Can and should the United States promote democratic change abroad? In his book Michael McFaul responds to this timely question with an unequivocal yes. He offers three interrelated arguments on behalf of this position. The first is that democracy is superior to alternative forms of government because of its political and economic, as well as its domestic and international benefits. The second is that the spread of democratic governance around the world serves the interests of the United States. The third is that, in fact, US democracy assistance has had positive effects on democratic development in other countries. For these reasons, McFaul maintains, that the United States should remain committed to actively supporting democratic development abroad. At the same time, however, McFaul acknowledges problems with aspects of the democracy promotion enterprise and offers a long list of suggestions about how the United States can do a better job of advancing democratic progress.

It would be hard to quibble with the first two arguments presented. It is very easy to construct a case in support of democratic progress abroad if we approach this issue from the vantage point of ordinary citizens or from the perspective of the values and interests of the United States. A considerable amount of research over the past decade has demonstrated that democratic regimes out perform other types of regimes in such critical areas as the provision of human rights, the physical quality of life, and rates of economic growth. Moreover, democracies have been shown to be less corrupt than other types of regimes, and they are also less likely to instigate international conflicts or to carry out and then maintain radical policies that have dire human consequences (see, for example, Morton Halperin, Joseph Siegle, and Michael M. Weinstein, The Democracy Advantage: How Democracies Promote Prosperity and Peace, rev. ed., 2010). Finally, a number of studies have argued convincingly that US support for democratic regimes and democratic progress enhances US influence in the international system. For all these reasons, it is far from controversial to claim that the United States should welcome democratic change abroad.

The same cannot be said, however, of the author's broader policy prescription. It is one thing to conclude that the US has reasons to “welcome” democratic progress in the international system and quite another to take the further step—one might even say leap—of asserting, as McFaul does, that the United States should play an active role in supporting the global spread of democratic governance. his rendition of the impact of US democracy aid is too rosy. If it rests upon a relatively selective reading of the available studies in this area, it also overstates the quality of those studies and the confidence we can have in drawing useful generalizations from them about the contributions of the United States to the global advance of democracy. These problems, in turn, lead us to wonder about the extraordinarily long list of recommendations that McFaul puts forward regarding needed improvements in US democracy assistance. If there is so much wrong with what the United States has done and how it has done it, then why should we be confident about its ability to advance democratic progress in the international system? At the same time, why should we assume that institutions such as the U.S. Agency for International Development (USAID) are capable of making the considerable changes that he proposes? Indeed, aspects of McFaul's argument bring to mind Mikhail Gorbachev's defense of Soviet socialism. The system, he argued, was fundamentally sound, despite the fact that it had performed very badly during the Stalinist and Brezhnev eras—that is, for nearly two-thirds of the time that the Communist Party had been in power. Gorbachev never considered the possibility that it was not the system's problems, but its rare and partial successes, that were the aberration.

We thus prefer a more modest set of recommendations for future US democracy assistance efforts. Our views on this subject have been strongly influenced by a study we have been conducting on the role of the United States and other international and domestic actors in a series of key elections that took place from the early 1990s to 2008 in mixed regimes in postcommunist Europe and Eurasia. All of these regimes straddled democracy and dictatorship by combining authoritarian politics and competitive elections—arrangements, it is important to recognize, that have proliferated over the course of the third wave and that are seen by McFaul as optimal targets for US democracy aid (Valerie Bunce and Sharon Wolchik, Democratizing Elections, International Diffusion, and U.S. Democracy Assistance, forthcoming).

Our study sheds new light on whether and how the United States should promote democracy abroad. While most analyses of democracy assistance rely on individual case studies (which may not be representative) or statistical assessments of how US democracy and governance aid relates to various measures of democratic performance (which are very removed from complex dynamics on the ground), we stake out a middle ground between these methodological extremes, employing a range of methods to compare 11 elections in a specific region sharing distinctive historical and structural features. This eclectic approach has a number of analytical advantages. For example, we focus on a key aspect of democratic change (the defeat of authoritarian leaders and, more generally, electoral turnover), as well as longer-term patterns in democratic performance; we compare successful and failed attempts by the opposition and its allies to win power; and we introduce a number of controls into the analysis, such as unity of the opposition and the character of postelection protests. Just as important, the United States (along with a variety of other international and local actors) was very involved in supporting democratic change in all of the countries we analyzed and in these particular electoral exercises. Our study thus allows us to address two key questions: What kinds of aid did the United States and other actors provide, and how much influence did the United States and these other actors exert on electoral results?

