VoxEU Column Migration

Visa policies and multilateral resistance to migration

Migration policy is a pressing issue, but our empirical understanding of it is wanting. This column introduces new estimation techniques for identifying the impact of immigration policies. The novelty is to account for third-country effects since migrants have more options than staying home and moving to the hoped-for destination. Looking at bilateral policies in isolation misses this externality. Disregarding such ‘multilateral resistance to migration’ leads to an underestimation of the effect of bilateral migration policies, and thus potentially leading to severe policy mistakes.

A large number of immigrants to the US do not have a legal residence permit. Most of these entered legally, and simply stayed. A report by the US General Accounting Office suggested that the number of overstayers amounted to approximately 2.3 million in the early 2000s, representing at least one fourth of the total number of illegal aliens in the country.

The flow of immigrants arriving in a destination through these channels depends on the ease of obtaining a visa, which depends on an individual's country of origin. Travel visas can impose substantial costs on visitors, filling out application forms for consular services, paying for processing fees, waiting with insufficient information, and the uncertainty of possible denial (Neumayer 2006). A visa waiver allows travellers to move across borders at substantially lower cost, and with greatly reduced uncertainty. This, in turn, suggests that the bilateral visa regime can also influence the scale of migration flows, as they determine the cost of legal entry.

Six member states (Austria, Belgium, France, Germany, Luxembourg and the Netherlands) have recently urged Cecilia Malmström, the European Commissioner for Home Affairs, to take urgent action against the surge in inflows from five eastern European countries (Albania, Bosnia, Macedonia, Montenegro and Serbia) whose citizens had been granted a visa waiver between 2009 and 2010 (Le Monde 2012).

Bertoli et al. (2011) provide descriptive evidence of the dramatic fall in Ecuadorean immigration following the removal, in August 2003, of the visa waiver that Spain had granted to Ecuador and to other former colonies in Latin America in the 1960s. Average monthly inflows fell from 7,862 in the 12 months before the change in the bilateral visa policy to 1,566 in the following 12 months. This paper represents an exception in the migration literature, which has either not accounted for the influence of visa policies on migration flows or produced non-robust results.

Modelling international migration

The econometric analysis of the determinants of international migration flows is usually conducted within the framework of a pseudo-gravity model, characterised by a long-standing tradition of “estimating bilateral migration flows as a function of characteristics in the source and destination countries only" (Hanson 2010, 4373) in a simple regression. By construction, this framework disregards third-country effects. This tradition can represent a binding constraint for the analysis of the effect of bilateral visa policies, as these are increasingly coordinated at the supranational level, with several destination countries that can simultaneously grant a visa waiver to the citizens of a country. This happened, for instance, in April 2001, when all the 15 signatories of the Schengen Treaty granted a visa-free access to Polish citizens as a step in the process of accession to the EU. The estimated effect of, say, the change in the Spanish visa policy towards Poland on Polish migration to Spain can be confounded by the visa waivers granted by the other 14 European countries, which improve the opportunities that Polish would-be migrants have for entering alternative destinations. If you use an estimation technique that, by construction, does not allow you to control for such a confounding effect, you are likely to obtain biased estimates, in this case biased estimates towards zero.

Our model

In Bertoli and Fernández-Huertas Moraga (2013), we start from a simple random utility maximisation model, i.e., a model in which individuals choose the location that offers them the highest utility. Our model describes the location-decision problem that potential migrants face, and we show that a relaxation of the distributional assumptions usually introduced in the literature creates the need to control for the time-varying attractiveness of alternative destinations to obtain unbiased estimates from a pseudo-gravity model. The confounding influence of the attractiveness of alternative destination countries, which we term multilateral resistance to migration, can be removed with the adoption of an estimator proposed by Pesaran (2006).

We apply this estimation approach to the analysis of the determinants of bilateral migration flows to Spain between 1997 and 2009, using quarterly migration data collected by the National Statistical Office on the basis of the local population registry and an originally built dataset on Spanish visa policies. The estimates based on the (restrictive) long-standing tradition in the literature suggest that visa policies do not significantly influence the size of migration flows to Spain. Still, this result is entirely driven by the bias induced by multilateral resistance to migration. When this is adequately controlled for through the proposed econometric approach, the coefficient of visa policies becomes highly statistically significant, and the data reveal that the introduction of a visa requirement for non-immigrant admission reduces the size of bilateral migration flows to Spain by 74%.

These results have been confirmed in Bertoli and Fernández-Huertas Moraga (2012) with a cross-sectional analysis that relies on migration data from Docquier et al. (2009) and the dataset on bilateral visa policies built by Neumayer (2006). Estimates based on the traditional approach adopted by the migration literature fail to unveil a significant role for visa policies, while a less restrictive estimation approach that controls for multilateral resistance to migration reveals that the introduction of a visa requirement lowers incoming flows between 40 and 47%. We also find that the introduction of a visa requirement by one destination increases flows toward other countries between 3 and 17%; such a sizeable diversion of migration flows signals a policy externality that had not been quantified by the literature, and that justifies the adoption of a supranational coordination in the provisions for non-immigrant admission.


Bertoli, S, Fernández-Huertas Moraga, J, Ortega, F (2011), “Immigration Policies and the Ecuadorian Exodus,” World Bank Economic Review, 25(1), 57-76.

Bertoli, S, Fernández-Huertas Moraga, J (2012), “Visa Policies, Networks and the Cliff at the Border,” IZA Discussion Paper 7094, Bonn.

Bertoli, S, Fernández-Huertas Moraga, J (2013), “Multilateral Resistance to Migration,” Journal of Development Economics, forthcoming, DOI: 10.1016/j.jdeveco.2012.12.001.

Docquier, F, Lowell, B L, Marfouk, A (2009), “A Gendered Assessment of Highly Skilled Emigration," Population and Development Review 35(2), 297-321.

Hanson, G H (2010), “International Migration and the Developing World," in D Rodrik and M Rosenzweig (eds.), Handbook of Development Economics, 4363-4414, Amsterdam, Holland, Volume 5.

Le Monde (2012), “L'afflux de migrants des Balkans préoccupe l'Union européenne”, 24 October.

Neumayer, E (2006), “Unequal access to foreign spaces: how states use visa restrictions to regulate mobility in a globalized world," Transactions of the Institute of British Geographers 31(1), 72-84.

Pesaran, M H (2006), “Estimation and Inference in Large Heterogeneous Panels with a Multifactor Error Structure," Econometrica 74(4), 967-1012.

2,730 Reads