Economic Institutions, Democracy, and Development

Source: Carnegie

Paper presented at the Conference on Democracy, Market Economy, and Development, sponsored by the World Bank and the government of the Republic of Korea, February 26-27, 1999

The complex relationship between forms of government, economic development, and underlying political and cultural values has been the subject of scholarly research for more than a hundred years. In more recent years, questions related to this triangular relationship have attracted the attention of policy-makers as there is a growing awareness that problems of governance, public policy, and economic performance are interconnected: solution to one set of problems may often be found in the solution to another set. A promising development as a result of such awareness is that multilateral institutions, Western aid-providers, and governments in many countries have placed the promotion of good governance on their policy agenda.

In this study I will attempt to address a set of issues that are at the heart of the puzzle: how do democracy and development influence each other and what policy and institutional reforms may have the most immediate and measurable effects on improving governance. In the first section of the paper, I will briefly summarize the theoretical literature on the relationship between democracy and development and discuss the claims made in the literature. I will then analyze recent data on the correlation between democracy, wealth, economic freedom, and corruption to explore the relationship between these variables. The third section applies some of the insights from our findings to the Chinese development experience since 1979. The final section will offer a modified, and less ambitious, approach to solving the dilemma between democracy and development.

I.  Democracy and Development

Numerous thinkers and scholars have written on this relationship. This brief survey will not include the vast literature on the relationship between markets and democracy although, unavoidably, similar issues may be addressed in the context of discussing democracy and development. Instead, this survey will narrowly focus on how political systems and economic development affect each other.

a. Development and Democracy: Does Wealth Lead to Democracy?

The impact of economic development on the transformation of political systems has long been thought to be direct and positive. Sustained economic development is supposed to lead to the emergence of democratic institutions and, eventually, democracy through a combination of factors produced by such development. First and most important, economic development will transform social structure and create a large enough middle class as the social basis of democracy. Second, economic development may, as its by-product, lead to the emergence of new political values (such as enhanced sense of individuality, personal autonomy, and value of personal freedom and choice) that support democratic institutions and practices. Third, a direct effect of economic development is the increase in the level of education. An educated citizenry is likely to be more knowledgeable about the political process and aware of their rights. Such a citizenry is more vigilant in defending its rights and possesses more effective means of doing so. Fourth, successful development will generate more economic wealth, which allows private-sector actors to accumulate resources and enhance their independence from the state, thus strengthening civil society as a counterweight to the state. Another beneficial effect of wealth is the increased possibility of resolving redistributional conflicts (because the bigger pie makes it more likely that every one will get a piece). Fifth, as successful development is more likely to occur in an open economy, such development may, in the process, promote extensive social, cultural, and political linkages with the international community. These linkages act to facilitate the flow of information (which undermines authoritarian rule) and constrain (through various external pressures) autocratic rulers.

Until recently, few have questioned these assumptions. Empirically, the case for a positive relationship between development and democracy appears unusually strong ? wealthy countries are more democratic than less wealthy countries (more discussion on this point later). But the data in Table 1 also show that wealth and democracy are closely correlated and may have a far more complex relationship than what the simple linear assumptions suggest about the effects of development on democracy, and vice versa. Academic research in the last few years have begun to shed some light on how economic development may affect the emergence and durability of democracy. In a path-breaking statistical study of the relationship between wealth and democracy, Adam Przeworski and Fernando Limongi disputed the notion that rising economic wealth leads to democracy. In a more nuanced argument, they find that wealth has a measurable effect on the survival rate of democracy, but not on the emergence rate of democracy. In other words, to the extent wealth is an independent variable, poor democracies are more likely to collapse than wealthy democracies. More specifically, Przeworski and Limongi?s analysis shows that rising wealth did not increase democracy. In fact, certain wealthy autocracies remain autocratic despite rising wealth. But the effect of wealth above a certain level on the durability of democracy is powerful and unambiguous: the probability of a democracy?s survival rises with wealth measured in per capita income. When per capita income is $1,000 (measured in purchasing power parity, or PPP), a democracy?s life expectancy is 8 years. When per capita income is between $2,001 and $3,000, the life expectancy of a democracy rises to 26 years. A democracy acquires immortality after per capita income is above $6000. Citing their statistical evidence that shows persistent autocracies above given wealth thresholds, Przeworski and Limongi reject the notion of linearity between wealth and democracy.

