China Is Not Ten Feet Tall
How Alarmism Undermines American Strategy
DISPELLING A MYTH
"Economic development makes democracy possible" asserts the U.S. State Department's Web site, subscribing to a highly influential argument: that poor countries must develop economically before they can democratize. But the historical data prove otherwise. Poor democracies have grown at least as fast as poor autocracies and have significantly outperformed the latter on most indicators of social well-being. They have also done much better at avoiding catastrophes. Dispelling the "development first, democracy later" argument is critical not only because it is wrong but also because it has led to atrocious policies-indeed, policies that have undermined international efforts to improve the lives of hundreds of millions of people in the developing world.
Those who believe that democracy can take hold only once a state has developed economically preach a go-slow approach to promoting democracy. But we and others who believe that countries often remain poor precisely because they retain autocratic political structures believe that a development-first strategy perpetuates a deadly cycle of poverty, conflict, and oppression.
Why has the development-first myth prevailed? First, it rests on a common-sense notion, put forward by political sociologist Seymour Martin Lipset and others some 45 years ago, that economic growth creates the necessary preconditions for democracy by expanding literacy, creating a secure middle class, and nurturing cosmopolitan attitudes. Second, it fits comfortably with the demands of the era of its origin, the Cold War, when about a third of countries qualified as democracies and very few of them were poor. Governance patterns appeared stuck, with countries trapped in opposing magnetic fields created by the Soviet bloc and the West. Pinning hopes for progress in the developing world on seemingly exceptional democratic examples such as India, Costa Rica, and Colombia appeared unrealistic under such conditions. Besides, the West was happy to bolster authoritarian governments that were not controlled by the Soviet Union to prevent them from turning communist.
The development-first thesis-which subscribes to the notion of an authoritarian advantage-has persisted in the post-Cold War world, despite the abysmal economic records of Latin American military governments, the "strongman" rulers in Africa, and the communist states in Eastern Europe and the former Soviet Union. This is largely because of the dazzling economic performance of certain eastern Asian autocracies: Singapore, Indonesia, South Korea, Taiwan, and, lately, China. Based on these countries' experiences, a variant of the development-first thesis has gained particularly wide appeal: strong, technocratic governance, insulated from the chaos of democratic politics, is the best way to pursue efficient and farsighted macroeconomic policies. According to this view, the experience of Russia in the 1990s and the faltering performance of young democracies in eastern Europe, Latin America, and Africa demonstrate the folly of attempting democracy too soon.
STRONGER, NICER, CALMER
As compelling as the development-first thesis sounds, the empirical evidence is clear: democracies consistently outperform autocracies in the developing world. But before proceeding, it is important to establish what we mean by democracy. Democracies are political systems characterized by popular participation, genuine competition for executive office, and institutional checks on power. We put this definition into practice using the Polity IV democracy index, devised by Ted Robert Gurr of the University of Maryland in 1990. The annual index gives each country a score between 0 (least democratic) and 10 (most democratic) based on the extent to which it exhibits the democratic characteristics listed above. To compare distinctive governance types, we categorize countries that score between 8 and 10 on this scale as democracies and those that score between 0 and 2 as autocracies.
Because everyone agrees that the most prosperous states in the world are well-established democracies, and because the real debate is over whether low-income democracies are capable of growing at a rate comparable to that of low-income authoritarian governments, this discussion is limited to countries with GDP per capita of under $2,000 (in constant 1995 dollar terms). We thus compare two groups of countries: low-income democracies and low-income autocracies.
The data, compiled from the World Bank's World Development Indicators from 1960 to the present, reveal a simple truth: low-income democracies have, on average, grown just as rapidly as low-income autocracies over the past 40 years. Outside of eastern Asia (about which more will be said later), the median per capita growth rates of poor democracies have been 50 percent higher than those of autocracies. Countries that have chosen the democratic path-such as the Dominican Republic, India, Latvia, Mozambique, Nicaragua, and Senegal-have typically outpaced their autocratic counterparts, such as Angola, the Republic of Congo, Syria, Uzbekistan, and Zimbabwe. Moreover, because 25 percent of the worst-performing authoritarian regimes, including Cuba, North Korea, and Somalia, have failed to document their performance, the growth shortfall for autocracies is even larger than the available data indicate.
The advantage poor democracies have over poor autocracies becomes even more apparent when the debate moves from growth rates to broader measures of well-being. Development can also be measured by social indicators such as life expectancy, access to clean drinking water, literacy rates, agricultural yields, and the quality of public-health services. On nearly all of these quality-of-life measures, low-income democracies dramatically outdo their autocratic counterparts.
