International migration is a global phenomenon and a debated issue in both political arenas and the media. The recent influx of migrants into EU countries has led to a surge in anti-immigrant sentiment, matched by a dramatic fall in levels of trust in European institutions and a marked increase in voter support for right-wing, anti-immigrant parties. However, when it comes to immigration, anecdotal and academic evidence suggests that individuals are often poorly informed about basic facts related to migrants. For example, a global survey carried out in 2014 by Ipsos MORI shows that public estimates of the share of migrants living in a given country are more than twice the actual figures. A study from 2018 by Alesina and coauthors confirms this bias in natives’ perceptions about both the number and the characteristics of immigrants.
Ensuring citizens have correct information about the size and characteristics of the immigrant population in their country is crucial. Indeed, recent studies show that, if better informed, citizens change their opinion about important policy issues: for example, individuals with more accurate beliefs about the migrant population in their country are more likely to support open immigration policies and higher levels of redistribution. In this context, the media play a key role in informing citizens about immigration and thus in shaping their beliefs and attitudes towards immigrants.
In our paper, we examine the effect of one of the most common sources of information – high-speed Internet – on attitudes toward migrants. The impact of high-speed Internet on attitudes towards migrants is a priori ambiguous. On the one hand, accessing news online allows Internet users to consume information without restrictions and at a reasonable cost. Internet can thus correct potentially biased beliefs about immigrants and their characteristics and improve natives’ opinions about immigrants. On the other hand, access to more information does not automatically translate into greater accuracy of perceptions. The almost unlimited amount of online resources may, in fact, lead to an information overflow. As a reaction, users may concentrate on online material that is concordant with their prior opinions and that could reinforce their original beliefs, thus causing larger polarization. An additional concern is that the rollout of broadband Internet, much like the diffusion of earlier technological innovations, may crowd out traditional and potentially more informative news sources, in the same way that the TV crowded out newspapers.
Our paper empirically examines the causal effect of broadband Internet diffusion on attitudes toward migrants in the context of Spain. To answer our research question, we exploit restricted-access data from a survey run in 1995, 1996, and every year from 2008 to 2012, which was specifically designed to elicit the opinion of Spanish natives about immigration. This unique dataset allows us to understand Spaniards’ perceptions of, and knowledge about, the immigrant population in the country and the nature of their concerns about immigration. We combine this dataset with granular information on broadband Internet diffusion in Spain and data on the characteristics of the voice telecommunication infrastructure in the 1990s. We use the latter to predict future broadband Internet penetration in 2008-2012, thus overcoming concerns related to the fact that the speed of broadband diffusion was influenced by other characteristics (like population density and economic development) that changed around the same time and that are also likely to influence attitudes towards migrants of the resident population. Landline penetration in the 1990s is a good predictor of future broadband diffusion because, in the early phase of its rollout, broadband Internet was supplied through the existing copper lines used to provide telephone services. Therefore, broadband Internet diffusion was easier in municipalities with higher landline penetration due to the lower costs of supplying high-speed connections. Using this predicted value for broadband diffusion in Spanish municipalities, we compare changes in attitudes toward immigrants across local areas, before and after the introduction of high-speed Internet.
Our paper provides three main sets of findings. First, looking at natives’ general attitudes towards migrants, we find that higher broadband Internet penetration leads to a significant increase in the share of respondents thinking that immigration is overall good for their country. This effect is larger for young respondents, who are likely to rely more on the Internet to inform themselves. Additionally, we find that high-speed Internet availability reduces natives’ concerns about the effect of immigration on the labor market and decreases the likelihood of considering immigration a severe problem facing Spanish society. Second, we provide suggestive evidence that higher Internet penetration is associated with better knowledge about the actual share of migrants living in Spain and a reduction in the percentage of respondents that overestimate this figure. Lastly, we find that respondents living in municipalities with higher Internet penetration are less likely to report having voted for the right-wing party Partido Popular. This result is confirmed using data from electoral records of national elections over 1996-2011: municipalities with higher Internet penetration saw a significant decrease in the share of votes cast in favor of PP in the post-Internet period.
Overall, our results suggest that broadband Internet access in its early phase improved the knowledge of native Spaniards of basic facts about immigration. This, in turn, let to more positive public attitudes towards immigrants, and also resulted in lower support for parties promoting stricter anti-immigrant policies. Our findings thus underscore the critical role that broadband Internet plays in shaping political behaviors and opinions. We note that our paper examines the role played by early high-speed Internet exposure on political outcomes and preferences. More specifically, the article focuses on those days when social media were not widespread, and fake news were not as common as today. The impact of different online activities, mobile phones, and social media use, as well as the spreading of fake news on attitudes and political polarization, is a question that is left for future research.