What Shakes It? What Shapes It?
©2005 International Monetary Fund
Reform: What Shakes It? What Shapes It?]
In an unliberalized financial system, the government plays a large role in determining who gives and receives credit and at what price. Governments may offer various rationales for maintaining such a system, including, for example, preserving financial stability or channeling resources to support government industrial policy, give priority sectors access to low-cost credit, or finance a budget deficit.
Proponents of liberalization point out that financial development is strongly associated with economic growth. They argue that the allocation of capital is more efficient in a competitive financial system and that higher real interest rates stimulate saving, thereby increasing the funds available to finance investment. Moreover, government-allocated credit tends to be characterized by poor lending decisions, weak repayment discipline, and government corruption, since those granted access to capital (usually at low rates) may buy influence to protect their favored positions.
While liberalization is not without its critics—who question the link between liberalization and higher saving rates and suggest that it may make some countries more vulnerable to crisis—financial sector reform was high on the agenda of policymakers during the last quarter of the twentieth century. Countries around the world, developed and developing alike, liberalized their financial systems—allowing markets to set interest rates, eliminating controls so that capital could flow freely across borders, and opening their doors to foreign financial firms.
But there were significant differences in the pace and scale of reform. Reforms were swift in some countries, sluggish in others; some countries merely tweaked their financial sectors, while others overhauled them. A large and technically sophisticated literature has examined the consequences of financial sector liberalization, but the factors triggering reform have received less attention. Although there have been case studies supporting various economic and political theories about the forces driving reform, little statistical testing has been done.
To overcome this gap, we compiled a cross-country database—the first large cross-country database focused specifically on financial liberalization—that allowed us to compare the experiences of different countries and analyze our findings objectively, in the hope of uncovering the factors that triggered reform and influenced its timing and extent—why, when, and how much.
This economic issue, which was prepared by Charles Gardner, is based on our study Financial Reform: What Shakes It? What Shapes It? (American Economic Review, March 2005). The working paper version (IMF Working Paper No. 03/70, April 2003), which presents our methodology and findings in greater detail, is available free of charge at www.imf.org/pubs.
Given the difficulties inherent in undertaking any kind of reform, including financial sector reform, it is not surprising that inertia often sets in. The uncertainty over who will win and who will lose, the opposition of elite or special interest groups—these are just two of the issues facing policymakers considering reform. And yet many countries did reform their financial sectors during the last quarter of the twentieth century. What enabled—or forced—them to challenge the status quo?
Previous studies have identified three possible triggers for reform: discrete events, or shocks; a learning process that grew out of new information or success with initial reform measures; and a governments political ideology, institutions, or structure. The lack of cross-country data on financial liberalization has made it difficult, however, to analyze the timing, direction, and size of policy changes in a way that allows patterns to emerge. Moreover, financial liberalization consists of both discrete changes that are easy to identify and gradual changes that become visible only over time, while existing measures of liberalization refer either to a one-time change in the rules or to continuous proxies, such as the level of financial development.Thus, what is needed, in addition to cross-country data, is an index that combines both types of changes—discrete and gradual—and that looks at reform setbacks as well as progress.
The database that we compiled covers 35 economies, from the poorest and least developed to the richest, over 197396, and tracks both kinds of changes. Based on these data, we constructed an index that allows for a more precise determination both of how large a role different types of events play in triggering reform and of how the type of event influences the timing of reform.
We used six measures of financial sector repressiveness, aggregated on an annual basis for each of the 24 years covered, to capture the different dimensions of financial reform:
In each of these categories, we gave the economies a score on a graded scale: 0 for fully repressed, 1 for partially repressed, 2 for largely liberalized, and 3 for fully liberalized. We used some guidelines to reduce the subjectivity of the scores. For example, interest rates were considered fully repressed if the government set all interest rates, partially repressed if interest rates were allowed to vary within a band or were subject to a ceiling, largely liberalized if some interest rates were allowed to be completely determined by the market (or if new floating-rate instruments were introduced), and fully liberalized if all interest rate restrictions were removed.
We adjusted scores to reflect policy changes. In some cases—such as when a country privatized all state-owned banks at once or abolished controls on all interest rates—scores in the relevant category jumped more than one unit. Reversals of liberalization—such as the imposition of capital or interest rate controls—lowered an economys score.
