Download Economic Freedom and Financial Development: International Evidence

Economic Freedom and Financial Development:
International Evidence
R.W. Hafer
Distinguished Research Professor
Southern Illinois University Edwardsville
May 2012: This is an initial draft and is being circulated for comments. Please do not quote or
cite without permission of the author.
1
1. Introduction
Over the past 20 years there has been a substantial amount of research on the relationship
between financial development and economic growth. The basic finding is that the level of
financial development is a good predictor of economic growth (among others, King and Levine,
1993a,b; Levine and Zervos, 1998, Rousseau and Wachtel, 1998, Levine, et al., 2000, Levine,
2003). This line of research has expanded to encompass, among others, studies of the property
rights on economic and financial development (Acemoglu and Johnson, 2005) and on the effects
of different legal and colonial histories on financial systems (LaPorta, et al., 1998; Mahoney,
2001; Klerman, 2011, among others.)
A concurrent line of research has investigated the varied sources of economic growth.
This wide-ranging body of work covers not only the usual factors of physical and human capital,
but also considers such a number of diverse institutional factors causes such as colonial
background, religious preferences (overviews can be found in Sala-i-Martin, 2002; Barro and
Sala-i-Martin, 2004; Loayza and Soto, 2002). A number of studies focusing on the institutional
factors that promote economic growth have employed indexes to measure economic freedom as
a proxy for the underlying institutions of a country, including the size of the government, the
country’s legal structure, and the level of regulation. The weight of evidence suggests that
countries with faster economic growth are characterized as having higher levels of economic
freedom (among others, see Gwartney, et al., 2006; Weede, 2006).
In this study we consider the question “Does more economic freedom lead to a higher
level of financial development?” If we find the answer to be yes, then we have discovered yet
another route through which improving social institutions affects economic growth. While there
is evidence that more economic freedom leads to improvements in financial systems (Hartarska
2
and Nadolnyak, 2007; Crabb, 2008; Enowbi-Batuo and Kupukile, 2009) to our knowledge there
has not been a study of the direct link between economic freedom and financial development
using a large cross-section of countries. This study attempts to fill that void.
The format of the paper is as follows. The next section describes the methodology and
data used in our study. Basically, we adopt the procedure set out in Levine, et al. (2000) to
analyze the effects of economic freedom on several measures of financial intermediary
development. Section 3 presents our regression results with conclusions offered in Section 4.
Looking ahead, we find that countries with higher levels of economic freedom are more likely to
experience relatively greater development of their financial intermediaries and, by extension,
improved economic conditions.
2. Methodology and data
Our objective is to assess the role of economic freedom in explaining differences in
financial intermediary development across countries. To do this, we adopt the approach used by
Levine, et al. (2000) (hereafter, LLB). To explain observed differences in financial intermediary
development measures, LLB estimate the regression
(1)
FINANCEi = α + β [LEGAL ORIGINi] + γ log(RGDPCAPi) + εi
where FINANCE represents some measure of financial development for the ith country included
in their sample, LEGAL ORIGIN is taken as an endowment measure that captures the origins of
the ith country’s legal system, RGDPCAP is the ith country’s real per capita GDP in the initial
year of the sample period, α, β, and γ are parameters to be estimated and ε is the error term. All
of the financial measures used in LLB are averages for the period 1960-1995. Thus, the log of
real per capita RGDP is the value for 1960.
3
Amongst possible measures of financial development, LLB focus on three.1 One
measure, Liquid Liabilities, is used as an overall measure of financial depth. This variable is
calculated as the liquid liabilities of the financial system, equal to currency plus demand deposits
and interest-bearing liabilities of banks and nonbank financial intermediaries, relative to GDP.
