The rapid spread of preferential trade agreements (PTAs) is one of the most interesting phenomena of the contemporary international political economy. Especially since the end of the Cold War, countries across the world have been signing a large number of PTAs, many of which go substantially beyond simple agreements regulating trade in goods. For nearly as long, Edward D. Mansfield and Helen V. Milner have done research on PTAs. In Votes, Vetoes and the Political Economy of International Trade Agreements, the culmination of this research program, they make a forceful case for the important role that political institutions play in the political economy of PTAs.

Mansfield and Milner argue that two variables are key in explaining which countries conclude trade agreements: regime type and the number of veto players. For one, democratic governments are more likely to sign PTAs than are autocracies, as they can use such agreements to convince the public and pro–free trade interest groups that the government is pursuing an open trade policy. The argument builds on the assumption that the median voter is more liberal than the government, but that he or she has limited information about governmental preferences and policies. When the country experiences economic difficulties, the median voter will punish the government, assuming that it implemented trade policies that extract rents, even though an exogenous shock may have brought about the economic downturn. To avoid being voted out of office, democratic governments can sign PTAs that credibly commit them to an open trade policy. Since autocracies are less dependent on the median voter, they do not need PTAs as a commitment device.

The second argument is that the number of veto players is negatively related to the probability of PTA ratification. As the number of veto players increases, protectionist interests find it easier to sway at least one of them and thus block the ratification of a PTA. While governments may be able to buy off individual veto players—for example, by including provisions in a PTA that soften the opposition by protectionist interests—as the number of such players increases, this policy becomes increasingly costly. It is also unlikely that foreign governments will accept the inclusion of ever more provisions that buy off domestic interests.

The authors use four brief case studies in the theoretical chapter to show the plausibility of their argument. Across three empirical chapters, they then carry out several systematic tests of the argument. Controlling for other influences on the probability of a PTA being signed, such as the presence or absence of a hegemon, the distance between two countries, and the size of the participating countries' economies, they find consistent support for their argument. Democracies turn out to be about 55% more likely to enter a trade agreement than are autocracies, and countries with a large number of veto players are 35% less likely to ratify a PTA than are countries with few veto players.

A particular strength of the book is the attempt not only to test the two main hypotheses but also to derive a series of auxiliary hypotheses and test them against the available data. Among these auxiliary hypotheses are 1) that democratic leaders who sign PTAs are in power for longer than those that do not, and 2) that left-wing parties are more likely to sign PTAs than are right-wing parties. These tests add considerable plausibility to the argument that relates regime type and veto players to the signing of PTAs.

Overall, this is a highly compelling book that deserves a wide readership. The authors managed to anticipate and defuse many potential objections to their argument. Moreover, the empirical examination serves as a model of excellent research. Nevertheless, a few questions emerge that merit further research.

First, the authors advance some evidence showing that the public is informed about trade agreements, but much of this evidence remains anecdotal. Clearly, many German voters know about the European Union and many US voters have heard about the North American Free Trade Agreement. But do German voters also know about the EU's association agreement with Jordan, and do US voters know about the US-Oman free trade agreement? And even if they know, can these agreements with relatively small trading partners serve as a credible commitment that the government is not giving in to rent seeking? The causal story thus may apply to some agreements (namely, those with relatively important trading partners) better than to others. One may even wonder why—given the specific argument advanced in this study—the US government, which can use NAFTA and membership in the World Trade Organization to signal commitment to an open trade policy, bothers about signing additional PTAs at all.

Second, I was left wondering whether the authors capture the effect of democracy or the effect of transition to democracy. The case studies that accompany the discussion on the effect of regime type deal with South Africa's decision to join the South African Development Community and the decision of Argentina and Brazil to create the Common Market of the Southern Cone. All three countries underwent a transition to democracy shortly before taking these decisions. The same applies to the many Central and East European countries that signed a large number of trade agreements just after emerging from dictatorship. It would be interesting to test whether there is anything particular about recent democracies that causes them to sign PTAs.

Third, as the authors readily admit, the empirical test of whether PTAs enhance the longevity of democratic leaders is only suggestive. In fact, the causal mechanism stipulates that PTAs are most helpful for the ability of governments to stay in office during economic downturns. An empirical test of this conditional effect could have offered more clear-cut support for the argument than the finding that democratic leaders that sign PTAs stay in power longer than leaders who do not. Several alternative hypotheses, including the reverse hypothesis that leaders need to control a stable majority in parliament to be able to sign PTAs, are compatible with the evidence offered for this step in the causal story.

Finally, the authors hardly question why many recent PTAs cover much more than tariffs. Many post-NAFTA PTAs contain provisions on intellectual property rights, investments, public procurement, and more. Inasmuch as it seems implausible that the median voter both is informed about all of these provisions and has a preference for more liberalization on all of them, the present argument does not seem able to shed light on this trend.

Mansfield and Milner's book serves as an excellent starting point for future research that will tackle these puzzles. However, it will also be of great interest to political scientists and political economists interested in broader questions of international cooperation. These readers will find many interesting implications of the argument for additional debates, such as those on the design of international institutions and on the role of power and domestic politics in international relations.