Naturally, liberalization tends to occur simultaneously in all six categories, but some are much more highly correlated than others. The highest correlations (ranging from 0.76 to 0.82) occur among the three measures most frequently used as indicators of financial repression—controls on credit, interest rates, and international financial transactions. The elimination of entry barriers and regulations is somewhat less correlated with the others, and financial sector privatization is the least correlated.
We obtained an index of overall financial liberalization for each economy in each year by aggregating all six measures. For the sake of convenience, we used the sum of the individual components, and, since each of the six indices can take on values between 0 and 3, the totals fell between 0 for full repression and 18 for full liberalization. Outcomes for 1996 ranged from 6 for Taiwan Province of China, India, and Nepal, to 18 for Canada, New Zealand, and the United Kingdom. (See pages 1418 for the year-by-year data for the 35 countries in the sample.)
The index offers a useful profile of liberalization over the period in the 35 individual economies, in the major regional groupings, and in the world. Globally, it shows that despite stops, gaps, and reversals, financial sector liberalization advanced through much of the world in the last quarter of the twentieth century. Countries in all income groups liberalized, although higher-income economies were more liberalized than lower-income economies (see Figure below, top panel).
While the trends appear smooth for country averages, individual countries experienced long periods during which no policy changes took place. To allow a closer, more informative look at the process, we created five categories for policy changes in each country-year. A decrease in the financial liberalization measure of 3 or more points is classified as a large reversal, a decrease of 1 or 2 points as a reversal, an increase of 1 or 2 points as a reform, an increase of 3 or more points as a large reform, and no policy changes as status quo observations.
Status quo observations represent the large majority of observations—over 76 percent of the whole sample. Reforms constitute another 15 percent of the sample, and large reforms account for another 5 percent. Reversals were relatively rare, and large reversals even more so. There were only two large reversals in the sample—Argentina in 1982 and Venezuela in 1994.
Countries in the major geographical regions tended to liberalize their financial sectors at roughly the same time and in roughly the same way (see Figure, bottom panel). For example, most of the reforms in Latin America were clustered in the late 1980s and early 1990s, with the exception of early reforms in Argentina and Chile in the 1970s. The two exceptions also illustrate that reform is not a steady march forward; both countries reversed policy reforms during the debt crisis of 198283.
The process of financial liberalization in East Asia was much more gradual. The East Asian economies took small steps to open up their financial sectors in the early 1980s; in most cases, the reform process stretched over a decade. In the South Asian countries, reforms occurred in the early to mid-1990s, except for Sri Lanka, which undertook a major reform effort in 1978; however, financial sectors in South Asia remained at least partially repressed even at the end of the sample period.
Four of the industrial countries in the study—Canada, Germany, the United Kingdom, and the United States—already had largely liberalized financial sectors at the start of the period. The rest of the countries in the Organization for Economic Cooperation and Development (OECD) started out with relatively repressed financial systems but gradually began to liberalize them in the late 1970s and early 1980s; their financial sectors are now largely or fully liberalized. Only New Zealand adopted a one-shot approach, undertaking most of its financial reforms in 198485. (For regional groupings, we included two OECD members, Korea and Mexico, in their regions rather than in the OECD group.)
Our database yielded further insights when analyzed from another angle—the political economy of reform. The starting point for our analysis was that the status quo would persist so long as the benefits of maintaining it outweighed the costs for those determining the timing and pace of reforms. We examined the impact of the three different triggers identified in the literature and described above—shocks; learning; and government ideology, institutions, and structure—from this perspective.
Shocks are one-time events that trigger an almost immediate realignment of decision-making forces—for example, a new government coming to power, changes in the global economy, and conditions imposed by international financial institutions such as the World Bank and the IMF on countries seeking to borrow.
According to the so-called honeymoon hypothesis advanced in earlier studies, policy reforms can be expected at the beginning of an electoral cycle, right after a new government has come to office, rather than later, when an incumbent government is seeking reelection—particularly if the benefits of reform will be realized over the long term and the costs felt in the short term. The evidence on this is mixed, however. It is true that the tendency toward maintaining the status quo weakens significantly during a chief executives first year in office, as shown by the drop of status quo observations in those periods from 78 percent to 70 percent (see Table 1), and the likelihood of large-scale reform increases markedly. But a change of government may also lead to a reversal of liberalization—for example, during 197396, new democratic leaders in Argentina, Bolivia, and Brazil pursued expansionist policies after they took office, delaying needed reform. In contrast, financial reform is not likely during periods of drastic political change, such as coups détat and the imposition or lifting of martial law. In fact, our index showed that there was a slight increase in the relative frequency of status quo observations during such periods.