People in low-income democracies live, on average, nine years longer than their counterparts in low-income autocracies, have a 40 percent greater chance of attending secondary school, and benefit from agricultural yields that are 25 percent higher. The latter figure is particularly relevant because some 70 percent of the people in poor countries live in the countryside. Higher levels of agricultural productivity mean more employment, capital, and food. Poor democracies also suffer 20 percent fewer infant deaths than poor autocracies. Development practitioners should pay particularly close attention to these figures because infant-mortality rates capture many features of social well-being, such as prenatal health care for women, nutrition, quality of drinking water, and girls' education.
Careful review of the data suggests that low-income democracies have another powerful advantage: they are better at avoiding calamities. Since 1960, poor autocracies have experienced severe economic contractions (falls of 10 percent or more in annual GDP) twice as often as poor democracies. Seventy percent of autocracies have experienced at least one such episode since 1980, whereas only 5 of the 80 worst examples of economic contraction over the last 40 years have occurred in democracies.
Viewed through this prism, many of the periods of rapid growth enjoyed by poor autocracies, which are frequently cited by development-first advocates, were little more than spurts to make up for ground lost during hard times. Consider Chile. Although often touted as a model of autocratic growth for its 13 years of economic expansion during Augusto Pinochet's 17-year rule, Chile also suffered two acute economic crises during this time: a 12 percent decline in GDP per capita in the mid-1970s and a 17 percent contraction in the early 1980s. It took until the mid-1980s for Chile to sustain a per capita income level higher than that of 1973, the year Pinochet seized power.
The frequent criticism that democracies pander to populist-driven interests to the overall detriment of the economy is demonstrably false. Poor democracies have, on average, not run higher deficits over the past 30 years than poor autocracies. Similarly, both poor democracies and poor autocracies spend almost the same on education and health. Democracies have just used their resources more effectively. Not coincidentally, low-income democracies typically score between 15 to 25 percent stronger on indices of corruption and rule of law than do autocracies.
Democracies also do a better job of avoiding humanitarian emergencies: the 87 largest refugee crises over the past 20 years originated in autocracies, and 80 percent of all internally displaced persons in 2003 were living under authoritarian regimes, even though such systems represented only a third of all states. The Nobel laureate and political economist Amartya Sen once famously observed that no democracy with a free press has ever experienced a major famine.
Some hold that "premature" democratization in low-income countries is responsible for enabling opportunistic politicians to fan ethnic and regional resentments, even armed conflict. According to this point of view, the iron fist of an autocratic leader can keep a fractious society intact. But this argument, too, fails to withstand empirical scrutiny. Poor countries fall into conflict often-about one year in every five since 1980. But poor democratizers fight less frequently than do poor authoritarian nations. In sub-Saharan Africa, where most civil conflict has occurred recently, countries undergoing democratic reform have experienced armed conflict half as often as the norm in the region.
Although the data show that poor democracies do a better job of generating material benefits for their citizens than poor authoritarian countries, there is, of course, variation in each category. Some democracies flounder; a few authoritarian regimes, especially in eastern Asia, have flourished. The latter cases show that development under authoritarian systems is possible. Yet this class of authoritarian governments is far from representative of most autocracies around the world. South Korea, Taiwan, Singapore, and Indonesia encouraged the private sector, pursued export-oriented growth strategies, and were heavily influenced by Western democracies when they adopted their particular economic and political institutions. Moreover, as China demonstrated by its appalling economic record through the late 1970s (when it began to adopt market-oriented economic policies), authoritarianism is not the distinguishing characteristic of its growth. (This is a point underscored by the disastrous economic performances of other eastern Asian autocracies: North Korea, Burma, Cambodia, and the Philippines under Ferdinand Marcos.) Thus, although exceptional cases exist, it is the preponderance of experience that should guide development policy. And the overall evidence is overwhelming: poor democracies have had a consistent development advantage over poor autocracies over the past 40 years.
The complementary assumption of the development-first argument is that democracy will eventually follow economic progress. Specifically, as a country emerges into middle-income status-Newsweek's Fareed Zakaria and others have touted per capita income levels of $6,000 as the target income threshold-the increasingly sophisticated population will inevitably call for greater political participation, leading to a successful democratic transition. Yet there is a serious practical problem with testing this assumption: few authoritarian countries have attained the middle-income category. Since 1960, only 16 autocratic countries have had per capita incomes above $2,000. Of these, only six-Taiwan, South Korea, Spain, Portugal, Greece, and, debatably, Mexico-adopted democracy in the aftermath of economic expansion. This provides scant basis to apply a development-first model to the entire developing world. For that matter, the $6,000 benchmark would suggest that all but 4 of the 87 countries currently undergoing a democratic transition, including Brazil, Kenya, the Philippines, Poland, and South Africa, are unfit for democracy.
