Migration Shocks and Voting: Evidence from Ukrainian Migration to Poland

Russia’s aggression against Ukraine triggered two massive inflows of Ukrainians into Poland: conflict-induced labor migration from 2014 onward and a mass refugee inflow after the Russian full-scale invasion in February 2022. We study how local exposure to each shock reshaped voting in Poland. The findings show that greater exposure to labor migrants reduced support for conservative parties in the short run and subsequently shifted voters toward pro-redistribution parties. Both migration waves reduced far-right voting, but this effect emerged only after Ukrainian migrants became salient in public debate and the far-right Konfederacja party adopted anti-Ukrainian rhetoric. The backlash against the far-right is about ten times stronger in areas more exposed to refugees than in areas more exposed to labor migrants.

Two Migration Waves, One Origin Country

Europe has absorbed several large migration waves over the past decade, often followed by a shift to the right in domestic politics. Russia’s invasion of Ukraine has led to the largest war-induced migration in recent European history, and many of the new arrivals have settled in post-communist countries that had long been sources of emigration rather than destinations. Poland stands out: between 2014 and 2023, it experienced two unexpected and very different waves of Ukrainian migration, which provides a rare opportunity to see how distinct types of migration affect local politics.

Before February 2022, Russia’s 2014 aggression and the economic turmoil it produced pushed large numbers of Ukrainian workers into Poland. While these migrants were not necessarily low-skilled, they mostly filled low- and medium-skilled positions, complementing rather than competing with Poland’s abundant supply of highly educated workers (Zuchowski 2025). Crucially, they had no access to Polish social benefits. The situation changed abruptly after Russia’s full-scale invasion in February 2022. Over a million Ukrainian refugees, mostly women and children fleeing an immediate threat to their lives, arrived in Poland. Under the EU Temporary Protection Directive, they received unrestricted access to the Polish labor market and to a broad set of social benefits. About 90 percent of Polish society supported taking in Ukrainian refugees in the immediate aftermath of the invasion. However, as war fatigue set in, the far-right Konfederacja party increasingly relied on anti-Ukrainian rhetoric, which became one of the defining features of its 2023 parliamentary campaign.

Measuring the Local Political Effects

We use county-level data to study how local exposure to each shock changed voting patterns in the Polish parliamentary elections of 2015, 2019, and 2023. Polish counties differ substantially in the number of Ukrainian workers and refugees they received, and we compare the change in vote shares since 2011 between counties with more and less exposure. Because migrants are not randomly distributed across counties, simply correlating migrant inflows with local outcomes could confuse cause and effect. For instance, migrants may settle where labor markets are already expanding. Thus, to isolate the causal effect of immigration, we use three instruments that predict where migrants settled for reasons unrelated to local economic conditions:  the distance to historical hotspots of Ukrainian networks created by the 1947 forced resettlement “Akcja Wisla”, the distance to the nearest Polish-Ukrainian border crossing, and a novel instrument based on the distance to the Polish cities that co-hosted UEFA Euro 2012. The intuition is that each of these instruments drew Ukrainians to certain locations through ethnic networks, lower travel costs, or the connections and visibility that the tournament generated, yet these historical and geographic features had no direct impact on contemporary voting behavior, allowing us to attribute observed effects to the migrant inflows. We classify Polish parties into three non-exclusive groups: conservative (versus liberal), pro-redistribution (versus pro-free market), and far-right (versus non-far-right).

Labor Migration: Away from Conservatives, Then Toward Redistribution

Figure 1 shows the estimated effect of local exposure to Ukrainian labor migrants on voting for the three party groups in the 2015, 2019, and 2023 parliamentary elections. The pattern is clearest for conservative parties: in the first election after the 2014 inflow, a one percentage point increase in the local share of Ukrainian workers is associated with a decrease in the combined conservative vote share of about 0.3 percentage points. For pro-redistribution parties, we detect no statistically significant movement in 2015, but by 2019, the same exposure corresponds to an increase of 0.7 to 0.9 percentage points. In other words, exposure to Ukrainian labor migrants first moved voters away from conservative parties and, over time, pulled them toward parties that promise more redistribution. Voting for far-right parties follows a different pattern. Through 2019, we detect no effect, even though Ukrainian workers had been arriving since 2014. Only in 2023, after Russia’s full-scale invasion made Ukrainian migration highly visible in public debate, does a negative effect on far-right voting emerge, with a one percentage point increase in the local share of labor migrants reducing far-right support by 0.15 to 0.27 percentage points. Empirical evidence on mechanisms from local labor markets provides an intuitive explanation for the first two results: counties more exposed to Ukrainian labor migrants experienced rising wages and falling unemployment, so voters first rewarded openness and then sought a stronger social safety net for themselves, knowing that labor migrants did not themselves draw on Polish social benefits.