The impact of balance of payments crises and banking crises, recessions (defined simply as negative economic growth), and high inflation (defined as an annual inflation rate of over 50 percent) is shown in Table 2 (page 8). During periods marked by any one of these four adverse economic events, the relative frequency of status quo observations falls. However, the extent, and even the direction, of the policy change varies with the nature of the adversity. In response to a balance of payments crisis, the likelihood of a large reform increases from 3.1 percent to 9.7 percent. In contrast, when a country is in a banking crisis, the likelihood of a large reform falls from 5.5 percent to 2.6 percent, and the incidence of reversals (big or small) increases from 2.3 percent to 9.5 percent. Recessions and high inflation increase the likelihood of both reforms and reversals.
Besides crises, external influences such as changes in the global economy—such as increases or declines in global interest rates—can influence the balance of decision-making power and upset the status quo. When global interest rates are low, for example, access to cheap international capital provides a strong incentive, strengthening reformers and increasing the likelihood of liberalizing reforms. In contrast, when rates are high, the incentive to reform declines, except in countries eager to signal their commitment to liberalization (in the hope of attracting future investment) by maintaining their openness even in the face of high world interest rates.
International financial institutions such as the IMF may also be able to induce reform (see Table 3). Their actions can strengthen the hand of domestic outsiders—that is, those who do not have much of a voice in the political process—who support reform. Such outsiders include groups without access to credit and savers who receive low returns on their financial assets. To their benefit, an external impetus to reforms may originate as a condition imposed by an international financial institution on its loans or may come from external advisors from multilateral institutions, universities, and think tanks, as well as nationals previously associated with such organizations. These outside forces played a significant role in several reform episodes in our sample.
Political economy theory highlights the importance of the uncertainty faced by individuals regarding the outcome of reform. If individuals or interest groups do not know who among them will benefit, they may oppose change even if it is socially optimal and a majority stands to gain. Even if some of the existing financial institutions may prosper after liberalization, uncertainty regarding the identities of the winners and losers may cause the financial sector as a whole to oppose reform.
But what are the mechanisms for discovery, and why is the status quo eventually abandoned and reforms undertaken? Here, learning made possible by the accumulation of new information is particularly relevant. If reform is a multistage process, then early reform may help interest groups assess whether they will benefit or lose. Early opponents of reforms thus sometimes become the strongest advocates for further reforms.
Following an initially successful program, strong political opposition to reform is often neutralized. If additional policy measures (and good luck) then provide additional economic gains, support for the reforms may strengthen and opposition weaken. New interests emerge favoring the altered economic policies.
Learning may also occur—and reforms may cascade—when technical or managerial expertise in undertaking reforms is lacking initially but gradually builds. Under these conditions, implementation of multistage reform programs may need to be protracted. As expertise is gained from initial reforms, further reforms are easier to carry out.
Recent investigations of the political economy of reform—and of the spread of democracy—have noted a clustering of reforms in certain regions during the same period, which suggests that countries may learn from their peers—neighboring nations with similar economic, social, and political structures and shared experiences. There is also considerable evidence that countries within a region compete for the same international pool of risk capital, which may provide an impetus for reform.
We assessed the relevance of the learning process by determining if the distribution of policy changes varied with the level of financial liberalization in a country. One might surmise a negative relationship, since countries with highly repressed financial sectors have the most potential for liberalizing, while liberalized economies have less to do, but this is not the case. Rather, countries with highly repressed financial systems tend to stay that way, as evidenced by a much higher proportion of status quo observations. But if these countries liberalize somewhat, the likelihood of further reforms increases substantially. Reforms are much more probable in countries whose financial sectors are in an intermediate range of liberalization—either partially repressed or largely liberalized. Of course, countries with highly liberalized financial sectors are less likely to undertake further reforms. It is interesting to note, however, that none of the countries that are fully liberalized reverse their reforms.
The relationship between the level of financial liberalization and the incidence of reform supports the idea that learning creates a self-sustaining dynamic in the reform process. Reforms do not have to be undertaken all at once—even small reforms are a victory, as they carry the seeds of future reforms.
Finally, some economists believe that reforms can be conditioned by a governments ideology, institutions, or economic and political structure. This category of trigger includes the political orientation of ruling parties, forms of government (presidential versus parliamentary, for example), electoral rules, legal systems, and a countrys openness to international capital and trade.