Having highlighted the superior performance of poor democracies over poor authoritarian regimes, we turn to the conceptual underpinnings of this pattern. Poor democracies outperform authoritarian countries because their institutions enable power to be shared and because they encourage openness and adaptability.
Democratic leaders have incentives to respond to the needs of common citizens. Otherwise, they find themselves out of office. And because ordinary people care about bread-and-butter issues, these concerns figure prominently in candidates' agendas. By contrast, the narrow clan-and patronage-based support on which autocratic leaders often rely for power gives them little incentive to focus on the general well-being of society.
The developmental advantage of democracies also stems from the checks and balances that characterize self-governing political systems. Power is not monopolized by any one individual or branch of government, even though a national leader may claim a popular mandate. Although democracy is a more cumbersome process, it reduces the scope for rash, narrowly conceived, or radical policies that can have disastrous economic consequences. Federated systems also place checks and balances on the various levels of government, thereby guarding against an overconcentration of power at the national level while allowing for flexibility to address local priorities.
Authoritarian regimes, by comparison, often turn political monopoly into economic monopoly. Only businesses and individuals closely tied to the ruling party are able to acquire the licenses, permits, credit, and other resources needed to succeed. Such preferential treatment diminishes competition and innovation and therefore reduces economic efficiency. Consumers get fewer choices and higher prices. When political allegiances also dictate access to education, housing, career options, and social status, the spectrum of opportunities available to political outsiders is severely narrowed. An integral virtue of democracies, therefore, is that they provide a sphere of private space, which, protected by law, nurtures inventiveness, independent action, and civic activity.
Democracies are open: they spur the flow of information. Organizations in and out of government regularly report findings, educate the public, and push political leaders to consider a full range of options, spreading good ideas from one sector to another. The free flow of ideas, every bit as much as the flow of goods, fosters efficient, customized, and effective policies. Put this way, development is an exercise in educating a population: to wash hands, improve farming techniques, eat nutritious food, or protect the environment, for example. And societies that promote the free flow of information have a distinct advantage in these efforts.
Information is best communicated through multiple and independent channels. For example, it was the active public-education campaign undertaken by the Ugandan government and nongovernmental organizations in the 1990s that dramatically reduced the transmission of HIV/AIDS in that country. Uganda was once the world leader in percentage of adult population infected, at roughly 30 percent, but by 2003, that rate had declined to 7 percent. By contrast, attempts to suppress information during the SARS epidemic in China allowed the disease to spread before the public became aware and concerted action could be taken. Once the epidemic was acknowledged, distrust of the government led many Chinese in infected areas to violate the government's quarantine. This example also confirms a larger proposition: democracies do a better job of correcting errors. Once private or public authorities make decisions in open societies, the results become known and corrective action, if needed, can be taken.
Openness also reduces the scope for corruption. An independent, investigative media creates higher expectations regarding transparency and disclosure of potential conflicts of interest. Paradoxically, greater openness in newly democratizing societies may at first lead to the perception that corruption is worsening. In Kenya, for example, a survey by Transparency International found that the perception of corruption was worse in 2003-the first year of the democratically elected government-than in the late 1990s, under the authoritarian rule of Daniel Aap Moi. Yet the same organization found that Kenyans paid an average of nine bribes in 2003, down from 29 in 2002, saving roughly 10 percent of their annual income.
Transparency does more than cut the cost of bribes, which, technically, merely transfer money from one citizen to another and do not thereby reduce average incomes. The World Bank estimates that corruption, which acts as a tax on legal commerce and makes returns less certain, costs the global economy five percent of its total value, or $1.5 trillion a year.
Adaptability is another beneficial feature of democracies. Democracies enhance political stability by establishing clear mechanisms for succession. This allows them to adapt smoothly to the death or electoral defeat of a leader, minimizing the scope for extralegal or coercive tactics to attain power. Development momentum is thus sustained even though specific policies change from one administration to the next.
Adherence to established means for transferring power reflects a commitment to the rule of law under a democracy: leaders can gain legitimacy in the eyes of the people only if they ascend to power through democratic processes. Political legitimacy grounded in the rule of law, in turn, provides the foundation for the application of legal norms in the conduct of government and business, and a rules-based regulation of the economy.
