Overestimating Refugees’ Economic Impact: An Analysis of the Prevailing Economic Literature on Forced Migration
The UNHCR reported in June 2016 that the number of number of refugees, asylum seekers, and internally displaced persons reached a record high of 65 million individuals worldwide. How will all these individuals impact the economies of the countries in which they find asylum? Do refugees, as some politicians claim, force domestic workers out of the labor market? Do refugees exert a substantially negatively net fiscal impact? This paper attempts to address these questions by analyzing the prevailing economic literature from 1990 until the present on refugees’ and immigration’s economic impact. I argue that, although the estimated economic effects of immigration and refugees vary, their overall impact is negligible. Refugees exert a slightly negative impact on domestic wages if domestic labor is immobile. At the same time, refugees’ net fiscal impact depends more on their tax contribution—which is a function of their labor market integration—than their consumption of publicly funded goods and services.
Refugee and forced migration issues have dominated recent political debates in Europe and other parts of the world. Those on the political Right claim that refugees threaten European national security, economic prosperity, and cultural traditions; those on the political Left claim that the influx of refugees represents a humanitarian crisis that demands accepting additional refugees. Perhaps the Economist’s attempt to reconcile these two opposing opinions puts it best: “Humanity dictates that the rich world admit refugees, irrespective of the economic impact. But the economics of the influx still matters.” Setting aside that international law does require states to provide refugees with asylum, this paper attempts to address the two most salient economic questions regarding refugees’ arrival in Europe.
First, do refugees displace domestic workers, leading to higher rates of unemployment and lower wages? Second, do refugees exert significant strains on public finances? While perfect data do not exist to answer either of these questions beyond the shadow of a doubt, evidence suggests that, in the short run, granting refugees asylum leads to a negligible overall effect on the labor market and public finance. In the long run, however, refugees positively contribute to the labor market and public finances, the extent of which depends mostly on the success of their integration into the economy.
Refugees and the Labor Market
Economists disagree about the precise nature and extent of immigration’s impact on the wages and unemployment rates in immigrant-receiving countries. On the one hand, some studies suggest that immigration has “essentially no effect on the wages or employment outcomes” of domestic workers. David Card’s famous analysis of the Mariel Boatlift found that refugee immigration had a positive, yet minimal impact on the Miami economy due to the city’s ability to absorb refugees into previously unexploited sectors. On the other hand, George Borjas argues that, because of labor mobility, the impact of immigration on unemployment and wages may be tenuous in regional labor markets, while simultaneously depressing labor market conditions at a national level. Borjas measured skilled and unskilled immigrant labor in terms of educational qualifications and found that a 10 percent increase in the labor force due to immigration resulted in a three to four point decrease in domestic workers’ wages.
However, other studies attempt to find some sort of middle ground, disputing both the argument that immigration has no effect on the labor market and the argument that immigration drastically depresses wages and employment. Gianmarco Ottaviano and Giovanni Peri, for instance, adopt the qualification bands from Borjas’ framework, but they assume that, even within those bands, immigrant and domestic workers are not perfect substitutes. In other words, immigrants with the exact same educational qualifications as domestic workers can function as “imperfect substitutes” because of labor market discrimination. Even if immigrants could do the same job as domestic workers, they don’t practically function as perfect substitutes because, in reality, they may not be hired by employers who consider them less capable because of their race, ethnicity, nationality, language, etc. As a result, Ottaviano and Peri find that immigration has “a small effect on the wages of native workers with no high school degree (between 0.6% and +1.7%) […and] a small positive effect on average native wages (+0.6%).” Moreover, Ottaviano and Peri also note that, given the standard error, this effect is not “significantly different from 0.” The largest impact on the labor market observed was on the wages of previous immigrants, which were found to have “a substantially negative effect (−6.7%).” Thus, even at the theoretical level, the effect of immigration on the labor market has been highly contested.
