The effect of low-skill immigration restrictions on

In Section 2 we build a basic model of the effect of restrictions on 1 A limitation of that work is that in the United States, natural randomization of high-skill work visa petitions occurs at the level of the individual foreign worker, not at the level of the firm. [...] The ? -value is for a two-sample test of the null hypothesis that the fraction of petitions receiving each lottery letter in the survey sample is equal to that fraction in the ∗ universe of petitions. [...] The final row gives the p-value of Fisher’s exact test of the null hypothesis that the lottery-result distribution across all five letters is equal in the sample and the universe. [...] 23 The results of this comparison are substantially the same when the sample of survey-reported petitions is re- stricted to the 328 petitions that were filed by the 251 firms in the core regression sample for the analysis to follow. [...] The ranking of industries in the sample and universe are similar, with the exception of forestry and construction: Forestry is overrepresented in the sample (ranked third in the sample, seventh in the universe) and construction is underrepresented in the sample (ranked eighth in the sample, third in the universe).