Finally, democratic structures adjust well to changing circumstances. Because policies in democracies flow from an elaborate process of trial and error, they can adapt to realities on the ground. When there is a constant flow of policies and ideas, there is pressure to amend, drop, or replace initiatives that do not work. Elections are the most distinctive junctures around which these adjustments occur. But even during a given leader's tenure, constant fine-tuning takes place. Democracies are distinctive, therefore, not because they always identify the best policy but because they institutionalize the right to change leaders or policies when things go poorly. This capacity for revitalization explains why citizens of established nations such as Argentina, Guatemala, Kenya, and South Africa spoke of living in a "new" country after recent democratic changes in leadership.
All in all, then, democracies present an enormously powerful set of institutions that propel development. The more representative, transparent, and accountable those governmental processes, the more likely policies and practices will respond to the basic priorities of the general population.
With the case for supporting democracies so compelling, it may come as a surprise that the United States, other industrialized democracies, and international financial institutions have not shown greater preference to countries on the path to democracy when providing economic assistance. Instead, existing rules have typically prevented democratic criteria from guiding funding decisions. As a result, as much official development assistance (as a percentage of GDP) has been provided to autocracies as to democracies. This is not just a Cold War phenomenon; the same patterns have applied since 1990. Nor does it reflect disproportionate levels of humanitarian assistance to crisis-riven autocracies; the lack of distinction survives even if only non-emergency assistance is considered or the poorest countries are removed from the sample. Despite increased rhetoric and funding for democracy-promotion projects, the simple fact is that the West does not tilt its development assistance to democracies. This can and should change.
The U.S. government and the multilateral financial institutions should adopt five new policies to prioritize democracies. First, a principle of "democratic selectivity" should be embraced. Countries that develop democratic institutions, and thereby adopt power-sharing arrangements, should be given preference when allocating development assistance. Tilting aid to democracies would not only enhance the effectiveness of that aid, it would give clear, powerful incentives to nonrepresentative governments to shift course.
The Millennium Challenge Account (MCA) proposed by President George W. Bush and created by the U.S. Congress in 2004 is a major step in the right direction. Under the MCA rules, democratic governance, transparency, existence of economic rights, and investments in health and education are held up as qualifying criteria for countries to receive assistance. Funding distributed by the MCA should be increased, as originally proposed by President Bush, and the MCA board should continue to exclude nondemocratic countries, even if they meet other eligibility criteria.
But the MCA alone is an inadequate development strategy. The U.S. Agency for International Development (USAID), still the major development actor in the U.S. government, should also offer preferential treatment to democracies and target its assistance to help countries undertaking democratic reforms. The United States should, of course, continue to respond to humanitarian crises, but these funds should be more closely circumscribed than at present so as to avoid inadvertently propping up repressive regimes.
Second, the charters of the World Bank, the International Monetary Fund (IMF), and the regional lending institutions should be amended to favor democratic regimes. Democracies should qualify for larger levels of funding, and their leaders, as legitimate representatives of their societies, should be granted considerable flexibility in identifying development priorities and strategies.
Under the articles of agreement signed at Bretton Woods in 1944, the World Bank and the IMF are currently prohibited from considering democratic legitimacy when making financing decisions. Thus, although a growing number of reformers in these institutions have come to appreciate the fundamental role that democracies play in enhancing development and economic stability, they are handcuffed from rewarding political reform. Moreover, because many of the multilateral development banks and bilateral donors take their lead from the World Bank and the IMF, the effects of this agnosticism have been amplified. To its credit, the World Bank in particular has supported more government-reform projects in recent years. But these projects focus primarily on improving the efficiency of the civil service, regulatory agencies, and control of corruption. Although commendable, they are insufficient because they overlook the reality that the foundation for a system of rule of law is the legal basis on which a society chooses its leaders.
The Bretton Woods provisions that prohibit using political criteria when disbursing aid were originally introduced to entice, albeit unsuccessfully, the Soviet Union into participating and to minimize the role of politics in macroeconomic policymaking. However, there is a major difference between "playing politics" in trying to influence the selection of a particular leader or political party and encouraging a country to adopt institutions that are representative, accountable, and responsive to its population. As the data above make clear, the latter political characteristics have an unambiguously positive impact on the very development goals prioritized by international financial institutions. Taking them into account, moreover, makes more geopolitical sense today than during the Cold War. Two-thirds of the world's states are now either democracies or on a democratic path, double the percentage in 1944. The European Bank for Reconstruction and Development (EBRD), the only regional development bank established since the end of the Cold War, explicitly considers democratic governance as an overarching objective, along with promoting market economies. The EBRD's profitability is proof that considering democratic orientation in financing decisions is an effective strategy.