Our research, together with other recent work, suggests that the controversies surrounding US democracy aid exist for very good reasons (see, for instance, Zoltan Barany and Robert G. Moser, eds., Is Democracy Exportable? 2009). Ambitious projects that seek to remake politics and economics, especially when they are fashioned by powerful outside actors who do not have to live with the consequences of their actions, are always risky endeavors (as argued, for example, by Nicholas Guilhot, The Democracy Makers: Human Rights and the Politics of Order, 2005; James Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed, 1998; and Janine Wedel, Shadow Elite: How the World's New Power Brokers Undermine Democracy, Government, and the Free Market, 2009). This is especially true when, as with democracy assistance, there is no consensus about “what is to be done”—to borrow from Vladimir Lenin, who was certainly one of the most ambitious political and economic architects in history. The fact is that there are competing theories of how democracies arise, sustain themselves, deepen, and break down, and there are, in addition, spirited debates about whether and, if so, how international actors can contribute—in negative, as well as positive ways—to these complex processes. Democracy assistance, in short, involves more whistling in the dark than McFaul seems to suggest.

Such assistance also seems to have far more variegated effects than the author recognizes. To evaluate the impact of US democracy aid, one needs to address some difficult questions. In particular, what specific measures did the United States undertake (US programs typically distinguish between support for the rule of law, elections, governance, and civil society)? Were these efforts ambitious or modest? Were the effects of support large or small, positive or negative—distinctions that remind us that the consequences of democracy assistance can be mapped on a two-by-two table? Finally, what did other actors and processes contribute to these outcomes? Because McFaul fails to address these hard questions in a systematic way or to refer to studies of democracy assistance that vary along each of these important dimensions, the reader cannot be certain that he has put his enthusiasm for US democracy promotion to a sufficiently tough test. Moreover, the book fails to provide a comparative standard that would enable the reader to judge whether the “successes” he discusses, such as the defeat of Slobodan Milosevic in Serbia in 2000, are typical or unusual events with respect to the degree of US engagement and the positive consequences of US support. We think the Milosevic case was in fact an unusual one, in part because of the extraordinary work of local actors. In the absence of this kind of information, moreover, it is hard to know what underpins success and, for that matter, the extent to which the United States acted alone or as part of a consortium of players.

One reason that McFaul avoids presenting a systematic assessment of US democracy aid may be because there are in fact a limited number of rigorous studies of this subject. It is ironic, for example, that a major conclusion of a recent effort by the United States Academy of Sciences to evaluate democracy assistance was for the need for more assessments of such assistance (National Research Councils of the National Academies, Committee on Evaluation of USAID, Democracy Assistance Programs, Improving Democracy Assistance: Building Knowledge Through Evaluations and Research, 2008). A key reason why there are so few assessments of democracy assistance is that such studies are devilishly difficult to carry out. Analysts do not agree on how to measure democratic progress (in part because available measures do not correlate very well when extreme scores are eliminated) or what would constitute the appropriate time span for assessing both the influences in play and the outcomes of those influences.

Moreover, the field of plausible causal factors is remarkably crowded, ranging from a variety of structural, institutional, and agency effects to domestic and international influences. Indeed, one discovers pretty quickly when doing fieldwork in this area that much of the day-to-day effort associated with democracy assistance is carried out by loose-knit transnational networks composed of a wide range of US-funded local and US organizations; private foundations based in the United States and Europe; the European Union and various European governmental organizations; and local opposition and civil society groups. Regional actors also play a role, especially as the newest members of the European Union, such as the Poles and the Slovaks, have shifted from being recipients to donors of democracy assistance, especially in their postcommunist neighborhood (Tsveta Petrova, “From Recipient to Donor: New Europe Promotes Democracy in the Neighborhood,” Ph.D. diss. in progress, Cornell University).