LLB note that this is a popular measure of the overall size of the financial sector, used in
previous studies (see King and Levine, 1993a, b and the references cited therein). It does,
however, have several drawbacks. LLB recognize that because it includes deposits among
financial intermediaries, this can give rise to a degree of double counting. It also may be a poor
measure of assessing the ability of the financial sector to reduce transactions costs and reduce
informational symmetries. Still, if one accepts the hypothesis that the general size of the
financial sector is positively correlated with the provision of financial services, then Liquid
Liabilities adequately serves in measuring the development of financial intermediation across
economies.
Following King and Levine (1993a, b), LLB also utilize a measure that equals the ratio of
commercial bank assets to commercial bank plus central bank assets. This variable, which we
label Bank Assets, is an attempt to determine how much of an economy’s savings is allocated by
commercial banks relative to its central bank. The motivation is that commercial banks are profit
maximizers and, therefore, more likely to identify and pursue investments than a central bank. In
addition, given their objective functions, commercial banks probably will (should) invest in more
oversight activities, be actively engaged in risk management and the allocation of financial
resources among savers and borrowers in an efficient and socially effective manner. Even so,
LLB warn that this measure may not be the most accurate measure of the quality and quantity of
financial services provided by intermediaries.
1
This discussion is drawn from LLB, pp. 37-39.
4
Lastly, the measure Private Credit equals the ratio of credits by financial intermediaries to
the private sector by GDP. Unlike the other two measures, Private Credit isolates the role of the
private sector. LLB refer to this measure as their “preferred” measure of financial intermediary
development, unlike the measure used by King and Levine (1993a, b) which includes credits
issued by central banks and governments. Although this measure, LLB concede, does not
capture the reduction in information and transactions costs thought to be the fundamental reason
for financial intermediation, they argue that it indicates “higher levels of financial services and
therefore greater financial intermediary development.”(39)
In LLB’s study, the legal origin of an economy is used to capture a country’s
“endowment” of legal tradition. The use of legal origin draws from previous work by LaPorta, et
al. (1998) who found that the origin of a country’s legal structure is important in establishing
rules that affect various aspects of financial transactions, such as contract enforcement,
accounting standards and rules over the use and allocation of credit. In LLB legal origin is
viewed as an “endowment” that each country has, much like colonial history or geographical
location, two control variables often used in economic growth studies. Based on Reynolds and
Flores (1996) LLB identify a country’s legal origin as stemming from one of four possible
sources: English, French, German or Scandinavian. Of these, the influence of the French legal
system is geographically the widest. The French Civil Code, written in 1804 under Napoleon’s
direction, extended to all conquered lands, which included Italy and the Netherlands. It was
further spread as France colonialized parts of Africa, Indochina, the Caribbean and the Near
East. It also influenced the legal traditions of Portugal and Spain, which then was passed to their
respective colonies. At the other end of the spectrum, the legal codes of the Scandinavian
countries are probably the most direct, unadulterated descendants of the original Roman Law.
5
Since the appearance of LaPorta, et al. (1998) and LLB, there have been numerous
studies that further examine the role that legal origins play in finance and economics (Acemoglu
and Johnson, 2005; Beck and Levine, 2005; Demirguc-Kunt and Levine, 2009; Miletkov and
Wintoki, 2011). Klerman, et al., 2011 have questioned the effectiveness of this measure
compared to one controlling for colonial history. With the debate unsettled, we adopt LLB’s
specification as a basis of analysis.
Finally, the estimated equation controls for the overall level of economic development by
including the initial level of real GDP per capita. LLB use real per capita RGDP in 1960 to
control for the level of overall economic development at the beginning of the period over which
financial intermediary development is being observed. This helps to “exogenize” each country’s
economic standing from subsequent financial development.
We take equation (1) as our “baseline” regression. To it we add our measures of
economic freedom to test the hypothesis that economic freedom has a positive effect on financial
intermediary development. Thus, the extended equation of interest becomes
(1)
FINANCEi = α + β [LEGAL ORIGINi] + γ log(RGDPCAPi) + λ [ECONOMIC
FREEDOMi] + εi
where FREEDOM is the Economic Freedom of the World index (Gwartney, et al., 2010). Given
a country’s endowment of legal origin and its overall level of economic development, estimating
equation (2) allows us to test whether a country’s levels of economic freedom explains observed
cross-country differences in the development of financial intermediaries. Following LLB’s use
of initial real per capita GDP, we specify that the freedom measure should be exogenous to the
subsequent financial development.