However, research by other political scientists shows that wealth produces a measurable, though modest, effect on democracy. A statistical analysis performed by John Londregan and Keith Poole reports that doubling per capita income has an especially large effect on moderately authoritarian countries. Typically, a doubling of per capita income would increase the degree of democracy in a moderately authoritarian country by 30 percent (compared to the status quo before the doubling). The effects of rising wealth on extremely authoritarian countries are small ? doubling per capita income leads to about 5 to 15 percent increase in the level of democracy in the most repressive regimes. The law of diminishing returns also applies here ? doubling capita income only marginally increases the degree of democracy in relatively open societies.

The two studies cited here may differ in their conclusions about the exact effects of economic development on democracy (i.e., whether it promotes or sustains democracy). The difference may be important to scholars, but should not obscure the central fact that development is beneficial to democracy.

b. Democracy and Development: Do Political Regimes Matter?

The relationship between democracy and development is far more contentious. The question whether political regimes affect economic growth was raised many years and remains unanswered. The idea that autocratic regimes have an advantage in economic development, although no longer taken seriously, was once quite fashionable. The advantages of autocratic regimes, to be sure, were not intrinsically derived. Rather, these regimes were supposed to have an edge in development mainly because they were said to lack the same disadvantages often associated with democracy. More specifically, these disadvantages include (1) insecure property rights of the wealthy (as a result of the enfranchisement of the poor, who are expected to use their voting power to redistribute wealth), (2) high propensity to consume (as a result of electoral politics and meeting voters? short-term demands), (3) rent-seeking by special interest groups that penetrate the open political process and use their influence to produce socially inefficient policies. As a theory, autocracy-good-for-development was thus extremely weak. While open political processes under democracy may lead to the above-described problems, there is nothing intrinsic to autocracy that would convince us that the same problems would not exist under autocratic rule. Property rights are by no means secure in an autocracy whose rulers are immune from institutional constraints (such as an independent judiciary or parliamentary competition) and have the capacity to impose confiscatory rates of taxation and refuse to honor their financial obligations. Similarly, autocrats have, historically, been known to plunder their societies for personal gains (such predatory behavior is partly motivated by the very insecurity of the autocrats themselves; such insecurity greatly reduces the time-horizon of autocrats). The vice of rent-seeking exists in autocracies, as well. Even the most absolutist rulers must rely on supporters to keep themselves in power. These rulers must constantly reward their constituents with favors to maintain their support. Such favors take the form of monopolies, subsidies, tax privileges, and licenses that greatly reduce economic efficiency.

The case that democracy promotes development rests on the central idea that the political institutions critical to economic development are more likely to exist and function effectively under democratic rule. These institutions include the rule of law which protects property rights, individual liberties which foster creativity and entrepreneurship, the freedom of expression which ensures the production and unimpeded flow of information, and institutional checks and balances that prevent massive theft of public wealth often observed in autocracies. One statistical study, of growth data for 115 countries from 1960 to 1980 claims that countries with high degrees of political openness achieved an average annual real per capita growth rate of 2.53 percent, compared with 1.41 percent in more closed political systems. This implies that more democratic countries may grow 80 percent faster than less democratic countries. Other well-known statistical studies, however, yield more ambiguous conclusions. One study examining GNP growth data from 94 countries for 1960-79 (excluding centrally planned economies, colonies, and capital surplus oil-exporting countries) reported that democracy has a weak negative overall effect on economic growth. Another statistical analysis of growth data for 100 countries from 1960 to 1990 reached a similar conclusion. But Barro?s study also suggests that the relationship between democracy and growth may be nonlinear and take an inverted u-form. According to his analysis, economic growth is likely to be slowest in the most politically repressed societies. But improvement in political rights and civil liberties in such societies tend to produce much higher growth. His data show that growth rates tend to peak when the level of democracy is in the middle-range and gradually taper off as the level of democracy rises.