Separating out the impact of the United States, therefore, is extraordinarily difficult. It is especially so since US democracy assistance in particular is very decentralized—to the point where the flow of information in most cases is from the field to Washington and not vice versa; chaotically organized; and features unusually high rates of personnel turnover (see, for instance, Thomas Melia, “Toolbox: The Democracy Bureaucracy,” The American Interest 1 [4]: 122–31, and, for ideas on how to reform the democracy assistance bureaucracy, Thomas Carothers, Revitalizing Democracy Assistance: The Challenge of USAID, 2009).

These considerations have two important implications. First, while it is easy to appreciate the value of democracy and to hope that more countries will become democratic in the future, it is much harder to be an unqualified champion of US democracy assistance. Second, for many of the same reasons, we are uncomfortable with accepting the long and ambitious “to do” list that McFaul provides. There are too many uncertainties associated with democracy assistance to be confident about embracing this mission whole-heartedly or rushing to accept the priorities he delineates and the reforms he advocates.

These implications recognized, however, we would nonetheless offer some modest observations, based upon the evidence we have gathered, about the conditions under which US democracy and governance aid is more likely to make positive contributions to democratic change. First, like McFaul (though for some additional reasons), we think that elections provide an important opportunity for democratic change. While we know that democracy cannot be reduced to elections, we also recognize that citizens around the world are prone to associate democracy with elections, that external support for free and fair elections is a less objectionable enterprise than other types of interventions associated with democratic development, and that elections in fact can have important consequences. For example, democratic improvements, if they take place, tend to occur following elections (Jan Teorell and Axel Hadenius. “Elections as Levers of Democratization: A Global Inquiry,” in Staffan Lindberg, ed., Democratization by Elections: A New Mode of Transition, 2009). At the same time, elections solve coordination problems—not just among ordinary citizens and opposition parties, but also among external supporters of democratic progress.

Second, some elections are more likely to produce democratic change than others. We found through our comparative analysis of elections in mixed regimes that the key difference between successful and failed attempts to defeat authoritarian leaders or their anointed successors was whether oppositions adopted a sophisticated approach to winning power: that is, one that included forming coalitions among opposition parties, running ambitious campaigns, establishing close alliances between opposition parties and civil society groups, and carrying out sophisticated voter registration and turnout drives, along with extensive vote monitoring exercises. What is important about these electoral strategies is that they were invented elsewhere (such as in the Philippines in the successful campaign against Ferdinand Marcos in 1986); they were transferred to the postcommunist region by US democracy promoters; and, as a result, they were amenable to a process of deliberate international diffusion across regions and states. Moreover, it was the United States, in conjunction with other international actors, which provided ideas, money, and training. At the same time, long-term investment in civil society, which is a distinctive component of the US approach to democratic change, also paid off during the campaigns, at the ballot box, and, if the regime tried to steal the election, in the streets. Strong civil societies also contributed to greater democratic progress after democratizing elections that ousted semiauthoritarian leaders and brought the opposition to power.

The United States, however, did not act alone in the electoral breakthroughs that played a decisive role in improving democratic performance in Croatia, Serbia, Slovakia, and Ukraine. These breakthroughs were, in fact, collaborative struggles carried out by the United States, the European Union, a variety of Western-based foundations, local opposition and civil society groups, and, finally, regional actors who had succeeded in defeating authoritarians in their own countries and wanted to help achieve a similar outcome in neighboring states. US influence, in short, depended upon, rather than substituted for, local and regional commitment and capacity.

Our research, in short, leads us to agree with McFaul that the US can play a useful role in assisting democratic development outside our borders. But, and this is a crucial qualification, it can do so only when certain relatively rare conditions exist. When domestic circumstances—including the long-term development of civil society—permit; when domestic actors are willing and able to carry out the ambitious, difficult, and sometimes tedious tasks outlined above; and when the US, instead of acting alone, works as part of a transnational coalition that includes other governmental and nongovernmental actors, then policies of democracy promotion can prove successful. But because such a fortunate confluence of conditions is so rare and so precarious, we ought to proceed with much greater caution than McFaul allows.