6
To ensure adequate country coverage, we use a sample period that begins in 1980. We
chose 1980 because even though freedom measures are available beginning in 1975, the sample
is quite limited. And since the freedom data appear only every five years in this early period,
1980 is the first year that serves the purpose of providing a large sample of countries. Thus, all
freedom measures are for 1980, the initial income measure is the (log of) per capita real GDP in
1980, and all financial development indicators are averaged over the period 1980-2009.2
Our expectation is that increased economic freedom leads to improved development of
financial intermediaries. Since this is, to our knowledge, the first study to directly test this
hypothesis, the expected effect derives from the facts that numerous studies have found that
countries with higher levels of economic freedom tend to be countries that also have better
economic growth, achieve higher levels of wealth, have a healthier population and one that is
happier (Norton, 1998; Esposto and Zaleski, 1999; Dawson, 2003; Welsch, 2003; Gwartney, et
al., 2006; Inglehart, et al., 2008; Gropper, et al., 2011). We also know that, as LLB and others
have found, countries that have better financial development also have faster economic growth.
This body of evidence, take together, suggests that the social institutions that characterize
economic freedom may overlap those that determine the development of financial intermediaries.
What we would like to know is whether the link between economic freedom and economic
success may work through the financial system. Hence our direct test of the relationship
between freedom and financial development.
In this study we use the overall measure of economic freedom (Sum Freedom) and its
major subcomponents to test the foregoing hypothesis. These measures use a 10-point scale, the
higher the value the greater the economic freedom. As more detailed definitions for our
2
In an earlier version, we experimented with using the LLB data set and including the freedom measure for 1975.
While the results were qualitatively similar to those reported below, we did not feel that the use of the 1975 freedom
measure adequately “exogenized” the variable. So, we opt to use data for the broader 1980-2009 sample.
7
measures are available from Gwartney, et al. (2010), for present purposes a brief description
should suffice. The components used include a measure of government size relative to the
economy (Government), a measure of the legal structure and property rights (Legal), whether the
government undertakes actions to protect the purchasing power of the currency (Money), how
free international trade is (Trade), and a measure of regulatory intervention (Regulation). In
addition, given the nature of this study, we also include in our list of freedom measures one that
specifically captures the degree of regulations of the credit system (Credit Market Regulation).
2.a. Data
The data for the financial intermediary development measures are taken from Beck and
Demirguc-Kunt (2009).3 This is an updated version of the data set used by LLB. In the LLB
study the financial variables were averaged over the period 1960-1995. Because we wish to test
the effect of economic freedom on financial development, as discussed above we advance the
initial date to 1980. Moving the initial date to 1980 allows us to include most of the original
LLB countries as well as include several additional ones: There are 80 countries included in our
base data set. In addition, updating the sample through 2009 means that our financial measures
are averages using data for the period 1980 through 2009. The control variables included in
LEGAL ORIGIN are taken from LLB. Any country not included in the original LLB data set is
assigned a legal origin using information taken from the CIA Factbook. Real per capita GDP for
1980 is taken from the Penn World Tables. Finally, the economic freedom measures for 1980
are obtained from Gwartney, et al. (2010).
In Table 1 we report the summary statistics for the variables used in this study. Even
with the changes made, the average financial development measures reported there are roughly
similar to those reported by LLB (see their Table 1). Our average Liquid Liabilities ratio is
3
We use the most recently updated version of this data, available online from the World Bank.
8
about 50 percent compared to 43 percent in LLB. Comparing the mean our Bank Asset ratio to
LLB’s is an even closer match: 80 percent compared with 78 percent, respectively. The Private
Credit ratio is the most unequal, ours being 49 percent and theirs 38 percent. Overall, we are
confident that our set of financial measures is comparable to those used in LLB, even after the
changes made to the sample.
