights. The high degree of correlation between these two institutions possibly captures

this phenomenon, with the Level 2 variable dominating the effect of the Level 3 variable.

Model 5, an extension of Model 4, tests for non-linearity of government consumption.

 13

Although the results seem to indicate decreasing returns when governments are too large,

they should be interpreted with caution, since our sample consists predominantly of

developing countries, where the average size of government is small. All of these

countries with smaller governments lie in the increasing portion of the growth­

government consumption curve. Further analysis with a larger sample of countries is

expected to yield more robust results.

Conclusion

This paper attempts to identify the most important institutional determinants of

differences in economic growth rates among countries. It provides an analysis of which

institutions prove to be growth-enhancing in a sample of forty three countries. The

results from the analysis are significant, and provide support for the historical evidence

presented by North and Thomas (1973), Rosenberg and Birdzell (1986) and North

(1990). They show that the security of property rights provides incentives for economic

growth in the world. Secure property rights also lead to an efficient allocation of

investment and to an efficient use of capital.

The results seem to indicate that smaller governments are "better". However,

government consumption merely reflects its size, and says nothing about the "quality",

i.e. its effectiveness. Dramatic changes in the global economy have fundamentally

changed the environment in which states operate, and the state is no longer seen merely

as a provider, but as facilitator and regulator. Since there is also a predominance of

"developing" countries in the sample, which lie in the "increasing" portion of the curve,

results for the size of government variable should be interpreted with caution.

This cross-country analysis does not account for inter-temporal changes in the

variables that could explain some of the variations in the growth rates. For future
 14

research, when additional data for the institutional indicators are available, the model

could be extended to include the temporal dimension and estimated as a panel. The

measures of institutional capital used in the study are far from perfect, and do not capture

all of the dimensions of institutions. Differences in Level 1 (Williamson, 2000)

institutions are also not captured in this study, a factor that could be a significant source

of variation in growth rates. A theoretical discussion of cultural differences and long-run

economic performance is provided by Lal (1998), which could be used as the building

block for empirical testing of this factor endowment as an important source of growth

differences.
 15

References

Ali, Abdiweli M. "Economic Freedom, Democracy and Growth." Journal ofPrivate

Enterprise, Fall 1997, 1-20.

Ali, Abdiweli M. and W. Mark Crain. "Institutional Distortions, Economic Freedom and
Growth." Draft Manuscript, April 1999, James M. Buchanan Center for Political
Economy.

Barro, Robert 1. "Economic Growth in a Cross Section of Countries." Quarterly Journal

ofEconomics, May 1991,407-43.

Barro, Robert 1. and Jong-Wha Lee. "International Comparisons of Educational

Attainment." Journal ofMonetary Economics, December 1993, 363-94.

Barro, Robert 1. "Democracy and Growth." Journal ofEconomic Growth, March 1996, 1­
27.

Dawson, John W. "Institutions, Investment and Growth: New Cross-Country and Panel
Data Evidence." Economic Inquiry, October 1998,603-19.

Easterly, W. and Ross Levine, "It's not factor accumulation: stylized facts and growth
models", Mimeo, World Bank and U. of Minnesota, September 1999.

Easton, Stephen T., and Michael A. Walker. "Income, Growth, and Economic Freedom."
American Economic Review, May 1997, 328-32.

Helliwell, John F. "Empirical Linkages between Democracy and Economic Growth."

NBER Working PaperNo. 4066,1992.

Keefer, Philip and Stephen Knack. "Institutions and Economic Performance: Cross­
Country Tests Using Alternative Institutional Measures." Economics and Politics,
November 1995,207-27.

Kormendi, Roger C. and Philip G.Meguire. "Macroeconomic Determinants of Growth:

Cross-Country Evidence." Journal ofMonetary Economics, September 1985, 141-63.

Lal, Deepak. Unintended Consequences: The Impact ofFactor Endowments, Culture,

and Politics on Long-Run Economic Performance (Ohlin Lectures, 7), MIT Press, 1998.

Mankiw, N.Gregory, David Romer and David N.Weil. "A Contribution to the Empirics
of Economic Growth." Quarterly Journal ofEconomics, May 1992,407-37.

Mankiw, N.Gregory. "The Growth of Nations. " Brookings Papers on Economic Activity,
1995,275-326.
 16

Mauro, Paolo. "Corruption and Growth." Quarterly Journal ofEconomics, August 1995,
681-712.

North, Douglass C. Institutions, Institutional Change, and Economic Performance. New

York: Cambridge University Press, 1990.

North, Douglass C. and Robert P. Thomas. The Rise ofthe Western World: A New
Economic History, Cambridge University Press, 1973.

Norton, Seth W. "Poverty, Property Rights, and Human Well-being: A Cross-national

Study." Cato Journal, Fall 1998,233-45.