These theoretical and empirical explorations may have advanced our understanding of the relationship between democracy and development, but they have left several important puzzles unresolved. The most important is the relationship between the rule of law and political regimes. This issue is central to answering the puzzle because economic historians have persuasively demonstrated that secure property rights constitute the institutional foundations of sustained economic development. The rule of law ? generally understood to mean the supremacy of legal norms and codes enforced by an independent judiciary ? has proved to be the most important institution that can protect property rights. The relationship between the rule of law and property rights is now so uncontroversial that the two concepts have almost become interchangeable when used in the context of economic development. But the relationship between the rule of law and political regimes is far more complex. Intuitively, democracy in general, and democratic institutions such as multi-party systems, competitive elections, and a free press in particular, should be viewed as part of the political foundations of the rule of law because the functioning of these institutions ensure that the rule of law will have its defenders (parties, candidates, and the media). More importantly, the competitive nature of democracy ensures that no single individual or political force will acquire so much political power as to overwhelm all other forces, which means that no individual or entity will be above the law or threaten the rule of law. Conversely, the rule of law has been thought as such an inseparable part of democracy that it is difficult to imagine a democracy without the rule of law.

In reality, the political foundations of the rule of law may be more complex and not well-understood. Democratic institutions may strengthen or defend the rule of law; but the same institutions also have the capacity to undermine the rule of law. Electoral politics may be a double-edged sword that can also undermine the rule of law in several ways. It may produce a majority that make new laws that deprive the minorities or unrepresented groups their rights (just witness the draconian laws passed recently in the U.S. against illegal immigration and convicted criminals). Democratically elected leaders can also rigger the judicial process through appointments; ultimately, institutions defending the autonomy of the judiciary are but the products of political decisions. Voters? preferences may change and such a change can lead to a conscious decision, even in the form of constitutional amendments, to reduce the autonomy of the judiciary. Recent American history is filled with episodes of attempts to revise the Constitution to pass laws that were declared unconstitutional by the Supreme Court (such as the law banning the burning of the American flag). That most such attempts have failed is an indication of the resilience of the American democracy, but also of the potential of democracy to weaken the rule of law. The existence of this potential ? a coalition?s ability to build a democratically formed majority to change the law ? makes property rights insecure. Admittedly, a complete redistribution of wealth under democracy has not taken place; this does not prove that such a possibility is zero. Rather, this fact may say something about the type of democracy we have (institutions that are embedded to protect property rights, not to promote mass sovereignty).

Secondly, it is unclear whether democracy is a pre-condition to the rule of law (while it seems indisputable that the rule of law must be a pre-condition to democracy). As historical evidence shows, a quite large number of non-democratic regimes (such as Kaiser?s Germany, pre-1945 Japan, Pinochet?s Chile, Hong Kong under colonial rule, Singapore, Franco?s Spain, and nearly all Western European countries before they became democratic in the mid-1800s) also had the capacity to maintain the rule of law. The claim that only popular sovereignty can guarantee the rule of law is thus questionable on empirical ground (as well as on theoretical ground). Of course, the political factors that help defend the rule of law may be different in democracies than in autocracies. But one central similarity is that the rule of law can maintain its effectiveness only when the rulers, whether democratically elected or not, are placed under certain political constraints (which need not be formal or constitutional). Of the autocracies (most of them in Western Europe) that were able to maintain the rule of law, nearly all of them were subject to powerful political constraints that defended property rights. Such constraints included the power of an independent aristocracy, the church, the rising urban capitalist class, and the ever-present external threat. These constraints compelled desirable political moderation by the rulers. In fact, it may be more accurate to say that it is political and institutional pluralism, rather than formal democracy, that has proved to be the ultimate defender of property rights. This observation may lead to a conclusion that is not wholly politically correct according to the post-Cold War doctrine of democratic enlargement. The first priority of the international community is not to promote formal democracy, but to promote political and institutional pluralism as an intermediate step.