The summary statistics for the other variables used in the study also are found in Table 1.
Real per capita GDP in 1980 averages about $8,912 with quite a wide distribution: from
Burundi’s $368 to Switzerland’s $29,774. Statistics for the different economic freedom
measures used also are reported. Comparatively speaking, the overall measure has a much
tighter distribution compared to its component: Sum Freedom ranges from about 3 to 7, while the
other measures range from a low of zero (Trade and Credit Market Regulation) to a high of 10
Legal and Credit Market Regulation).
Table 2 reports the simple bivariate correlations between the variables. All of the
reported correlations are significant at the five-percent level or better except for those associated
with the Government and one of the Regulation economic freedom measures. Thus we adopt the
convention of reporting correlations that are not significant in bold. What we find is that
economic freedom and financial intermediary development measures are, on a whole, positively
correlated. This suggests that countries with higher levels of economic freedom in 1980 also are
those that experienced greater financial development in the ensuing decades. Moreover, note
that real per capita GDP and economic freedom (save the Government measure) are all
significantly and positively correlated. This corroborates earlier findings that greater economic
freedom is associated with higher levels of economic well-being.
9
The correlations in Table 2 also reveal that the level of economic freedom in 1980 is
positively correlated with subsequent financial development in nearly every case. To see this
graphically, we plot the financial intermediary development measures against the summary
economic freedom measure in Figures 1-3. Although the tightness of the relation varies across
finance measures, the “best-fit” line clearly indicates a positive relationship between economic
freedom and financial development across our sample of countries.4 The question to which we
now turn is how robust is that relationship once we control for other factors.
3. Regression results
To answer that question we estimate equations (1) and (2). Table 3 reports the results for
both. The “baseline” results are found in the first column under each financial measure. The
results uniformly indicate that the initial level of (log) real per capita GDP in 1980 is a highly
significant factor in explaining observed variations in the financial intermediary development
measures averaged over the 1980-2009 period. In every instance the estimated coefficient is
positive and significant at greater than a one percent level of significance. This is qualitatively
the same outcome as reported by LLB (Table 2).
Our results on the role of legal origin are not as similar to those of LLB, however. They
find that the “major message is that countries with a German legal origin have better developed
financial institutions.”(43) With the Scandinavian legal origin captured in the constant, they find
that the estimated coefficients on the English and French legal origin dummies are usually
negative while those for German legal origin are positive. Based on our sample of countries and
time period, we find that there is little difference on the effects of legal origin when the
dependent financial variable is Bank Assets, but that there are significant differences when
4
The “outlier” in Figure 1 is Japan. As found by LLB, Japan has one of the highest (in our case the highest) values
for the Liquid Liabilities measure.
10
Liquid Liabilities is the development variable being explained. We should note that the
inclusion of the income variable often had significant effects on the sign and significance of the
estimated legal origin coefficient in the LLB study. Overall, the reported R-squared and Pvalues for the regression F-test indicate that our baseline equation, as in LLB, explains a
significant amount of the cross-country variation of the three financial intermediary development
measures.
The second column under each financial measure in Table 3 reports the outcome of
estimating equation (2). One thing to note is that, even though real per capita RGDP and the
Sum Freedom measures are correlated (see Table 2), each exerts an independent effect on the
development in financial intermediaries. Although the size of the coefficient on income is
reduced slightly once economic freedom is included, its significance is not. And, in general,
adding economic freedom does little to the estimated effects from legal origin dummies. Only
for the Private Credit measure do we see a notable impact: the estimated negative coefficient on
French legal origin is now significant, and the coefficient on German legal origin is not.