Rosenberg, Nathan and L.E.Birdzell. How the West Grew Rich: The Economic
Transformation ofthe Industrial World, New York: Basic Books, 1986.

Scully, Gerald W. "The Institutional Framework and Economic Development." Journal

ofPolitical Economy, June 1988, 652-62.

Solow, Robert M. "A Contribution to the Theory of Economic Growth." Quarterly

Journal ofEconomics , February 1956,65-94.

Summers, Robert, and Alan Heston. "The Penn World Table (Mark 5): An Expanded Set
ofIntemational Comparisons, 1950-1988." Quarterly Journal ofEconomics, May 1991,
327-68.

The Fraser Institute. Gwartney, Jim and Robert Lawson. "Economic Freedom in the
World: 2000 Annual Report." Vancouver, 2000. Data retrieved from
http://www.freetheworld.com

The Freedom House. Annual Survey ofFreedom Country Scores 1972-73 to 1999-2000,
Washington DC, 2000.

Williamson, Oliver E. "The New Institutional Economics: Taking Stock, Looking

Ahead." Journal ofEconomic Literature, September 2000,595-613.

World Development Report. The State in a Changing World, The World Bank, 1997.
 17

Table 1: Description of variables and their expected sign in the estimated model

Variable Source Description Expected Sign

Growth rate of real Penn World Tables, Mark 5.6 Dependent Variable Not applicable
GDP per worker

Initial Income Penn World Tables, Mark 5.6 1985 International Uncertain
Prices

Labor Force Growth Easterly and Levine, 1999 Computed from number Negative
and Depreciation of workers

Investment Share Penn World Tables, Mark 5.6 Real Investment share Positive
ofGDP [%] 1985
International Prices

Human Capital Barro and Lee, 2000 Secondary School Positive

Enrollment [%]

Government The Fraser Institute Scale of 0-10, with Uncertain

Consumption higher values indicating
smaller governments Nonlinear?

Governance IRIS Center, University of Scale of 0-6, with Uncertain

Maryland higher values indicating
better ratings

Security of property IRIS Center, University of 0-10, with higher values Positive
rights Maryland indicating better ratings

Political Freedom Gastil Index from "Freedom in Scale of 1-7, with 1 Uncertain
the World" representing most free Nonlinear?

Dummy for Oil World Bank classification l=OIL Positive

Exporters (OIL) O=Other
 18

Table 2: Factors affecting economic growth

Dependent Variable: Growth Rate of GDP per worker, 1975-1990

Variable Model I Model 2 Model 3 Model 4 Model 5

Intercept -0.041 -0.053 -0.30 1*** -0.306*** 0.248
(0.473) (0.312) (0.004) (0.002) (0.319)

Initial Income -0.012** -0.015*** -0.013** -0.011 ** -0.010**

(0.032) (0.007) (0.029) (0.027) (0.035)

Labor Force Growth -0.033 -0.046** -0.082*** -0.079*** -0.078**

and Depreciation (0.138) (0.038) (0.001) (0.000) (0.000)

Investment Share 0.019*** 0.020*** 0.023*** 0.022*** 0.025***

(0.001) (0.000) • (0.000) (0.000) (0.000)

Human Capital 0.006 0.008 -0.003 0.003 0.001

(0.312) (0.200) (0.585) (0.562) (0.833)

Oil Exporters 0.029** 0.043*** 0.041 *** 0.046***

(0.027) (0.002) (0.001) (0.000)

Security of Property 0.015 0.021 ** 0.024***

Rights (0.268) (0.041) (0.011)

Governance 0.005
(0.566)

Political Freedom -0.003

(0.638)

Government 0.055** 0.049** -0.538**

Consumption (0.018) (0.019) (0.0341)

Government 0.150**
Consumption squared (0.021)

R2 0.34 0.42 0.59 0.58 0.64

Adjusted R2 0.27 0.34 0.47 0.50 0.56

RootMSE 0.018 0.017 0.015 0.010 0.013

Sample size 43 43 43 43 43

All variables in logs. Parantheses contain p-values.