Later, even more tweaks were made to the traditional methodology used to study the economic effects of migration. Stephen Nickell of the University of Oxford and Jumana Saleheen of the Bank of England recently studiedmigration’s impact on average British wages in any given region of the country between 1992 and 2014. Crucially, Nickell and Saleheen measure skill distribution by occupation, a clever methodological tweak considering “that it is often very tricky to accurately compare education qualifications across countries.” In addition, treating skill distribution as a function of occupationhelps to translate the economics of migration directly into the jargon of public discourse, which treats immigrants principally by occupation rather than by educational attainment, such as with the “stereotype of the Polish plumber—used widely as a symbol of cheap labor.” Ultimately, Nickell and Saleheen findthat migration exerts “a statistically significant, small, negative impact on the average occupational wage rates of the regions” studied. The largest effect on wages observed related to semi-skilled and unskilled labor, where a 10% increase in migrant labor resulted in a 2% decline in the average wage. Nickell and Saleheen’s occupational measure of qualification might be said to be more accurate than educational measures such as Borjas’ considering that, oftentimes, educational credentials do not transfer between countries. Therefore, Nickell and Saleheen’s findings suggest that refugees immigrating to Europe may adversely affect the labor market, but not nearly to the extent that some politicians claim.
Moreover—and with specific regard to refugees—the Economist notes that the wage-dampening may “even have positive side-effects” for the domestic labor market. A recent paper by Mette Foged and Giovanni Peri finds that, in Denmark between 1991 and 2008, domestic workers pushed out of low-skilled industries by refugees changed jobs to other, “less manual and more cognitive” labor-intensive sectors. Such jobs included “legislators and senior [government] officials,” “corporate managers,” and even “skilled agricultural and fishery” sectors. By contrast, the proportion of refugees composing manual skilled sectors such as “machine operators,” “drivers,” and “mining laborers” rose substantially, resulting in “positive or null wage effects and positive or null employment effects” for domestic populations over the long run. So, to the extent that refugees substitute for domestic labor—however imperfect that substitution may be—their overall economic impact also depends on the abilities of displaced domestic workers to find employment in other sectors. Additionally, evidence exists from Congolese refugee camps in Rwanda to suggest that one additional refugee receiving cash aid contributes an estimated $205 to $253 to the local economy. Taking the difference between contributions and per-refugee cash aid, refugees yielded a positive individual contribution of between $70 $126 annually. Most of the refugees’ individual contributions resulted from spillovers with the local economy, such as the “purchase [of] goods and services from host-country businesses outside the camps.” If refugees displace workers who move into other sectors of the economy and experience higher wages, then they also positively contribute to the sales of local businesses.
Refugees and Public Finances
Refugees exert a similarly ambiguous impact on public finance as they do on the labor market. In fact, a 2013 OECD report notes that including or excluding non-personal sources of tax revenue, such as corporate income taxes, as well as non-excludable goods like roads, in an analysis of immigrants’ net public fiscal impact “often changes the sign of the impact” itself. Estimates of immigrants’ net fiscal impact thus vary depending on the methodology employed, although the report’s main findings suggest that—however measured—the impact “rarely exceeds [plus or minus] 0.5% of GDP in a given year.” In fact, the OECD observed the highest impact on public finance in Luxembourg and Switzerland, where immigrants positively contributed an estimated 2% of GDP to the public purse. Compared to domestic populations, however, the OECD report found that, on average, immigrants have a lower net fiscal contribution overall.
This is an especially salient concern in the short run, because refugees can potentially exacerbate strains on the public purse, contributing to increases in demand for public services while the supply of those resources remains temporarily fixed. In fact, precisely because of the protections afforded to asylum seekers under international law, “additional public spending for […] housing, food, health, and education, will increase aggregate demand,” therefore making such services more costly to provide, all else equal. However, in many cases, the short-term costs of accommodating asylum seekers are borne by international donors rather than governments. In fact, University of Oxford Professor Emeritus Roger Zetter notes that global programs to accommodate refugees in the short term total 8.4 billion USD globally, but that economists “rarely analyze the economic outcomes of their program[s].” Instead, they “tend to assess the impacts and costs for the host community” as a percentage of GDP regardless of whether or not the government actually pays for the accommodations provided to refugees. Such analyses are frankly misleading because, while aggregate demand for publicly funded goods may increase in the short run, the cost of meeting such a higher demand puts strain on NGOs, the UNHCR, and other international donors, not on governments.
