Journal of Financial Economics 69 (2003) 5–50
The great reversals: the politics of financial
development in the twentieth century
$
Raghuram G. Rajan*, Luigi Zingales
The University of Chicago Graduate School of Business, 1101 E. 58th St., Chicago, IL 60637, USA
Abstract
The state of development of the financial sector does not change monotonically over time.
In particular, by most measures, countries were more financially developed in 1913 than in
1980 and only recently have they surpassed their 1913 levels. To explain these changes, we
propose an interest group theory of financial development where incumbents oppose financial
development because it breeds competition. The theory predicts that incumbents’ opposition
will be weaker when an economy allows both cross-border trade and capital flows. This theory
can go some way in accounting for the cross-country differences in, and the time-series
variation of, financial development.
r 2003 Elsevier Science B.V. All rights reserved.
JEL classification: G100; G180; G200; G380; O160; P000
Keywords: Financial markets; Growth; Politics; Financial development; Reversals; Trade; Capital flows
$
This paper is a development of some ideas in a previous working paper entitled ‘‘The Politics of
Financial Development.’’ We thank the Bradley Foundation, the George J. Stigler Center for the Study of
the Economy and the State, the Center for Research in Securities Prices, the Kauffman Foundation, and
the World Bank for funding support. Rajan also thanks the National Science Foundation and M.I.T. for
research support. Claudio Callegari, Henrik Cronqvist, Shola Fafunso, Isidro Ferrer, Jorg Kukies, Roger
Laeven, Jamers Mello, Galina Ovtcharova, Nahid Rahman, Sofia Ramos, Ruy Ribeiro, Amir Sasson, and
Elfani Wen provided excellent research assistantship and Arnoud Boot, Pratip Kar, Claus Parum, Kristian
Rydqvist, and Elu Von Thadden provided invaluable help. We benefited from comments by Lucian
Bebchuk, Stijn Claessens, Peter Hogfeldt, Louis Kaplow, Colin Mayer, Mark Ramseyer, Eric Rasmussen,
Mark Roe, Andrei Shleifer, Richard Sylla, and an anonymous referee.
*Corresponding author. Tel.: +1-773-702-4437; fax: +1-773-834-8172.
E-mail address: raghuram.rajan@gsb.uchicago.edu (R.G. Rajan).
0304-405X/03/$ - see front matter r 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0304-405X(03)00125-9
1. Introduction
There is a growing body of evidence indicating that the development of a country’s
financial sector greatly facilitates its economic growth (e.g., Demirguc-Kunt and
Maksimovic, 1998; King and Levine, 1993; Jayaratne and Strahan, 1996; Rajan and
Zingales, 1998a). Why then do so many countries still have underdeveloped financial
sectors?
The simple answer, and one favored by many economists, is the absence of
demand. Certainly demand is a prime driver of financial development, but it cannot
be the only explanation. Demand (as proxied for by level of industrialization or
economic development) cannot explain why countries at similar levels of economic
development differ so much in the level of their financial development. For instance,
why was France’s stock market much bigger as a fraction of its gross domestic
product (GDP) than markets in the United States in 1913, even though the per capita
GDP in the United States was not any lower than France’s? It is hard to imagine that
the demand for financing in the United States at that time was inadequate. At the
time, the demand for more, and cheaper, credit was a recurrent theme in political
debates in the United States, and it was among the most industrialized countries in
the world even then.
An alternative explanation is that there are structural impediments to supply rising
to meet demand. Perhaps a country does not have the necessary levels of social
capital (Guiso et al., 2000) or ‘‘savoir faire’’ to create a viable financial sector (e.g.,
Bencivenga and Smith, 1991; Greenwood and Jovanovic, 1990). Or perhaps it has
not inherited the right legal, cultural, or political system. In particular, the seminal
work of La Porta et al. (1997, 1998) shows that countries with a Common Law origin
seem to have better minority investor protection, and furthermore, these countries
have more highly developed equity markets. There has been some debate as to the
precise channel through which a country’s institutional inheritance affects its
financial development (e.g., Berglof and Von Thadden, 1999; Coffee, 2000; Holmen
and Hogfeldt, 2000; La Porta, et al., 1999a, 1999b; Rajan and Zingales, 1999; Stulz
and Williamson, 2001). Some question whether the influence of certain forms of Civil
Law heritage can be distinguished from the influence of a Common Law heritage
(e.g., Beck et al., 1999). Yet, there is a burgeoning literature suggesting that a
country’s ‘‘structure’’ matters.