Focusing on the results for the economic freedom measure, it is positive and statistically
significant at the five-percent level or better in every regression. Even after controlling for initial
income and legal origin, countries with higher levels of economic freedom in 1980 experienced,
on average, a higher average level of development in financial intermediation. Not only is the
freedom variable statistically significant, but its economic effects are not miniscule. A onestandard-deviation increase in economic freedom results in about a 7.2 point increase in the
Liquid Liabilities ratio, which amounts to about one-quarter of its standard deviation. For the
Private Credit measure, the effect is a bit larger: a one-standard deviation increase in economic
freedom produces, on average, about a 13 point increase, or about a one-third of a standard
11
deviation. A similar increase in economic freedom raises the average country’s Bank Assets
ratio by over 5 points, or one-third of a standard deviation.
Overall, the reported R-squares and the P-values of the regression F-test indicate that a
significant portion of the variability in these financial intermediary development indicators is
being explained by the variables used, including economic freedom.
Before turning to the results based on using the components of the overall economic
freedom index, we should report that we tried several other variants of equation (2). We
recognize that this is a lean specification, relying only on a handful of control variables. But
what others to include? We experimented with other variables that often are used to account for
various aspects of social and human capital in a country. The list includes life expectancy,
educational attainment (updated Barro-Lee (1996) estimates of average years completed and
percent of the adult population with a BA) and an overall measure of human development, the
so-called Human Development Index (HDI). Values for each measure, like income and
economic freedom were for 1980.5 We do not report the estimation results not only to save
space, but also because including these measures (individually and as a group) did not alter our
finding that economic freedom has a positive and significant effect on the three financial
intermediary development indicators. (These results are available upon request.) Not too
surprising, though, is the fact that including life expectancy and HDI significantly reduce the
statistical significance of initial income. This reflects the fact that all three are highly correlated
5
Due to data limitations for the HDI measure, we use 1985 observations for Cyprus and Uganda. Countries not
included in this estimation include the Bahamas, Madagascar, and Nigeria. For these countries the index is not
available in any year reasonably close to 1980.
12
and, in effect, are capturing similar initial conditions.6 Richer countries tend to have higher life
expectancies and by construct a higher HDI index.
Even though the financial development indexes are based on data that run through 2009
and may not be significantly impacted by the financial crises and recessions that began a year
earlier, we nevertheless want to make sure that no untoward influences from this turbulent period
are driving our results. Consequently, we re-estimated the equations averaging the financial
development data using data from 1980 to 2005. Again we do report those results (they are
available upon request) since they are qualitatively unchanged from those found in Table 3. We
find that a higher level in economic freedom is associated with a higher level of financial
intermediary development over this truncated sample.
Given the results in Table 3 and our other experiments, we are confident in concluding
that an increase in economic freedom is an important factor in the development of a country’s
financial intermediaries and, given previous research, in the economic well-being of its citizens.
3.a. Results using freedom components
We have thus far established that greater financial intermediary development is
associated with higher levels of economic freedom. Does this hold when we consider the more
narrow measures of economic freedom? The answer is found in Table 4.
To conserve space, and since it is our focus, we report only the estimated coefficients on
the six component measures of freedom used to replace the Summary Freedom measure in
estimating equation (2). Overall, the results indicate that there is a positive relationship between
the development of financial intermediaries and the more narrow measures of economic
freedom. An improved legal system and protection of property rights increases both the Private
6
The correlation between income and HDI is 0.85; for income and life expectancy the correlation is 0.74. This is
not true for the correlation with freedom, however. The correlation between HDI and freedom is 0.53 and between
life expectancy and freedom the correlation is 0.49.
13
Credit and Bank Assets indicators of financial development.7 The effect from a one-standard
deviation increase in this measure of freedom yields an increase of a 30 to 40 percent increase
relative to a one standard deviation in these financial variables. Governments that undertake
policies to protect the purchasing power of their currency, usually through low-inflation policies,
also have, on average, higher Liquid Liabilities and Private Credit measures. Although we find
that Regulation, here a measure of general business regulation by the government, is significant
only for the Bank Assets measure, the more specific Credit Market Regulation measure
positively and significantly effects both Private Credit and Bank Assets.8 Economically, a one
standard deviation increase in the Credit Market Regulation measure (a reduction in the amount
of regulation) increases the Private Credit and Bank Assets measures by about one-fifth of their
respective standard deviations. This indicates that countries with more intrusive regulations that
hinder activities by credit market participants, reduced the development of their financial
intermediaries.
