Moving Beyond the Policy of No Policy: Emigration from Mexico and Central America

Latin American Politics and Society, Winter 2004 by Rosenblum, Marc R

ABSTRACT

Do Caribbean Basin states influence U.S. immigration policy? Although the terrorist attacks of September 2001 eventually derailed migration talks, before that time Mexico and the United States appeared poised to negotiate a major bilateral agreement, largely on Mexico's terms. Drawing on 88 detailed interviews conducted with Mexican and other Caribbean Basin elites, this article examines sending-state preferences for emigration and their capacity to influence policy outcomes. The informants considered migration to be the most problematic issue on the bilateral agenda, but also saw migration policy as relatively open to source-state influence. A case study of Mexican emigration policymaking details the national and transnational changes that make migration increasingly an intermestic policy issue.

On September 5, 2001, Presidents Vicente Fox and George W. Bush met in Washington to sign off on a framework for a bilateral migration accord. The conservative Bush administration had devoted surprising diplomatic attention to migration, holding three binational cabinet-level meetings and three presidential summits in its first 19 months. In part, this attention reflected the growing influence of U.S. Latinos as potential swing voters. But U.S. domestic politics do not explain the most noteworthy feature of the migration framework negotiations: the five items on the agenda, as well as the language employed in summit statements, were taken directly from Mexican proposals.1

Although progress toward an agreement was derailed by the terrorist attacks the following week, Mexico's ability to set the bilateral migration agenda raises questions about when and how Latin American migrant-sending states influence U.S. migration policy. The literature on migration and on U.S.-Latin American relations is skeptical about this possibility, but this study will argue that migration policymaking is increasingly transnational. In particular, migrant source states have the desire and the ability to influence U.S. migration outcomes, with varying degrees of success.

Neither the literature on U.S. immigration policy nor that on U.S.Latin American relations emphasizes a Latin American role in migration policymaking. The results of 88 in-depth interviews conducted for this study with Latin American elites, mainly from Mexico, El Salvador, and Nicaragua, between 1998 and 2001 begin to fill that gap. These interviews offer insight into poorly studied source-state preferences and highlight a potential role for Latin America in the migration policy process. The Mexican case helps to place these interview findings in context by exploring these issues in greater detail, including a discussion of how changes at the national and international levels encouraged new Mexican policymaking strategies. Other Caribbean Basin cases also contribute to the discussion of the period before the events of September 2001.

THE POLITICS OF EMIGRATION POLICY

Do migrant-sending states influence U.S. immigration policy? On its face, immigration would seem to be the "quintessential intermestic issue" (Lowenthal 1999, 124; also see Domínguez and Fernández de Castro 2001; Manning 1977; Rosenblum 2000). That is, policies made in the United States are felt in migrant-sending states, and those states may have the desire and ability to influence U.S. policy outcomes. But this expectation is not generally supported by the existing literature. While there are ample reasons to expect that Caribbean Basin states should care about emigration, exactly how they weigh its pros and cons has not been well studied. In any event, most analysts are skeptical that migrant-sending states actually play a policymaking role.

Emigration undoubtedly has profound effects on the economies and societies of high-emigration Caribbean Basin states. On the economic side, emigration benefits states as a "safety valve" to relieve low employment rates and raise wages (O'Rourke and Williamson 1999). Indeed, source-state wages-and not U.S. wages-are important predictors of emigration (Massey et al. 1998), especially in the Mexican case (Hanson and Spilimbergo 1999).

Emigration is also a source of hard currency through migrant remittances. Official remittances to the top five Latin American countries of origin (Mexico, the Dominican Republic, El Salvador, Guatemala, and Colombia) have grown 26 percent annually since 1980, reaching $8 billion in 2000 (Lowell and de la Garza 2000). Total inflows-including untracked hand-carried cash flows-may be one-and-a-half times this level (Haus 1999; Lowell and de la Garza 2000). As a result, remittances are the largest source of foreign exchange for El Salvador, Nicaragua, and other small Caribbean Basin states (Díaz Briquets and Pérez-López 1997; Mahler 1999; Vega and Despradel 1999) and the third- or fourth-largest source of hard currency in the larger Mexican economy (Urrutía 2000). Even these figures understate the value of remittances, which are an efficient form of social welfare targeted directly to migrants' families (Meyers 1998).