There are other implications, however, of structural theories of financial
development. For instance, once a country has overcome the structural impediments,
the supply of finance should rise to meet demand. In other words, we should not see
measures of financial development waxing and waning independent of demand.
Similarly, conditional on demand, the relative position of different countries should
not change dramatically over time. If some countries have a system that is pre-
disposed towards finance, that pre-disposition should continue to be relatively strong
since structural factors are relatively time-invariant.
To test these implications, we collect various indicators of financial development
for developed countries over the twentieth century. By most measures, countries
were more financially developed in 1913 than in 1980 and only recently have they
R.G. Rajan, L. Zingales / Journal of Financial Economics 69 (2003) 5–506
surpassed their 1913 levels. Furthermore, even after controlling for the different
levels of industrialization, the pattern across countries is quite different from the
1990s. In 1913, France’s stock market capitalization (as a fraction of GDP) was
almost twice that of the United States (0.78 vs. 0.39) even though the French Civil
Code has never been friendly to investors (La Porta et al., 1998). By 1980, roles had
reversed dramatically. France’s capitalization was now barely one-fourth the
capitalization in the United States (0.09 vs. 0.46). And in 1999, the two countries
seem to be converging (1.17 vs. 1.52). More generally, in 1913, the main countries of
continental Europe were more developed financially than the United States. What is
especially interesting is that indicators of financial development fell in all countries
after 1929, reaching their nadir around 1980. Since then, there has been a revival of
financial markets.
In fact, in contrast to the findings of La Porta et al. (1997) for the 1990s, we find
that countries with Common Law systems were not more financially developed in
1913. There is some indication that these differences had to do with differences in
financial infrastructure. Tilly (1992) indicates that corporate share issues in Germany
in the beginning of the Twentieth Century were greater than in England. He suggests
this is because of the ‘‘paucity of information and relatively weak financial controls
on the operations of company founders and insiders’’ (p. 103) in England. The
common wisdom today is the reverse, that German corporations are much less
transparent than corporations in the United Kingdom, as reflected by their lower
scores on accounting standards.
The disruption in demand caused by the Great Depression and World War II are
not sufficient to explain the reversal in financial markets. The economies of the
hardest-hit countries recovered within a decade or two. Why did it take financial
markets until the late 1980s to stage a recovery? Moreover, such a delay was not seen
after the World War I.
All this is not to suggest that structural theories are incorrect, but that they are
incomplete. A theory with a more variable factor is needed to explain both the time-
series variation in financial development as well as the cross-sectional differences. In
our view, the strength of political forces in favor of financial development is a major
variable factor. The challenge for such a theory is to identify who is opposed to
something as economically beneficial as financial development. We believe that
incumbents, in the financial sector and in industry, can be hostile to arm’s length
markets. This is because arm’s length financial markets do not respect the value of
incumbency and instead can give birth to competition. There are occasions, however,
when the incentives, or the ability, of incumbents to oppose development is muted.
In particular, we argue that when a country’s borders are open to both trade and
capital flows, we see the opposition to financial development will be most muted and
development will flourish.
Of course, the decision to open to trade and capital flows is also partly political.
This raises two questions. First, why do some countries become more open than
others, or open up at some times rather than at others—do the incumbents not
oppose opening up? And second, how can we provide evidence of a causal link rather
than simply a correlation: How can we argue that the link between openness and
R.G. Rajan, L. Zingales / Journal of Financial Economics 69 (2003) 5–50 7
financial development should be interpreted as one causing the other rather than
simply as evidence that incumbents who favor openness also favor financial
development?
Let us answer the first question first. Some countries have no choice. Because they
are small, or because they are close to other countries, they are likely to have more
trade. Therefore, these countries are likely open for reasons that are not political.
Also, even if the decision is political, countries’ decisions whether to open up are
likely strategic complements. If important parts of the world are open, then natural
leakages across borders (the gray trade, smuggling, under-invoicing, over-invoicing,
etc.) are likely to be high and make it hard for a country to remain closed. Moreover,
groups that are in favor of openness (for example, exporters) are likely to gain in
prospective profitability and strength relative to those who rely on controls, and they
are likely to have more success in pressing for openness (e.g., Becker, 1983). The
economic importance of other countries that are open can be thought of as largely
exogenous to a country’s domestic politics.