Among the three countries with the highest numbers of Syrian refugees—Turkey, Lebanon, and Jordan—GDP is expected to rise, while estimates of the costs of accommodating refugees are paltry. The OECD predicts that Turkey’s GDP growth will “remain close to 4% per annum in 2016 and 2017.” Meanwhile, the 5.37 billion euros that Turkey spent between 2012 and 2015 on “the perfect refugee camp[s]” amount to less than 0.2% of GDP per year. Turkey, importantly, is one of the only countries paying the entire costs of short term asylum accommodations out of pocket, “except for some relatively minor international donations” Similar trends have been observed in Lebanon and Jordan, where GDP growth far outpaces the short term costs of accommodating refugees largely due to the fact that—in contrast to Turkey—refugee camps and resettlement programs are funded principally by NGOs and the UNHCR.
Yet even following the short term costs of accommodating refugees, their net fiscal impact over time depends more on the success of their integration into the labor market than their raw consumption of publicly provided goods and services. Joakim Ruist from the University of Gothenburg, for instance, suggeststhat the net fiscal contribution of refugees in Sweden steadily increases from approximately 10,000 kronor (approximately 1,100 USD) during the first year of residence to over 30,000 kronor (approximately 3,300 USD) during their seventh year. In a similar vein, the IMF observed that, depending on the speed of labor market integration, “the level of GDP could be about 0.25 percent higher for the EU as a whole and between 0.5 and 1.1 percent higher in the three main destination countries (Austria, Germany, Sweden)” by 2020.
Importantly, Ruist found that four-fifths of refugees’ net fiscal impact has been estimated to result from their smaller contributions to tax revenue, while only one-fifth was due to “higher per-capita public costs.” In other words, the net fiscal impact of refugees has more to do with their limited contribution to government revenue than their increased demand for public services, suggesting that refugees are more than capable of paying for the public services they consume if successfully and fully integrated into the labor market. This further justifies the need to focus on integrating refugees into the labor market of their destination country, as opposed to simply denying asylum claims based on perceptions that refugees will “steal” domestic jobs.
Furthermore, attempting to estimate refugees’ net fiscal impact based on previous models of migration like Borjas’ wrongly assumes that refugees have a reasonably similar economic profile as other immigrants. In reality, the net fiscal impact of any immigrant varies depending on both the economic profile of the immigrant and the economic conditions of the receiving countries. For example, immigration to Europe between 2007 and 2009 heavily strained public finances because “lots of [immigrants] were pensioners, who tend to drain the public finances,” according to the Economist. By contrast, most refugees fleeing Syria, who “constitute[d] the biggest national group migrating to Europe in 2015,” are both younger and more skilled than those fleeing the last “refugee crisis” in Europe—that of the former Yugoslavia in the 1990s. The median age of Syrian refugees in 2014 was 23 years-old, and compared to the former Yugoslavia, the ratio of “youth cohorts” to “near-retirement-cohorts” has declined from 1.3 to 0.7 since 1990. Moreover, the International Labor Organization reported in 2014 that an average of 56% of Syrian refugees residing in Lebanese camps worked either in skilled or semi-skilled sectors. Therefore, the impact of refugees on the economy overall, and specifically on a country’s public finances, depends on the economic profile of the refugees with particular regard for working-age and job skills.
According to the Pew Research Center, a record 1.3 million people applied for asylum in Europe in 2015, nearly double the previous record of 700,000 set in 1992, and the number of forced migrants across the globe continues to rise. Therefore, understanding refugees’ impact on European economies will become hugely important as more and more are granted asylum and resettled in their new homes. While previous studies of immigration’s impact on the labor market and public finances is somewhat ambiguous, the prevailing economic literature suggests that, in the short run, refugees will have a slightly negative impact on average wages and employment rates in substitutable sectors, and a slightly negative impact on public finances. However, in the long run, if prevailing economic scholarship holds true, wages will stabilize while those forced out of employment by refugees will find work in other sectors, and refugees will yield a net positive fiscal contribution. Finally, considering the fact that refugees hardly function as perfect substitutes for domestic labor, their integration into the labor market would bolster the overall net fiscal expenditure of immigrant-receiving countries.