Of all the component measures tested, two stand out. One is the Government Size
measure. It stands out because it takes an unexpected negative sign. This suggests that
economies with larger governments actually have improved development in their financial
intermediaries. This observation is tempered by the fact that it never achieves statistical
significance. The other freedom measure that deserves special mention is the Trade variable,
which captures the country’s freedom in international trade. What the results in Table 4 show is
that, regardless of the financial indicator, greater freedom to engage in international trade leads
to a greater development in a country’s financial intermediaries. And the economic effects are,
7
This result is based on a reduced sample due to lack of data for the Legal variable. This sample consists of 67
countries.
8
Due to data limitations, this result for Credit Market Regulation is based on a sample that omits Paraguay. For
Regulation, Central African Republic, El Salvador, Haiti and Paraguay are omitted.
14
on average, the largest of the freedom measures used. A one standard deviation increase in
Trade generates an increase in each financial intermediary development indicator that is greater
than one-third of their respective standard deviations.
4. Conclusion
In summary, the evidence presented in Tables 3 and 4 lead us to conclude that greater
economic freedom leads to an increase in the development of a country’s financial
intermediaries. Even after controlling for several conditioning variables, economic freedom is
found to have a positive and statistically significant effect on three popular indicators of financial
development.
We consider this study as a “pre-quel” to previous work in both the financial
development and economic growth literature. Many studies have examined the factors that cause
economic growth, one of which is economic freedom, and others have tested for the effects of
financial development on economic growth. No one to our knowledge has asked whether the
finance-growth nexus reflects indirect influences from economic freedom that operate through
the financial system. Our findings suggest that this is a viable explanation. Countries that have
higher levels of economic freedom, on average, exhibit greater levels of financial intermediary
development. And since these countries are likely to grow faster as a result, our results provide
yet another link from economic freedom to improved economic well-being.
We recognize that this study is only a start-point for further analysis. Our standard crosssectional analysis could be altered to account for changes across time and space. Examining the
link between economic freedom and financial development could be done using more recent data
on freedom by using panel estimation techniques. Moreover, examining the raw data for the
financial development indicators shows that some countries have experienced significant
15
changes in the ratios between 1980 and 2009. Our cross-section estimation admittedly ignores
such changes. Testing for temporal effects from changes in freedom to changes in financial
development also is an area for further research.
16
References
Acemoglu, D., and Johnson, S. 2005. Unbundling institutions. Journal of Political Economy
113(5): 949-995.
Barro, R.J., and Lee, J. 1996. International measures of schooling years and schooling quality.
American Economic Review, Papers and Proceedings 86, 218-223.
Barro, R.J., and Sala-i-Martin, X. 2004. Economic Growth. Cambridge: MIT Press.
Beck, T., and Levine, R. 2005. Legal institutions and financial development. Menard and
Shirley, eds. Handbook of New Institutional Economics 251-278.
Beck, T., and Demirguc-Kunt, A. 2009. Financial institutions and markets across countries and
over time: data and analysis. World Bank Policy Research Working Paper No. 4943 (May).
Crabb, P. 2008. Economic freedom and the success of microfinance institutions. Journal of
Development Entrepreneurship 13(2): 205-219.
Dawson, J.W. 2003. Causality in the freedom-growth relationship. European Journal of
Political Economy 19: 479-495.
Demirguc-Kunt, A., and Levine, R. 2009. Finance and inequality: theory and evidence. Annual
Review of Financial Economics 1(1): 287-318.
Enowbi-Batuo, M., and Kupukile, M. 2009. How can economic and political liberalisation
improve financial development in African countries? Journal of Financial Policy 2(1): 35-59
Esposto, A., and Zaleski, P.A. 1999. Economic freedom and the quality of life: an empirical
analysis. Constitutional Political Economy 10: 185-197.