II.  Democracy, Development, Economic Freedom and Corruption: Some Recent Evidence

To understand why, although democracies as a whole may have many embedded institutional advantages in economic development, such advantages are not apparent or real for poor countries, we perform a simple analysis of recent data on per capita income, political rights and civil liberties, economic freedom, and perceived corruption in 83-159 countries. (See Appendix 1 for detailed regression results) In this exercise, we assume a direct relationship between good governance and sustained economic development. Since it is difficult to establish measures of governance, we use a substitute, the corruption perceptions index compiled by an NGO, on the assumption that less corrupt countries have better governance. We report the following findings:

Wealth, Regimes, and Economic Freedom

The data in Table 1, which contains information on per capita GDP income (measured in $PPP), combined measures of political rights and civil liberties, and scores of economic freedom for 159 countries, show a close relationship between wealth and democracy (measured here by the combined scores of political rights and civil liberties). Regression analysis (equation No. 4 in Appendix 1) demonstrates a statistically significant and positive effect per capita income has on democracy. Each increase of $PPP1,000 raises the measured degree of democracy by 0.287.

Regression results also indicate that wealth has a small but positive effect on economic freedom. Each increase in $PPP1,000 would lead to a small increase (0.0775) in the economic freedom index. Political regimes, in comparison, have a bigger impact on economic freedom. Regression results show that an increase of 1 in the index of autocracy (for example, from 10 to 11 in the Freedom House index) would result in a reduction of economic freedom by 0.125. This implies that, everything else being equal, the effect of increasing political rights and civil liberties by 1 (on a scale of 2 to 14) would produce almost twice as large an effect on increasing economic freedom.

Wealth, Regimes, Economic Freedom, and Governance

To understand the complex relationship between per capita income, political regimes, economic freedom, and corruption perceptions (a proxy for governance), we perform a multivariate regression on data from 83 countries (Equation No. 1). The statistically significant results show that wealth, democracy, and economic freedom, as expected, all contribute to good governance (as reflected in the perceived corruption index). But of the three independent variables, economic freedom has the most powerful effect on corruption. After adjustment (because the range in the economic freedom index is 1.25 to 5.0 and that in the freedom index is 2 to 14), a unit increase in economic freedom has about three times the effect on governance as a unit increase in political freedom. By comparison, increase in wealth has the smallest impact. A rise of per capita GDP of $PPP1,000 leads to an increase of 0.187 on the corruption perceptions index (a rise on the index indicates less corruption).

These statistical findings reconfirm our intuitive understanding of the relationship of wealth, economic freedom, political regimes, and governance. In countries with greater economic freedom, fewer opportunities for rent-seeking exist; there should be less corruption in these countries. Of course, the relationship can also work in reverse: more corrupt countries have less economic freedom because groups or individuals benefiting from corruption would use their influence to limit economic freedom.

The finding that political regimes have a positive (although modest) effect on governance is not surprising, either, since in our earlier discussion on regimes and economic development, there is a stronger case to be made for sustained economic development under democracy than under autocracy. But one should also be cautious in reading too much into this finding. In fact, many democratic countries are known to have relatively poor governance. How do we explain this puzzle?

In our sample of 83 countries for which the data on corruption perceptions were available, we have 63 countries with median and above median scores of political rights and civil liberties. (The freest countries score 2 while the least free countries score 14, with the median score being 8). Twenty countries in the sample have scores of 9-14 and are considered "not free." We assume that freer countries should be less corrupt and thus regard as an anomaly a country with a median freedom score but an above-median score of corruption perception (more corrupt). Of our 63 countries with median freedom scores and better, 40 (or 63 percent) are found to have above-median (hence worse than median) scores on the corruption perceptions index (in other words, countries ranked at 27^th and above). A closer examination of these "anomalous" countries shows that nearly all of them are relatively new democracies or semi-authoritarian regimes in transition whose scores of political rights and civil liberties suggest that they are not consolidated democracies. Moreover, their income is low compared to established democracies (most with per capita GDP income under PPP$10,000). Of the 37 countries with above-median freedom scores but with per capita GDP income below PPP$10,000, only three countries have better-than-median corruption perceptions scores (considered less corrupt). This shows that (1) although democracies may not be able to eliminate corruption, more established and consolidated democracies have less corruption than new democracies; (2) the likelihood that a new democracy or a transitioning semi-authoritarian regime with a per capita GDP income of less than PPP$10,000 has a worse-than-median corruption score is quite high (almost 92 percent).