With respect to ideology, the conventional view is that right-wing governments are more receptive to market-oriented reforms such as financial liberalization. However, we found the opposite to be true—left-wing governments liberalized more frequently than right-wing governments. One possible explanation is that party ideology and credibility interact in interesting ways. If the electorate cannot tell whether a governments policies are being motivated by social concerns or simply by the politicians ideological preferences, then policy reforms become more credible when undertaken by political parties normally associated with ideologies opposed to those reforms. For example, a reform-minded Labor Party government coming to power in New Zealand during a currency crisis in 1984 was able to undertake a program of rapid economic liberalization.
Some economists have argued that presidential forms of government can overcome the logjam arising from conflicting interests and push through reforms more easily than parliamentary governments. Others have emphasized the importance of legal systems, while yet others have suggested that, in economies that are more open to international investment and trade, insider opposition to liberalization is weaker, since there is more to gain from new opportunities.
While ideology and structure can have a long-term influence on the direction of reform, the question remains whether these overarching forces changed the observed dynamics of financial sector reform in the last quarter of the twentieth century. Was the period long enough? Or was there a common ideology behind the apparent differences that, together with the forces of globalization, pushed policymakers to undertake reforms?
We found that no one form of government was likelier than any other to initiate reform. Nor did we find a significant relationship between legal systems and financial reform. Looking at the sample as a whole, we did not find a significant relationship between liberalization and openness to trade and foreign investment, but, breaking the sample down, we found that openness was strongly associated with liberalization in highly repressed countries but less so in more liberalized countries. Trade openness was also associated with a greater tendency to privatize banks.
Liberalization thus occurred as a combination of discrete changes and gradual learning. Since shocks resulted in setbacks to liberalization almost as often as they resulted in reform measures, learning—whether from successful reforms on the domestic front or from observation of neighboring countries—was essential to the dynamic that sustained widespread reform. Five conclusions about the dynamic of reform emerged from our statistical analysis:
1. The countries with the most repressed financial sectors were the most likely to stay that way. However, countries whose financial sector became even slightly less restrictive after some initial reforms were far likelier to undertake further reforms.
The self-sustaining nature of reforms can be explained in several ways. Initial reforms tended to strengthen those who benefited from (and lobbied for) them relative to those who opposed them. Initial reforms may also have caused changes that necessitated further reforms. An interesting example is Japan, which received its initial reform impetus in the 1970s from the need to finance large fiscal deficits. This led to the development of the government bond market, which, in turn, created demand for a more open corporate bond market. As the corporate bond market grew, commercial banks prevented from participating in it experienced substantial revenue loss. Ultimately, the government had to liberalize the scope of banks activities—an example of reforms in one area triggering further reforms in other areas.
2. Regional influences were important. Countries in a given region faced greater pressure to liberalize their systems the more repressive their own financial systems were relative to the regions leader. In part, this was because countries in the same region possessed similar characteristics and were likely to be motivated by similar objectives and to be competing for the same pool of international capital. The effect of competition sharpened over time as the level of liberalization increased.
3. Crises that posed a serious challenge to the status quo triggered action. Pressure to do something built rapidly when things went wrong. However, different types of crises had strikingly different effects. Balance of payments crises, for example, increased the likelihood of reform, while banking crises discouraged—or even reversed—it. The nationalization of banks during or following a banking crisis often dealt a setback to reform. In addition, since banking crises made the fragility of the banking sector evident, it was hardly surprising that governments did not push hard for reforms that could have further undermined banks franchise values in the short run. Instead, governments attempting to resolve banking crises by encouraging strong banks to take over weak banks were apt to offer incentives such as temporary monopoly power.
4. Reforms seemed to follow certain kinds of events. For example, a new government was most likely to undertake reforms in the first year after it assumed power, and this effect was strongest in countries with repressed financial sectors. Reforms also accelerated following a decline in U.S. interest rates, while a rise in rates set them back. And the conditions associated with IMF-financed structural adjustment programs appear to have had a strong influence initially in countries with relatively repressed financial systems, but this influence decreased over time.
5. Among factors such as a countrys political ideology and economic structure, only openness to international investment and trade affected the pace of reform, stimulating reform in countries with relatively repressive financial systems. Right-wing governments were no more reform-oriented than left-wing governments. If anything, left-wing governments were a bit more likely to introduce reforms, although the difference was not significant. Static factors such as the type of government or legal system were also not influential.