Comment: on Estimating Elasticities of Substition

The literature that attempts to measure the labor market impact of immigration grew explosively in the past decade. This growth was accompanied by a noticeable methodological shift. The more recent studies no longer represent atheoretic attempts to measure the correlation between wages and immigration-induced supply shocks. Beginning with the introduction of the nested CES framework into the immigration context by Borjas (2003), the recent literature explicitly strives to estimate the various elasticities of substitution that are mainly responsible for the presence (or absence) of an impact of immigration on the native wage structure. The papers by Ottaviano and Peri (2011) and Manacorda, Manning, and Wadsworth (2011) in this issue represent additional efforts to estimate the relevant elasticities in the American and British contexts, respectively. Despite the closer link between the empirical work and the underlying factor demand theory in recent work, there is still a great deal of disagreement on whether immigration has an impact on the earnings of native-born workers. As an example, Card (2009) argues that the effects of immigrants on US wages are small, whereas Borjas (2003) and Aydemir and Borjas (2007) suggest the opposite, that recent immigration has reduced US wages, particularly for low-skilled natives. Card (2009) notes that there are two elasticities of substitution whose magnitudes are at the core of the debate. The first is the elasticity of substitution between similarly skilled immigrants and natives: “If immigrants and natives in the same skill group are imperfect substitutes, the competitive effects of additional immigrant inflows are concentrated among immigrants themselves, lessening the impacts on natives” (Card 2009, p. 2). The impact on natives could be further weakened if the elasticity