Autocracies do not fare better in this area. In fact, our analysis shows that they may do considerably worse. (Of course, the small sample size raises legitimate questions about the validity of this claim). The first sign that autocracies may be more corrupt (as measured by the simple method used here) is that, of the 20 autocracies (with freedom scores of 9-14) in the sample, 18 of them (or 90 percent) have worse-than-median corruption scores (ranked higher than 27^th ). Two have better-than-median corruption scores. Thus, controlling for income, autocracies may have a greater likelihood of being more corrupt (90 percent vs. 63 percent) than democracies. For autocracies with per capita income of less than PPP$10,000, 18 out of 18 (100 percent) have worse-than-median corruption scores. This figure is slightly higher than that for countries with better-than-median freedom scores (92 percent).

The findings of our statistical analysis echo the widely held view that institutions play an important role in maintaining good governance. As far as governance, as measured by corruption, is concerned, countries with more political rights and civil liberties have a slight edge over those with fewer political rights and civil liberties. But the advantage enjoyed by economically more free countries over economically less free countries is much greater. This conclusion supports the view that, to the extent that political and economic institutions influence economic development, the top priority of the international community is not to promote democratic reforms immediately (although such reforms are beneficial), but to promote reforms in key economic institutions that have a direct impact on economic freedom. In reality, of course, democratic political reforms and economic institutional reforms may be closely interrelated. In some cases, economic institutional reforms cannot proceed without important democratic reforms that dislodge rent-seeking groups from power. But in thinking about promoting democracy as a means of economic development, one must be cautious in both raising expectations and devising policies. Our finding that new and unconsolidated democracies tend to have above-median corruption scores should warn us that the positive effects of democratic reforms may not be immediate. In fact, it usually takes a relatively long period of time for democratic reforms to yield benefits in the area of economic reform (although the improvement in the area of political rights and civil liberties is more immediate and visible).

III.  Governance and Development: Will the Chinese Miracle Turn into Mirage?

There is a widely shared belief that China?s economic development in the last two decades has been extremely successful ? the country achieved sustained high growth rates, quadrupled its GDP per capita income, raised millions of people out of poverty, increased efficiency, and established broad linkages with the world economy. There is much less consensus, however, on whether governance in China has improved correspondingly. There is still less agreement on whether China?s impressive economic gains could be sustained without important reforms in its political institutions and improvement in its governance.

On the surface, it may be tempting to cite the East Asian Miracle in general, and China?s developmental success in the last two decades in particular, as proof that governance-enhancing institutional reforms are not necessary for sustained economic development. China has obtained its remarkable gains, it is said, without undertaking many of the critical reforms considered as the foundations for long-term growth (such as the rule of law, financial regulatory regimes, transparency, and a system of accountability). This reading of the East Asian Miracle and the Chinese reform experience misses several important lessons. In the Chinese case, without denying the real rapid gains in economic development China has achieved (although the precise extent of those gains remain disputable due to reliability of data), one may argue that, far from being unique, China?s experience proves the crucial link between good governance and sustainable development. Specifically, China?s rapid growth in the last two decades benefited from many one-off factors that can no longer be counted on to sustain the country?s development momentum. China, for example, enjoyed relative political stability since 1979 (compared to the preceding 140 years of continuous foreign invasion, national disintegration, civil war, and domestic turmoil). The most important institutional reform ? the dismantling of the commune system in rural areas ? produced huge one-off efficiency gains and freed up rural labor markets. China?s gradual economic liberalization, however limited, was a positive contributing factor. Unlike many transition economies in the former Soviet bloc, the Chinese economy had better economic fundamentals (absence of a monetary overhang, high savings rates, a young and relatively well-educated and healthy labor force, a relatively small socialist state, and a flexible central-local political structure conducive to local initiatives). The combination of enlightened government policy and lure of the huge potential market also attracted enormous foreign direct investment, mostly from the Chinese diaspora. However, the benefits produced by most of these factors have been wearing off in recent years while more deeply embedded problems are surfacing. If China?s late leader Deng Xiaoping called his economic reform the "second revolution," his successor, Jiang Zemin, now faces the prospect of leading the "third revolution" in order to sustain reform. Unlike Deng?s reforms which focused on economic liberalization, Jiang?s reform must focus on building the institutional foundations of development.

