Since the Pinochet regime in Chile privatized its public pension system in 1981, several Latin American countries have enacted public pension reforms, usually incorporating some structural reform or privatization. Why have some countries privatized their public pensions (Chile, Mexico, Bolivia), while others add optional private pensions (Argentina, Uruguay), create parallel systems with both public and private pensions (Colombia, Peru), or fail to enact any structural reform (Brazil)? Retiring the State will be required reading for those interested in the answer to this question.

On the basis of a qualitative comparison of pension reforms in Mexico, Argentina, and Brazil in the 1990s, Raúl Madrid develops an explanation of pension privatization that combines economic, ideational, and political factors. The explanation is complex because it claims that the policy process proceeds in path-dependent stages and that different factors predominate at different stages. Initially, whether pension reform appears on the policy agenda depends on economic factors, such as the cost of the public pension system and the need for domestic savings. Once reform reaches the public agenda, the executive formulates a reform proposal. At this stage, a privatization proposal, in contrast to a parametric reform, is most likely if the Chilean model is salient, the World Bank participates in policy discussions, and domestic policymakers are neoliberal economists. If policymakers develop a privatization proposal, its ultimate fate will be determined during the enactment stage of the policy process. During this final stage, political factors, such as executive control of the legislature and party discipline, explain whether privatization proposals are enacted with little change, modified, or blocked. At least two results are possible at each stage in the policy process: reform or no reform on agenda; parametric or privatization proposal; and blocked, modified, or enacted privatization proposal.

This description of the policy process is reasonable, yet it implies at least five (and almost certainly many more) possible outcomes, depending on the path a reform proposal takes. Madrid's argument is derived from only three observations: Mexico's 1995 pension privatization, Argentina's 1993 creation of a mixed public/private pension system, and Brazil's 1998 parametric pension reform. In the early 1990s, however, the Mexican executive abandoned a full privatization proposal in favor of a less contentious reform (p. 74), and both Presidents Collor and Franco in Brazil abandoned pension reform proposals due to “overwhelming opposition from both inside and outside the government” and “staunch resistance” (p. 145). To be consistent, the explanation should explain the failed reform attempts as well as those that succeed. Regrettably, Madrid does not give these additional observations methodical analysis. The reader is left wondering whether changes in Madrid's explanatory factors were significant enough in Mexico after 1992 to explain why privatization became possible in 1995. Was President Cardoso's administration able to pass reforms that failed under his predecessors due to changes in executive control of Congress and party discipline? Madrid's overall argument would be more forceful had he more systematically addressed all the variations in his country studies and used these failed reform observations to reinforce the explanation of enacted reforms.

Fortunately, Madrid's analysis does not end with his three country studies. He then tests his explanation of pension privatization on a broader sample of middle- and upper-income countries using quantitative methods. Although some may criticize the choice of statistical models, the quantitative tests are simple and straightforward. They largely confirm the pattern suggested by the qualitative analysis. The accompanying discussion of additional cases of pension reform in Latin America and the postcommunist world will be useful and appealing to a wide audience. Madrid offers a careful description of pension reforms, explanatory factors, and policy process for several countries in both regions. The consistency with which he can explain a wide range of outcomes in many countries is a convincing test of his argument and demonstrates a strong understanding of the reform process in a wide range of cases.

Overall, this book makes at least two important contributions to the literatures on second-generation market reforms and policy diffusion. First, Madrid demonstrates the discreet way in which the World Bank has influenced pension reform in Latin America. Many Latin American countries were considering pension privatization before the publication of the World Bank's Averting the Old Age Crisis (1994). Further, World Bank loans often followed, rather than preceded, reform and seldom were large enough to finance the full costs of privatization, according to the author. Instead, the World Bank “create[d] an ideational climate that was propitious for pension privatization” by providing technical assistance through pension reform missions in several countries that “enabled World Bank representatives to maintain extended contacts with host country social security policymakers” (p. 176). His description of World Bank influence in pension reform echoes Judith A. Teichman's use of international policy networks among World Bank and domestic policymakers to explain market reforms of the 1980s and 1990s in Latin America (The Politics of Freeing Markets in Latin America, 2001). Second, Madrid correctly emphasizes that the Chilean pension model and the positive externalities associated with privatization were a powerful influence on policymakers in the early 1990s in Latin America. Although recent studies have called into question the effects of pension privatization on domestic savings and economic growth in Chile and elsewhere, policymakers in Latin America were swayed in the early 1990s by limited but powerful evidence from Chile. Thus, Madrid's book offers a potent illustration of how policy innovations, including those with short histories or even limited evidence of success, can become important reform models for other countries and quickly diffuse through international policy networks. In these two ways, this book makes important contributions to our understanding of market reform and policy diffusion.

In Retiring the State, Madrid has pulled together an insightful and complex description of pension reform in Mexico, Argentina, and Brazil. He has also offered a careful and convincing explanation of the policy process that produces pension privatization. As its subtitle suggests, this book is likely to generate great debate and inform the growing literatures on pension reform, policy diffusion, and second-generation market reforms in Latin America and beyond.