Gropper, D.M., Lawson, R.A., and Thorne, J.T., Jr. 2011. Economic freedom and happiness.
Cato Journal 31(2): 237-255.
Gwartney, J.D., Holcombe, R.G., and Lawson, R.A. 2006. Institutions and the impact of
investment on growth. Kyklos 59: 255-273.
Gwartney, J.D., Hall, J.C., and Lawson, R.A. 2010. Economic Freedom of the World: 2010
Annual Report. The Fraser Institute.
Hartarska, V., and Nadolnyak, D. 2007. Do regulated microfinance institutions achieve better
sustainability and outreach? Cross-country evidence. Applied Economics 39: 1207-1222.
Inglehart, R., Foa, R., Peterson, C., and Welzel, C. 2008. Development, freedom, and rising
happiness: a global perspective (1981-2007). Perspectives of psychology science 3(4): 264-285.
17
King, R.G., and Levine, R. 1993a. Finance and growth: Schumpeter might be right. Quarterly
Journal of Economics 108: 717-738.
King, R.G., and Levine, R. 1993b. Finance, entrepreneurship, and growth: theory and evidence.
Journal of Monetary Economics 32, 513-542.
Klerman, D.M., Mahoney, P.G., Spamann, H. and Weinstein, M.I. 2011. Legal origin or colonial
history? Journal of Legal Analysis 3(2): 379-409.
LaPorta, R., Lopez-de-Silanes, F., Shleifer, A., and Vishny, R.W. 1998. Law and finance.
Journal of Political Economy 106: 1113-1155.
Levine, R. 2003. More on finance and growth: more finance, more growth? St. Louis Federal
Reserve Bank Review July/August: 31-46.
Levine, R., and Zervos, S. 1998. Stock markets, banks, and economic growth. American
Economic Review 88:537-558.
Levine, R., Loayza, N., and Beck, T. 2000. Financial intermediation and growth: Causality and
causes. Journal of Monetary Economics 46: 31-77.
Loayza, N., and Soto, R. 2002. The sources of economic growth: an overview. Loayza and Soto,
eds., Economic Growth: Sources, Trends and Cycles. Santiago: Central Bank of Chile.
Mahoney, P.G. 2001. The common law and economic growth: Hayek might be right. Journal of
Legal Studies 30(2): 503-525.
Miletkov, M., and Wontoki, M.B., 2011. Financial development and the evolution of property
rights and legal institutions.
Norton, S.W. 1998. Poverty, property rights, and human well-being: a cross-national study. Cato
Journal 18(2): 233-245.
Reynolds, T.H., and Flores, A.A. 1996. Foreign law: current sources of codes and basic
legislation in the jurisdiction of the world. Fred B. Rothman and Co., Littleton, Co.
Rousseau, P.L., and Wachtel, P. 1998. Financial intermediation and economic performance:
historical evidence from five industrial countries. Journal of Money, Credit, and Banking 30:
657-678.
Sala-i-Martin, X. 2002. 15 years of new growth economics: what have we learned? Central
Bank of Chile Working Paper No. 172.
Weede, E. 2006. Economic freedom and development: new calculations and interpretations.
Cato Journal 26(3): 511-524.
18
Welsch, H. 2003. Freedom and rationality as predictors of cross-national happiness patterns: the
role of income as a mediating variable. Journal of Happiness Studies 4: 295-321.
White, H. 1980. A heteroskedasticity-consistent covariance matrix estimator and a direct test for
heteroskedasticity, Econometrica 817-838.