Figure 1. Ukrainian labor migration and vote shares in Polish parliamentary elections (2015, 2019, 2023)

Source: Mykhailyshyna and Zuchowski (2026), Figure 2. Each point reports the estimated change in the local vote share of pro-redistribution, conservative, or far-right parties for a 1-percentage-point increase in the local share of Ukrainian labor migrants, using OLS and three instrumental-variables specifications (Akcja Wisla, Euro 2012, and Border). Bars show 95 percent confidence intervals.

Refugees: A Sharp Backlash Against the Far-right

The picture looks very different for the 2022 refugee inflow, summarized in Figure 2. Local exposure to Ukrainian refugees has no statistically significant effect on either the conservative or the pro-redistribution vote share in 2023. The null effect on redistribution fits the fact that, unlike earlier labor migrants, Ukrainian refugees were eligible for Polish social benefits: expanding redistribution would be shared with migrants rather than captured only by natives. The null effect on conservatives likely reflects the broad cross-party solidarity with Ukraine in the immediate aftermath of the invasion, with both conservative and liberal parties initially taking a similar pro-refugee stance. What shows up strongly is an effect on the far-right: a one-percentage-point increase in the local share of Ukrainian refugees reduces the far-right vote share by 1.1 to 1.9 percentage points, roughly ten times the corresponding effect for labor migrants. The most likely explanation combines political salience with direct contact. During the 2023 campaign, the far-right Konfederacja party made opposition to Ukrainian refugees a central theme, using slogans such as “Poland only for Poles” and attacking government spending on refugee aid. In counties with more direct exposure to refugees, that rhetoric appears to have backfired: voters who had personally seen Ukrainian refugees integrate into local labor markets and daily life became less, not more, receptive to anti-Ukrainian messaging, a pattern consistent with Allport’s contact hypothesis (Allport 1954).

Figure 2. Ukrainian refugee inflow and vote shares in the 2023 Polish parliamentary election

Source: Mykhailyshyna and Zuchowski (2026), Figure 3. Each point reports the estimated change in the local vote share of pro-redistribution, conservative, or far-right parties for a one percentage point increase in the local share of Ukrainian refugees (based on PESEL registrations), using OLS and three instrumental-variable specifications (Akcja Wisla, Euro 2012, and Border). Bars show 95 percent confidence intervals.

Conclusion

Ukrainian migration to Poland shows that the political effect of immigration depends not only on how many migrants arrive but also on who they are, how they integrate into local labor markets, and how salient they become in national debate. Labor migrants who complemented Polish workers moved voters away from conservatives and, over time, toward pro-redistribution parties. Refugees who were highly visible, eligible for social benefits, and explicitly targeted by far-right rhetoric triggered a strong backlash against the far-right in areas with direct contact. These results cut against the assumption that migrant inflows mechanically strengthen anti-immigrant parties: under the right conditions, local contact and a positive economic experience can push voters in the opposite direction. For policymakers designing refugee and migration frameworks in the EU and beyond, the Polish case suggests that integration into local labor markets, clear rules on access to benefits, and the nature of political discourse around migrants matter at least as much as the sheer scale of inflows.

References

• Allport, Gordon W., Kenneth Clark, and Thomas Pettigrew, 1954. The Nature of Prejudice, Addison-Wesley, Cambridge, MA.
• Mykhailyshyna, Dariia, and David Zuchowski, 2026. “Migration shocks and voting: Evidence from Ukrainian migration to Poland”, Econ4UA Working Paper No. 25.
• Zuchowski, David, 2025. “Migration response to an immigration shock: Evidence from Russia’s aggression against Ukraine”, Journal of Economic Geography, 25(1), 21-40.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.