This third revolution is absolutely essential for several reasons. First, China?s past success is no guarantee for its future success. Second, China?s past success was built on relatively low base ? when the overall conditions were so poor, a small improvement can produce huge one-off gains. Third, in addition to delaying tackling its most difficult economic problems (especially the crisis of the state-owned enterprises), the Chinese government has only make minor progress in improving the institutions of governance. For instance, China?s legal system remains rudimentary and poorly equipped in enforcing contracts and protecting property rights. China?s unitary state institutions are seriously at odds with its federalist political reality, creating serious frictions between the central government and local governments. China?s civil society remains weak and heavily controlled by the state, resulting in decreasing social cohesion and rising social anomie (or normless behavior). As the government is fearful of the political consequences of democratization, there are few open participatory channels (such as the electoral and legislative processes) through which popular discontent and demands may be expressed. This directly causes social unrest to be both more frequent and take a more violent form.

All of the above institutional weaknesses will likely cause governance to deteriorate. This outcome will inevitably jeopardize China?s economic performance as a result of rising political instability (due to public frustration) and increasing economic costs of poor governance (corruption, prohibitive transaction costs due to a weak legal system, and opportunistic behavior by local governments). Conservative estimate put the direct economic costs of official corruption in China at around $10 billion or about 1 percent of GDP in 1998. The indirect costs could be much higher, as demonstrated in the case of the government?s financial losses in its grain procurement fund. Through a combination of interventionist policies, direct theft, and misappropriation of funds, government agents and grain brokers contributed to the loss of more than $25 billion of public funds budgeted for the purchase of grains produced by farmers in a six-year period. Government auditors showed that $10 billion (or 40 percent) of the funds lost were either stolen or diverted to sideline businesses operated by government officials in charge of the funds.

The dilemma facing the Chinese government today is that it is under tremendous external pressure and gradually rising internal demand to liberalize its political system while elite resistance to such democratizing reform remains strong. Moreover, China?s current governance institutions ? such as the legal system, quasi-federalist institutions, and civic organizations ? are perhaps too underdeveloped to withstand the political shocks of full democratization. The last thing the Chinese leaders want is a premature political opening resulting in a spectacular Soviet-style collapse.

But refusal to reform for fear of instability will only delay, but not remove, the day of reckoning. If anything, delay will further complicate the problem and increase the eventual costs of solving the same problems. If social science research is any guide for policy, Chinese leaders may be well advised to expend their limited political capital on improving governance institutions that have a direct impact on economic efficiency while putting off the most dramatic democratizing reforms. Recent events in China, such as the reform of the central bank and the fiscal system, suggest that this may be the route the country?s leaders have chosen.

IV.  Conclusion: Putting Economic Institutions First

The findings of this paper have several implications for policy. First and foremost, if increase in wealth has a relatively modest effect on improving governance, it would be naïve to count on accelerating economic development alone as a means of achieving good governance (obviously, such a strategy would surely fail because sustained development is surely impossible without good governance). Secondly, it is also naïve to believe that democratization alone will improve governance immediately. To be sure, countries scoring higher on the index of political rights and civil liberties have a slight edge in governance. But most young democracies have relatively poor governance records. The most effective means of obtaining good governance is to develop institutions governing economic activities and increasing economic freedom.

Economic historians such as North have made a strong case that institutions are vital to the long-term economic development of any society. Recent empirical works of the East Asian development experience have reached similar conclusions. However, an appreciation of the role of institutions in economic development is not enough. There is little doubt that such appreciation, differing only in the degree of sophistication, exists today at the elite level in most countries. The most serious challenge is political ? how to build economic institutions in an inhospitable political environment characteristic of many undemocratic countries and newly democratized countries.

As demonstrated by our theoretical discussion of why the rule of law may not function fully under democracy and by our empirical example showing the coexistence of democracy and serious corruption in many developing countries, democratization is not a panacea for development (despite its long-term promises). Therefore, a two-pronged approach must be considered.

