Backlash or Legitimization? Voter Responses to Pro-immigration Reforms

• In Event: Opinion and Immigration

Abstract

Can significant pro-immigration reforms—especially in the contexts where most voters are anti-immigration—be counter-productive? To address this question, this paper estimates the impact of immigration policy on individual attitudes and voting behavior by exploiting the variation in timing of major changes to immigration legislation across the last forty years in OECD countries.

According to public opinion polls, many voters across high-income countries do not want to increase immigration or want to reduce it even further (Gallup 2019). Consequently, many politicians and pundits alike tend to dismiss any possibility of a significant pro-immigration reform as “politically unfeasible.” In line with arguments, there has been a large academic literature documenting the “group threat” that immigrants pose to natives and the related “political backlash” to increasing immigration in terms of the rising salience of the issue and voting for anti-immigration parties (Abrajano and Hajnal 2015, Kaufmann and Goodwin 2018, Dennison 2019).

The extent to which immigration policy itself influences political behavior beyond its effects on immigration levels, however, is still unclear (for some initial exploration, see Abou-Chadi and Helbling 2017). It has now been widely recognized that, in addition to reflecting public preferences, government policy can significantly shape voters’ attitudes and behavior (e.g., Bar‐Gill and Fershtman 2005). On the one hand, a significant pro-immigration reform can cause a political backlash among anti-immigration voters and change their voting behavior even if does not immediately trigger changes in immigration flows. On the other hand, it is also possible such changes can alter the underlying social norms (Tankard and Paluck 2016) and legitimize immigration across the electorate. Relatedly, there can be corresponding reactive and normative responses to anti-immigration policy changes (e.g., Schwartz, Simon, and Hudson 2019).

To disentangle the potential backlash and legitimization effects of immigration reforms, I construct an original dataset linking the best available individual-level public opinion surveys (ESS, WVS, GALLUP) and country-year policy data (IMPIC, DEMIG) for 1980-2014 across 32 OECD countries. I then exploit the variation in the timing of immigration reforms to estimate difference-in-differences models of attitudes and vote choice while controlling for individual characteristics, country characteristics, linear country-specific time trends, as well as country, year, and month fixed effects.

Overall, I find that significant pro-immigration reforms are associated with slightly higher levels of anti-immigration and populist voting in subsequent years. However, I also demonstrate that pro-immigration policy changes are actually associated with reductions in anti-immigration attitudes and populist voting across most groups of voters in the long run. Importantly, these legitimizing effects hold across a variety demographic groups, including more and less anti-immigration voters. To complement the identification of policy effects on political behavior achieved by the difference-in-difference assumptions, I also conduct a number of placebo tests confirming that these effects emerge only after adoption of pro-immigration policies and that immigration policy change does not relate to public attitudes toward other issues.

In the end, I discuss the theoretical and policy implication of my findings. Theoretically, the results imply that immigration policy can have a direct legitimizing influence on immigration preferences above and beyond its indirect, often mixed, effects via changing immigration levels. More practically, these results also suggest that pro-immigration reforms are unlikely to eventually backfire due to political backlash.

Author

• Alexander Kustov, University of North Carolina at Charlotte