Over the past 25 years, political scientists Edward Mansfield and Helen Milner have made major contributions, both individual and joint, to our understanding of international political economy, particularly in the areas of trade agreements and domestic politics, respectively. The blend of theoretical insight and empirical savvy that characterized their earlier work continues in this book, as does their skill in drawing on economics, international relations, and international law along with their own discipline of political science. Readers from all these disciplines will find this book engaging and cogently presented. The empirical chapters of the book (Chapters 3 through 5) are accessible for readers possessing either familiarity with the basics of modern empirical work or a willingness to take on faith clear verbal presentations of the empirical findings.

The question central to Votes, Vetoes, and the Political Economy of International Trade Agreements is why a rational national leader would choose to enter a trade agreement that constrains the leader's future ability to set policy. Mansfield and Milner (M&M) argue that answering this question requires greater attention to domestic politics than has been typical in the voluminous economics and political-science literatures on trade agreements. The M&M story goes like this: A rational leader must balance protectionist special interests willing to contribute money for favors against voters who prefer less protectionism but lack the ability to ascertain just how much their leader is caving in to special interests' demands for protection. In particular, if the economy performs poorly, the leader does not want to be thrown out of office by voters who assume wrongly that the poor performance results from protection granted by the leader to protectionist contributors. A trade agreement represents an opportunity for the leader to make a visible commitment that voters can monitor because, with the agreement's anti-protection policy terms in place, trading partners' complaints about noncompliance alert voters to episodes of rent-seeking protectionism by the leader. This story could apply to leaders' decisions to join the multilateral General Agreement on Tariffs and Trade/World Trade Organization (GATT/WTO) framework, but the ‘International Trade Agreements' in M&M's title refer to the bilateral or minilateral preferential trade agreements (PTAs) that have bloomed recently – like a thousand flowers or a field of weeds, depending on whom you ask.

Two major testable propositions follow from the M&M story:

M&M test these two propositions both directly and indirectly and find considerable support for both using data for all reciprocal PTAs during the post-World War II period (1950–2005).

Any reader who closes this book unconvinced of the importance of domestic politics in PTAs must have opened the book with very strong priors. But does anyone – academic expert, policymaker, or just interested observer of trade policy – believe that domestic politics do not matter? M&M emphasize not just that domestic politics matter but that, when rational leaders calculate the benefits and costs of entering a trade agreement, those benefits and costs are in part or largely domestic political ones. The authors give three reasons for their focus on political rather than economic costs and benefits: (1) Economic costs and benefits have been emphasized by other authors, so a focus on understanding the political side rectifies an existing imbalance in the literature. (2) Economic costs and benefits are embedded in political ones because economic interests, at least in part, define special interests and other political forces. (3) Much of the literature by economists finds that the net economic effect of PTAs may be rather small. Such groups typically increase members' intra-group trade (the trade-creation effect) but also bias trade inefficiently away from non-members (the trade-diversion effect), leaving an ambiguous net effect on members' welfare. And many preferential trade groups actually grant preferences that are quite small relative to liberalization under the GATT/WTO system, adding to the evidence that such agreements' net economic effects may not be large.

Trade-policy experts, working across disciplines, have come a long way in our understanding of PTAs. Important key pieces of the puzzle like trade creation and diversion, credible commitments, and dispute-settlement mechanisms are in place. M&M contribute by building a stronger link to domestic politics. But, for a phenomenon so widely present in the current international political economy, there is still a lot we do not know (or, at least, do not all agree on). Before we judge ourselves too harshly, it is important to recognize explicitly the complexity of the phenomenon itself. A single country makes thousands of trade-policy decisions: Each product, import or export, coming from or going to each trading partner, is assigned a particular level of openness or protection. Those levels of openness or protection may be the same or similar across products and trading partners, or they may be very different; each may be decided individually, or the decisions may be bunched into groups. Each of those decisions may be implemented unilaterally, or negotiated with anywhere from one to 150+ other countries. A trade agreement may be open or closed to new members. So the M&M question about why a leader enters a PTA is one sub-question in a complex and multidimensional decision space. Why change trade policy at all? Why commit to specific future trade policies? Why the particular set of trade policies covered by an agreement? Why the particular set of co-signers? Why not try to accomplish the same policy goals through the WTO? The M&M domestic-politics story obviously addresses some of these sub-questions more directly and successfully than others. A rational democratic leader maximizes the chance of remaining in office by committing to a specific set of future trade policies with partners who will then alert voters if the leader caves in too much to protectionist domestic special-interest groups.

The reader is left to muse about answers to the other questions and to ponder their connections to the domestic-politics story. A few of many possible examples:

M&M briefly survey PTAs through history and note that their design and effects differed during the nineteenth-century, interwar, and post-World War II periods (pp. 9–13). These differences, along with the always-present data limitations, explain the postwar focus in the book. But it is interesting to think about the extent to which M&M's political cost and benefit calculus might apply to the earlier periods with their very different democracy/veto-player contexts.

PTAs differ significantly in their balance between trade creation and trade diversion. They differ significantly in the extent to which they provide protection to domestic industries through their incorporation of stringent rules of origin (for example, NAFTA's sourcing rules for textiles, apparel, and automobiles). Even individual PTAs change over time. For example, Mercosur has tolerated increasingly egregious protection and recently even changed its membership (suspending Paraguay and accepting Venezuela as a member) under questionable governance procedures (The Economist, 14 July 2012, pp. 31–32). Clearly, these differences between agreements and changes in agreements over time reflect in part domestic political forces. M&M note that PTAs' ‘flexibility’ may be part of their appeal (p. 167), but it would be useful to tie this ‘flexibility’ more directly into the book's central ‘commitment’ storyline.

Another margin on which PTAs differ greatly is in the degree to which they actually achieve their stated goals. Some agreements (e.g., the European Union, NAFTA) preceded dramatic shifts in members' trade toward other members. Many other groups espoused lofty goals but actually changed little in terms of either trade policy or trade patterns. M&M focus on groups' aspirations rather than accomplishments (p. 141). This is understandable given that they ask the ex ante question about why a leader undertakes an agreement, but the failure of so many groups to deliver on their promises ex post might be expected to affect PTAs' efficacy as commitment devices. And what is the role of domestic politics (in particular, democracy or prevalence of veto-players) in explaining the failure of so many groups to deliver?

Near the end of the book, M&M ask whether ‘the world needs some four hundred trade agreements’. If actual trade liberalization were all that was going on, the answer clearly would be no – just as NAFTA, were it actually a free-trade agreement, would not need to be more than two-thousand pages long. The M&M hypothesis, placed in this context, may be somewhat disquieting. Does it imply that a leader does not get a domestic political commitment benefit from agreements already put in place by predecessors? Does each leader have to make new agreements to achieve the desired commitment benefits? If so, then we might expect leaders eventually to produce a lot more than 400 agreements, especially if democracy spreads. After all, barring unusual circumstances, a country can enter the WTO only once; but the number of potential new PTAs is limited only by the daunting combinatorics of the number of states.

My main quibble with the book stems from how the authors frame their contribution. They repeatedly (perhaps too repeatedly) work to distinguish their argument from existing interest-group-based theories of trade policy. However, their brief case studies of Mexico and Japan are hardly distinguishable from well-known existing approaches. This does not mean that the book fails to make real contributions. The core contributions are, on the theory side, teasing out the implications of democracy and veto-players for a leader's interest in PTAs and, on the empirical side, carefully testing those implications. The reader, whether rookie or veteran, will be amply rewarded with valuable insights. And with lots of intriguing further questions to ponder.