19
Table 1
Summary statistics
Variable
Mean
Std Dev
Max
Min
Liquid Liabilities
0.499
0.298
1.913
0.120
Private Credit
0.493
0.370
1.498
0.036
Bank Assets
0.799
0.177
0.995
0.253
RGDP/cap
8,912.00
8343.66
29774.94
368.19
Sum Freedom
5.520
0.965
7.660
3.210
Government
5.102
1.477
9.100
1.600
Legal
5.504
2.720
10.000
0.900
Money
6.265
1.922
9.600
0.800
Trade
5.076
20.12
9.000
0.000
Credit Market
Regulation
5.910
2.259
10.000
0.000
Regulation
5.850
1.771
9.000
0.500
20
Table 2
Correlations
Liquid
Liabilities
Private
Credit
Bank
Assets
Real
GDP/capita
Sum
Freedom
LiqLiab
PrvtCred
0.834
BnkAssets
0.557
0.689
RGDP/cap
0.570
0.739
0.616
Sum Freedom 0.520
0.644
0.608
0.586
Government
-0.135
-0.111
-.047
-0.133
0.343
Legal
0.474
0.718
0.671
0.735
0.705
Money
0.408
0.352
0.264
0.432
0.712
Trade
0.565
0.698
0.639
0.587
0.691
Credit Market
Regulation
0.355
0.507
0.532
0.466
0.688
Regulation
0.260
0.384
0.232
.0594
0.194
_________
Note: Correlations in bold are not significant at the five percent level of significance. All others
are significant at least at the fiver percent level.
21
Table 3
Baseline regression results
Variable
Measure of financial intermediary development
Liquid liabilities
Private credit
Bank assets
RGDP
0.125
(0.000)
0.090
(0.000)
0.199
(0.000)
0.139
(0.000)
0.103
(0.000)
0.081
(0.000)
British
0.278
(0.000)
0.212
(0.002)
0.064
(0.528)
-0.049
(0.553)
0.0384
(0.172)
-0.012
(0.667)
French
0.202
(0.006)
0.135
(0.056)
-0.070
(0.454)
-0.186
(0.017)
0.027
(0.435)
-0.023
(0.499)
German
0.482
(0.005)
0.398
(0.008)
0.252
(0.036)
0.107
(0.248)
0.021
(0.708)
-0.042
(0.459)
Sum Freedom
Adj-R2
Prob(F-test)
__________
0.075
(0.026)
0.383
0.000
0.414
0.000
0.131
(0.000)
0.560
0.000
0.630
0.000
0.057
(0.006)
0.442
0.000
0.500
0.000
Notes: All equations include a constant term. Probability levels reported in parentheses. All
equations are estimated using White’s (1980) heteroskedasticity-consistent standard errors.
22
Table 4
Regression results using freedom components
Liquid
liabilities
Private
credit
Bank
assets
Government
-0.014
(0.350)
-0.003
(0.862)
0.001
(0.968)
Legal
0.010
(0.570)
0.051
(0.004)
0.019
(0.024)
Money
0.032
(0.010)
0.026
(0.090)
0.006
(0.424)
Trade
0.050
(0.000)
0.077
(0.000)
0.030
(0.002)
0.011
(0.365)
0.035
(0.014)
0.018
(0.068)
0.011
(0.572)
0.017
(0.270)
0.021
(0.035)
Freedom measure
Credit Market
Regulation
Regulation
__________
Notes: All regression include constant, legal origin dummies and log of real per capita GDP in
1980. Probability levels reported in parentheses. All equations are estimated using White’s
(1980) heteroskedasticity-consistent standard errors.
23
Figure 1
Scatter plot of liquid liabilities and freedom
2.0
LIQLIAB
1.6
1.2
0.8
0.4
0.0
3
4
5
6
7
8
SUMFREEDOM1980
24
Figure 2
Scatter plot of private credit and freedom
1.6
1.4
PRVTCREDIT
1.2
1.0
0.8
0.6
0.4
0.2
0.0
3
4
5
6
7
8
SUMFREEDOM1980
25
Figure 3
Scatter plot of bank assets and freedom
1.6
1.4
PRVTCREDIT
1.2
1.0
0.8
0.6
0.4
0.2
0.0
3
4
5
6
7
8
SUMFREEDOM1980
26