Simpler evidence on immigration and institutions: An assessment

Jamie Bologna Pavlik, Estefania Lujan Padilla, Benjamin Powell

Research output: Contribution to journalArticlepeer-review


The “New Economic Case for Immigration Restrictions” has challenged the consensus view that moving from existing restrictive immigration policies to unrestricted, or free, immigration would generate massive global gains in output- the so-called “trillion-dollar bills on the sidewalk” (Clemens 2011). Those making the new case for immigration restrictions generally posit that factors responsible for the low productivity in immigrants’ origin countries could migrate with immigrants and undermine the high productivity in destination countries, thus wiping out the forecasted trillion-dollar gains. The new economic case for immigration restrictions is an empirical conjecture. However, it was presented (by, e.g., Borjas 2014; 2015) as a theoretical possibility without supporting empirical evidence that immigration has, in fact, carried such a negative externality. The new economic case for immigration restrictions does not specify the channel through which the factors responsible for low productivity in origin countries are transmitted to destination countries, but the potential impact immigrants have on formal or informal institutions (norms) are one plausible channel. As George Borjas (2015, 961) asks, “What would happen to the institutions and social norms that govern economic exchanges in specific countries after the entry/exit of perhaps hundreds of millions of people?" One of us (Powell), along with coauthors J. R. Clark, Robert Lawson, Alex Nowrasteh, and Ryan Murphy, published the first paper examining how immigration impacts one formal institution related to productivity-economic freedom (Clark et al. 2015). Numerous papers have since investigated how immigration impacts other formal and informal institutions related to productivity. These papers have used cross-country and cross-state econometric analyses, synthetic control analyses, and analytical narrative case studies.4 Our paper (Bologna Pavlik et al. 2019) was one such study and investigated the impact that immigration could have on corruption in a cross-country analysis. The paper by Garett Jones and Ryan Fraser (2021) in this issue of Econ Journal Watch is framed explicitly as a critique of our paper (Bologna Pavlik et al. 2019) and as a critique of Clark et al. (2015) and Eugen Dimant, Tim Krieger, and Margarete Redlin (2015). Specifically, Jones and Fraser write “Unfortunately, when reporting the relationship between immigration from relatively poor or corrupt countries and subsequent changes in institutional quality, none of the important papers just cited reports simple correlations or scatterplots to give readers a sense of the underlying data. We rectify that omission here. We draw the attention to the potential of overcontrol bias-in particular, of controlling for proxies of the dependent variable-to obscure strong, important statistical relationships in data” (2021, 3-4). Unfortunately for Jones and Fraser, their critique seriously misses its mark. If Jones and Fraser (or the editor and referees of this journal) compared their own results with regards to corruption to the findings in our paper, they would find that they reconfirm our result: there is essentially no relationship between immigration and corruption in destination countries. Similarly, if Jones and Fraser had read Clark et al. (2015) more carefully they would have realized that that paper never examines the flows of immigrants from relatively poorer or more corrupt countries as Jones and Fraser claim. Thus, we are unsure of how that paper could suffer from the overcontrol bias that Jones and Fraser claim that they correct for. In short, Jones and Fraser’s central claim that these studies suffer from overcontrol bias is simply false. Two of the three papers Jones and Fraser (2021) claim suffer from overcontrol bias examine the impact of immigration on corruption-Bologna Pavlik et al. (2019) and Dimant, Krieger, and Redlin (2015)-while Clark et al. (2015) is the only paper that examines the impact of immigration on economic freedom that they claim suffers from overcontrol bias. Jones and Fraser (2021) use the data set from our paper (Bologna Pavlik et al. 2019) to attempt to show how overcontrol bias masks important relationships between immigration and both economic freedom and corruption. Although they use our data set, our paper never examined the impact immigration had on economic freedom; Clark et al. (2015) is the only paper cited by Jones and Fraser that examined that relationship. Economic freedom, not corruption, is the dependent variable in most of the statistically significant results reported by Jones and Fraser (2021). In fact, the phrase economic freedom appears 20 times in their paper, and the abbreviation for its measure, “EFW,” appears 15 times, while the word corruption appears 33 times and the abbreviation for its measure, “COFC,” appears only 5 times. It seems reasonable to conclude that Jones and Fraser (2021) is as much a critique of Clark et al. (2015) as it is of our paper (Bologna Pavlik et al. 2019) and to evaluate it as such.5 We proceed briefly as follows. The next section illustrates how Jones and Fraser (2021) reconfirm the findings of Bologna Pavlik et al. (2019). The section after illustrates how their critique does not apply to Clark et al. (2015). We then critically examine the potential value in Jones and Fraser’s analysis as a standalone empirical contribution to the literature on the relationship between immigration and economic freedom.

Original languageEnglish
Pages (from-to)21-34
Number of pages14
JournalEcon Journal Watch
Issue number1
StatePublished - Mar 2021

Fingerprint Dive into the research topics of 'Simpler evidence on immigration and institutions: An assessment'. Together they form a unique fingerprint.

Cite this