Third, "democracy impact statements" should become an integral part of international development assistance. Much like environmental impact statements, such statements would include assessments of all major development initiatives to determine the extent to which they support or undermine democratic institutions and processes. After all, aid policies are not enacted in a political vacuum; economic reforms can damage political and social institutions. Adopting an otherwise smart economic reform that winds up throwing democratic reformers out office, for example, makes little long-term sense. These threats need to be taken into account, especially in fledgling democracies, since most backsliding occurs within three years of a transition to democracy.
The intertwining of politics and economics points to the importance of timely and customized assistance in countries that have recently started down a democratic path. Democratizers need to deliver tangible benefits-a "democracy dividend"-to the general citizenry during the early years of the transition. Democracy impact statements would better equip development agencies to help. Of course, even with a democracy impact statement in hand, the challenges facing reformers are complex. Building pro-growth, pro-democracy coalitions is made more difficult by the highly skewed distributions of income and power typically inherited from narrow autocratic systems. And for many societies emerging from autocracy, coalition building is a first-time experience in collective action.
Fourth, aid provided for security purposes should be separated from aid for development. The need to support authoritarian allies on security grounds, including gaining their cooperation on antiterrorism measures, continues to act as a significant constraint on a broader U.S. commitment to democracy-even after the Cold War. Washington's political and economic backing of General Pervez Musharraf's government in Pakistan is a case in point. Although justified on the grounds of enhancing regional stability, this kind of Faustian pact overlooks the reality that autocracies are at the heart of most instances of civil conflict, governance failures, transnational terrorism, and nuclear proliferation-the very perils the United States most wants to prevent.
At the very least, the United States should set up separate funding streams for security and development. When support for an autocratic government is deemed to be vital to U.S. security interests, aid should be committed explicitly by the president in the form of a time-limited "security waiver." Because the justification for such support would be security, funding would be drawn from defense rather than development accounts. Such clear earmarking of funds would make the tradeoffs involved more explicit and minimize situations in which the United States aligns itself with autocratic leaders at the expense of tolerant, forward-looking reformers. The crucial underlying message would be that democracy is Washington's default policy.
Fifth, the United States must create a cohesive development strategy. The goals of alleviating poverty and advancing development involve political, social, and security, as well as economic, considerations. Currently, the Treasury Department, working through the multilateral financial institutions, sets U.S. development policy-and predictably emphasizes economic stability. But development involves much more than macroeconomic policies. Putting development concerns center stage will require the steady involvement of agencies such as USAID, the Millennium Challenge Corporation (MCC), the State Department, the Trade Development Authority, and the Overseas Private Investment Corporation, all of which have a focus on development. The policies of a number of other U.S. agencies-including the Office of the U.S. Trade Representative, the Department of Commerce, the Department of Labor, the Environmental Protection Agency, the CIA, and the Department of Defense-should also be taken into greater account, since they have direct, and sometimes damaging, impacts on development and democracy.
Fashioning an integrated, effective development strategy will require reconciling diverse and often competing interests. To do so, a Development Policy Coordination Council, an executive-level interagency council, should be created. This body would comprise four standing representatives: the secretary of state (who would chair the council), the secretary of the Treasury, the head of the MCC, and the administrator of USAID. It would formulate U.S. development policy, provide guidance to multilateral institutions, and reconcile the aims of all U.S. government agencies whose actions affect the developing world.
We reject a "development first, democracy later" approach because experience shows that democracy often flourishes in poor countries. Moreover, evidence reveals that countries frequently remain poor precisely because they retain autocratic political structures. A development-first strategy thus risks perpetuating the deadly cycle of poverty, conflict, and oppression.
By contrast, a democracy-centered development strategy presupposes not only that poor countries can successfully democratize but also that democracy brings political checks and balances, responsiveness to citizen priorities, openness, and self-correcting mechanisms-all of which contribute to steady growth and superior living conditions. Establishing domestic institutions that hold leaders accountable to their citizenry, moreover, has the potential to shift the burdens of oversight for development initiatives from international institutions to national political structures. Such a transfer of responsibility would alleviate the administrative burden faced by international agencies and foster development strategies better adapted to local needs.
Alleviating poverty and advancing democracy are long, difficult processes susceptible to periodic setbacks. But these struggles should be contrasted with the incomparably worse hardships frequently suffered under autocracies: economic stagnation, humanitarian crises, and conflict. In helping the developing world rid itself of these scourges, the United States and other industrialized countries must make democracy central to their development agendas.