First, in countries where democratization has already occurred, the top priority must be given to the establishment and consolidation of those institutions that have the most immediate, direct, and powerful impact on macroeconomic stability, security of property rights, and free trade. These steps necessarily involve the strengthening of the independence of the central bank, the regulatory regime of the financial system, the courts, and competition regimes. This institutional reform process may encounter political difficulties created by the new democratic institutions and processes. There is no conclusive evidence on whether the democratization process itself will aid or hinder these steps.

Second, in countries where democratization has yet to occur, the emphasis of change should be placed more on the establishment and strengthening of the same economic institutions than on the direct promotion of democracy (or more crudely, elections). The strategy of direct democratic promotion, however desirable, faces enormous odds (strong resistance from the ruling elite and weak socioeconomic and institutional foundations in those societies). The success of direct democratic promotion is highly uncertain; the beneficial effects of democratization on economic development, even if one assumes the success of democratic promotion, may not be immediately substantial. On the other hand, a strategy emphasizing institution-building without regime change may be more practical and yield more immediate results. Such a strategy aims at finding allies within the ruling autocratic regime who (for complex reasons) have a long-term interest in the development of these institutions. Moreover, this step poses less threat to the regime than the democratic promotion strategy, and is likely to encounter less resistance. Considering that reformers (whether domestic or international) have limited political capital, this strategy may be the most cost-effective since it makes the building of a reform coalition (with critical participation from elite members inside the regime) more achievable. The long-term benefits of this indirect approach ? building economic institutions ahead of democratic institutions ? may be twofold. Economically, such institutions will undoubtedly contribute to sustained long-term growth. Politically, these institutions will not only promote the eventual development of democracy (through sustained growth) but help insulate future democratic institutions and processes from the temptation of rent-seeking (since the rents available will be greatly diminished as a result of the operation of strong economic institutions). A final advantage of having strong economic institutions before democratization is that the existence and operation of these institutions will likely increase the likelihood of democratic survival and consolidation during the post-transition phase. As the statistical analysis performed by some political scientists show, poor democracies tend to die of economic crisis. Such crises may not have the same devastating impact if those countries have strong economic institutions capable of containing them.

There are, however, several serious political risks for advocating and implementing this strategy. Despite the convergence of the long-term objectives of this economic-institutions-first (EIF) strategy and the democracy promotion strategy, the former may attract less political support inside leading industrial democracies because the pay-off of the strategy is long-term and uncertain while pressures for short-term results are considerable. The EIF strategy is especially vulnerable to criticisms from human rights groups if it is applied in authoritarian countries with unacceptable rights records. Finally, the EIF strategy does not guarantee success. Since the strategy is predicated upon finding elite-level allies inside the ruling hierarchy who share a common interest in building these institutions, its fate is closely tied with that of the indigenous reformers ? a highly unpredictable factor over which external actors have little control. The challenge for the international development community in pursuing the EFI strategy is therefore primarily political ? building domestic political coalitions in support of such a strategy and finding political allies within recipient countries to implement the same strategy.

Table 1. Wealth, Democracy, and Economic Freedom

Notes:

1. Measures of Freedom consist of two separate indicators, political rights and civil liberties. The combined score of political rights and civil liberties is between 2 and 14, 2 being the most free and 14 being the most unfree. Freedom House considers countries with scores of between 2 and 5 "free;" those scoring between 6 and 10 as "partly free;" those scoring between 11 and 14 as "unfree." The median score is 8.

2. Measures of Economic Freedom consist of one index in which the freest economy (Hong Kong) scores 1.25 and the least free economy (North Korea) scores 5.0. The median therefore should be 3.125. Countries with scores near the median (such as Mexico, with a score of 3.15) are ranked 85^th.

Sources: United Nations, Human Development Report 1998; World Bank, World Development Report 1996; Freedom House, Freedom in the World 1997/98; Bryan Johnson et al., Index of Economic Freedom 1999.

Table 2: Wealth, Democracy, Economic Freedom, and Corruption

Sources: United Nations, Human Development Report 1998;World Bank, World Development Report 1996; Freedom House, Freedom in the World 1997/98; Bryan Johnson et al., Index of Economic Freedom 1999; Transparency International, "Corruption Perceptions Index 1998."