These observations suggest ways to test whether openness has a causal effect.
First, in examining the link between trade openness and financial development, we
instrument trade openness with a measure of natural openness (largely based on a
country’s distance from its trading partners) developed by Frankel and Romer
(1999). We thus focus on the exogenous component of a country’s trade. Because
distance matters less for capital, we do not have a similar instrument for cross-border
capital flows. But precisely because capital is more mobile, the strategic
complementarities in cross-border capital flows are likely to be stronger. So we
can use world-wide cross-border capital flows over time as an exogenous measure of
whether countries are more open to capital flows. International capital mobility is
high both in the beginning and towards the end of the twentieth century for most
countries. Thus, we test in the cross-section of countries if financial development is
positively correlated with the exogenous component of a country’s openness to trade
(correcting for the demand for finance), both in the beginning of the century and
towards the end of the century, and it is.
By contrast, in the intermediate periods (from the 1930s to the 1970s) when cross-
border capital flows had dwindled to a trickle for a variety of reasons, we find that
trade openness did not have as strong a positive correlation (if at all) with financial
development. These findings suggest that it takes the combination of openness in
product and financial markets to mute incumbent incentives to oppose financial
development. They also suggest a rationale for why indicators of financial
development fell between the 1930s and the 1970s. Cross-border flows, especially
of capital, were relatively small, so incumbents could oppose financial development
without constraints.
We are, of course, not the first to point to the influence of private interests on
financial development, though our focus is quite different from previous work.
Jensen (1991) argues that legislation motivated by potential targets crimped the
market for corporate control even while it was having salutary effects on US
industry. Kroszner and Strahan (1999) explain the timing of financial liberalization
across states in the United States in the 1970s and 1980s with variables that relate to
R.G. Rajan, L. Zingales / Journal of Financial Economics 69 (2003) 5–508
the power of private interest groups. Morck et al. (2000) find that the share prices of
heir-controlled Canadian firms fell on news that the Canada–US free-trade
agreement would be ratified. One reason they suggest is that the treaty had a
provision for greater capital market openness, which would reduce the advantage
heir-controlled firms had from access to capital. Bebchuk and Roe (1999) argue that
corporate governance regimes will be strongly influenced by the initial positions of
owners. Our paper is related to all these in that we also emphasize the role of private
interests in retarding financial development, but we differ in that we attempt to find
general patterns across countries.
We will postpone a discussion of the other related literature until we present the
theoretical reasoning and tests. The rest of the paper is as follows. Section 2 describes
how we collect the data and presents measures of financial-sector development in
different countries at various points in the twentieth century. Section 3 presents our
interest group theory of why some countries develop their financial systems (and
others not) and argues why this could explain the reversals in the data. Section 4 tests
both the time-series and cross-sectional implications of this theory. Section 5
concludes.
2. Evolution of financial development over the twentieth century
We are faced with two problems in analyzing the historical evolution of financial
development over the twentieth century. First, it is difficult to obtain reliable sources
for historical information about financial markets. In Appendix A, we describe how
we deal with this problem. The second problem is how to measure financial
development.
2.1. What do we mean by financial development?
The right measure would capture the ease with which any entrepreneur or
company with a sound project can obtain finance, and the confidence with which
investors anticipate an adequate return. Presumably, also, a developed financial
sector can gauge, subdivide, and spread difficult risks, letting them rest where they
can best be borne. Finally, it should do all this at low cost.
In our view, the most important word in the above definition is ‘‘any.’’ In a perfect
financial system, it will be the quality of the underlying assets or ideas that will
determine whether finance is forthcoming, and the identity of the owner (to the
extent it is orthogonal to the owner’s capability of carrying out the project) will be
irrelevant. Because our focus is on how easy it is to raise finance without prior
connections or wealth, our measures of financial development will emphasize the
availability of arm’s length market finance (and if the data were available, the
availability of non-relationship-based bank finance).
This choice is not innocuous. In some financial systems, capital is easily available
for anyone within a circle of firms and financiers, but it does not percolate outside
(e.g., Hellwig, 2000; Rajan and Zingales, 1998b). Most investment opportunities
R.G. Rajan, L. Zingales / Journal of Financial Economics 69 (2003) 5–50 9