*, ** and *** denote significance at the 10%,5% and 1% level.
 19

Table 3: List of countries

Bangladesh Mexico
Benin Nicaragua
Chile Pakistan
Colombia Panama
Costa Rica Papua New Guinea
Denmark* Peru
Dominican Republic Philippines
Ecuador Portugal*
Ghana Sierra Leone
Greece* 'Singapore*
Guatemala South Africa
Honduras Switzerland*
Iceland* Syrian Arab Republic+
Indonesia Thailand
Ireland Trinidad and Tobago
Italy* Tunisia+
Jamaica Turkey
Kenya United Kingdom*
Korea, Rep. Uruguay
Malawi Venezuela
Malaysia Zambia
Mali

* High-income economies
+ Oil exporters
 20

Table 4: Table of Income group by Region *

~ion Income
High-Income High-Income Low Income Lower Upper Total
Non-GECD GECD Middle Middle
Income Income

East Asia and 3 2 7

Pacific

Europe and 2 3
Central Asia

Latin America and 2 7 6 15

Caribbean

Middle East and 2 2

North Africa

South Asia 2 2

Sub-Saharan 7 8
Africa

Western Europe 6 6

TotaJ 8 12 13 9 43

* World Bank Classification, 1990

 21

Table 5: Descriptive Statistics

Variable Sample Mean Standard Minimum Maximum

Size Deviation
Growth rate ofGDP 43 0.010 0.021 -0.030 0.060
per worker

Initial Income 43 9581 6845 1085 27173

Labor force growth 43 0.074 0.010 0.054 0.092

Investment Share 43 17.43· 7.29 1.28 35.87

Human Capital 43 23.02 12.96 1.95 54.05

Governance 43 2.67 1.01 0.66 5.23

Political Freedom 43 3.66 1.72 1.00 6.64

Security of Property 43 5.37 1.38 2.65 8.65

Rights
Size of government 43 7.94 1.21 4.60 10.00
 22

Table 6: Matrix of correlation coefficients

Variable Growth rate Initial Labor force Investment Human Governance Political Security of Size of
ofGDP per Income growth share capital freedom property government
worker rights
Growth rate of 1.00
GDP per worker

Initial Income 0.18 1.00

(0.4542)

Labor force -0.22 -0.45 1.00

growth (0.1564) (0.0022)

Investment share 0.48 0.56 -0.19 1.00

(0.0010) (0.0001) (0.2208)

Human Capital 0.28 • 0.77 -0.44 0.57 1.00

(0.0653) «0.0001) (0.0030) «0.0001)

Governance 0.27 0.25 -0.01 0.26 0.32 1.00

(0.0804) (0.1022) (0.9608) (0.0951) (0.0375)

Political freedom -0.05 -0.68 0.52 -0.43 -0.53 -0.13 1.00

(0.7718) «0.0001) (0.0003) (0.0041) (0.0002) (0.4006)

Security of 0.40 0.09 -0.01 0.29 0.25 0.73 -0.12 1.00

property rights (0.0084) (0.5770) (0.9523) (0.0554) (0.1089) «0.0001) (0.4550)

Size of -0.04 -0.54 0.58 -0.44 -0.41 -0.01 0.53 0.08 1.00
government (0.8164) (0.0002) «0.0001) (0.0030) (0.0059) (0.9710) (0.0002) (0.6030)

Parantheses contain p-values for the null hypothesis: Rho=O

 RESIDUALS
~
I I
b o o o o ~ o
o.j::>. 'w=' o
N
o
,..... o
o,..... o
N
o
w '='
.j::>.

lDN
KOR
MYS

SGP
THA
GR,
PRT

--
TUR
CHL
COL

DoK
G

ECU
GTM :
~
IJQ
=
"'l

...
~

::0
t:j
JAM M
ME ...
t""
I
()
NrC
"'tl
0 PAN t""
~ PER
...,0
~ ..,
0
~
~
~

8
~

~
SYR j >­t""
TUN
BGD
I ~

PAK

KEN

MI

SLE

GBR
IRL
ISL N
w
ITA
 RESIDUALS
I I I I

-
;::>
-
0 0 0 0 0 0 0
0 0w 0 0 0 0 0 0
~ N 0 N w ~

I I I I I I I
IDN
KOR
MYS
L

-
PNG
SGP
THA
<*e
PRT
TUR

-H
COL "%j
CRI iQ'
DOM
.,
=
~
N
ECU
GTM s=
0
~
trj
JAM J t""
MEX N
I
NIC
n ""C
0 PAN t""
~..., PER . 0
""'3
~ 0
I
T
IJJ 1TDV I ~

~
rf.)
.....
~
cj

BGD J ~ I ~
rf.)
PAK

N
.l:;:.
 RESIDUALS
~ ~
I
0 b 0 0 0 0
0 0
v)
0 0
..... 0..... 0 0 0oj:>.
,
oj:>. N
I I
0
!
N
I
W

I
ION
KOR

MYS

SGP

THA

~
is'
=
~
l.iJ

~
0
~
~
t""l
.&:;.
I

n "'d
0 t""l
~~ 0
~

~
\/.J
T
lTn"
0
~

~
rJ.J
SYR I
---=====- I S
e
BGD ~
~ I ~
rJ.J
PAK

KEN

--
MLI
v lWI

-SLE
ZAF
2MB

-
CHE
DNK

GBR

IRL

ISL J \ I N
VI
ITA