1. Measures of Freedom consist of two separate indicators, political rights and civil liberties. The combined score of political rights and civil liberties is between 2 and 14, 2 being the most free and 14 being the most unfree. Freedom House considers countries with scores of between 2 and 5 "free;" those scoring between 6 and 10 as "partly free;" those scoring between 11 and 14 as "unfree." The median score is 8.

2. Measures of Economic Freedom consist of one index in which the freest economy (Hong Kong) scores 1.25 and the least free economy (North Korea) scores 5.0. The median therefore should be 3.125. Countries with scores near the median (such as Mexico, with a score of 3.15) are ranked 85^th.

3. Ranking of corruption is based on the data provided by Transparency International (1998). The score for the least corrupt country is 10, the most corruption 1.4; the median score is 5.7. The country with the median score is Estonia and ranked 26^th.

Appendix: Regression Results

1. Wealth, Regimes, Economic Freedom, and Corruption: regress corrupti __ppp liberty ecofree

Source | SS df MS Number of obs = 83

---------+------------------------------ F( 3, 79) = 68.09

Model | 333.667451 3 111.222484 Prob > F = 0.0000

Residual | 129.045494 79 1.63348727 R-squared = 0.7211

---------+------------------------------ Adj R-squared = 0.7105

Total | 462.712945 82 5.6428408 Root MSE = 1.2781

------------------------------------------------------------------------------

corrupti | Coef. Std. Err. t P>|t| [95% Conf. Interval]

---------+--------------------------------------------------------------------

__ppp | .0001878 .0000286 6.566 0.000 .0001309 .0002447

liberty | -.0923626 .0530806 -1.740 0.086 -.1980169 .0132918

ecofree | -.6975913 .3573465 -1.952 0.054 -1.408872 .0136892

_cons | 5.647138 1.156452 4.883 0.000 3.345278 7.948999

------------------------------------------------------------------------------

2. Wealth and Economic Freedom: regress ecofree on ppp

Source | SS df MS Number of obs = 159

---------+------------------------------ F( 1, 157) = 95.44

Model | 46.7181288 1 46.7181288 Prob > F = 0.0000

Residual | 76.8498541 157 .489489517 R-squared = 0.3781

---------+------------------------------ Adj R-squared = 0.3741

Total | 123.567983 158 .782075841 Root MSE = .69964

------------------------------------------------------------------------------

ecofree | Coef. Std. Err. t P>|t| [95% Conf. Interval]

---------+--------------------------------------------------------------------

__ppp | -.0000775 7.94e-06 -9.769 0.000 -.0000932 -.0000619

_cons | 3.691814 .0768198 48.058 0.000 3.54008 3.843547

------------------------------------------------------------------------------

3. Regimes and Economic Freedom: regress ecofree liberty

Source | SS df MS Number of obs = 159

---------+------------------------------ F( 1, 157) = 75.50

Model | 40.1273795 1 40.1273795 Prob > F = 0.0000

Residual | 83.4406034 157 .531468811 R-squared = 0.3247

---------+------------------------------ Adj R-squared = 0.3204

Total | 123.567983 158 .782075841 Root MSE = .72902

------------------------------------------------------------------------------

ecofree | Coef. Std. Err. t P>|t| [95% Conf. Interval]

---------+--------------------------------------------------------------------

liberty | .125407 .0144325 8.689 0.000 .0969002 .1539138

_cons | 2.25706 .1202015 18.777 0.000 2.019639 2.49448

------------------------------------------------------------------------------

4. Wealth and Regimes: regress liberty __ppp

Source | SS df MS Number of obs = 159

---------+------------------------------ F( 1, 157) = 52.71

Model | 641.316932 1 641.316932 Prob > F = 0.0000

Residual | 1910.1925 157 12.1668312 R-squared = 0.2513

---------+------------------------------ Adj R-squared = 0.2466

Total | 2551.50943 158 16.1487939 Root MSE = 3.4881

------------------------------------------------------------------------------

liberty | Coef. Std. Err. t P>|t| [95% Conf. Interval]

---------+--------------------------------------------------------------------

__ppp | -.0002872 .0000396 -7.260 0.000 -.0003654 -.0002091

_cons | 9.224977 .3829926 24.087 0.000 8.468495 9.98146

------------------------------------------------------------------------------