Le blog du CEPII

International migration and the new role of preferential trade agreements

An international organization (uniquely) in charge of managing international migration flows is missing. However recent Preferential Trade Agreements (PTAs) include migration related provisions able to manage migration flows. Further, the signature of a PTA itself boosts bilateral migration flows.
By Gianluca Orefice
 Facts & Figures, July 17, 2012

The managing of labour migration is the responsibility of several international organizations (World Trade Organization, UNHRC, International Labour Office and International Organization for Migration) and it passes through several instruments (Bilateral Migration Agreements, GATS mode IV under multilateral negotiations and PTAs). All these organizations have their own rules concerning migration which often overlap; thus, there is no unique international institution unambiguously managing international migration, and the flows of workers among countries have no proper worldwide governance. Hollifield (2000) and Panizzon (2010) define the former ambiguous framework as the ‘missing regime’ for international labour migration.

This is not the case for international flows of goods which are multilaterally managed by WTO and (more frequently in recent years) bilaterally by Preferential Trade Agreements (PTAs)[2]. The slow development in multilateral negotiations (Doha agenda) is making bilateral negotiations an increasing phenomenon all over the world. The new evidence by Horn et al. (2010) on the contents of PTAs suggested a new possible role for PTAs: filling the missing regime of international migration. In facts, PTAs often include also migration related provisions; hence a new possible link between PTAs and bilateral migration flows is here under the lens. If the new migration related contents of PTAs have a role in affecting bilateral migration flows, PTAs might become a new instrument for governing international flows of migrants.

A first naive evidence on the possible link between PTAs and migration flows is shown in figure 1, where it is clear the contemporaneous increase in both the number of countries having a PTA in force and the bilateral migration flows. This evidence contrasts with the traditional factor contents model of trade which predicts substitutability between free trade and international movement of factors: PTAs by boosting trade in goods imply the equalization of factor prices, reducing the incentive for migration of workers. This does not seem to be the case in figure 1 where freer trade corresponds with more migration flows. However, figure 1 is too naive and the increase in bilateral migration flows could be due to countries without PTAs.  

Figure 1. Increasing trend in migration flows and number of PTAs member countries

Source: WTO and OECD dataset.

Figure 2 shows the same evidence in a more precise way. It shows the average bilateral flows of migrants before and after the entry in force of a PTA by considering only country pairs which have signed a PTA over the period 1998-2008. It is straightforward as after the signature of a PTA the average bilateral flow of migrants has an upward jump.  
Figure 2. Average bilateral migration flows before and after the signature of a PTA (time=t0)
Source: author on OECD data.
Thus it seems that PTAs increase the bilateral migration flows among member countries. Econometric evidence of this link is provided in a recent CEPII working paper (Orefice 2012). But, why PTAs, traditionally thought to boost trade in goods, affect also the international migration flows? What are the channels through which PTAs affect migration flows?

As suggested by the literature on the determinants of migration flows, migration suffers two kind of costs: monetary and non monetary costs. The monetary cost of migration is valuable in terms of money and it relates mainly with travel costs (it stays outside the purpose of this short essay). Non-monetary cost of migration relates with all other sources of costs affecting the decision of migrate which are not valuable in terms of money: (i) difficulties in obtaining documents to migrate into a certain country (which is a “cost” and depends on both origin and destination country); (ii) how easy is for a new migrant to be integrated in the destination country (given the country of origin), it depends on the cultural proximity of the origin-destination couple.

Horn et al. (2010) show that PTAs also contain migration related measures such as visa-and-asylum, labour market and GATS related measures; former provisions potentially affect the non-monetary cost of migration and thus migration flows among PTAs’ member countries. Visa-and-asylum provision, by making the procedures for obtaining a permit to stay in the destination country easier, might reduce the non-monetary cost of migration among PTAs member countries. Labour market related provisions point to harmonize labour markets in all signatory countries and could favour migration into member countries by making easier the procedure for obtaining a job in the destination country. Finally, GATS replicating provisions by allowing free temporary migration among PTAs members allow new migrants to join labour market (and business networks) in destination countries and makes more likely future long term stay in that country. For example, the Australia-Singapore agreement (2003) includes provisions on visa-and-asylum and GATS, but not on labour market harmonization. Differently, the US-Chile agreement (2004) contains labour market integration and GATS provisions but not visa-and-asylum one [3]. Recent results in Orefice (2012) show that the inclusion on visa-and-asylum and labour migration provisions in PTAs stimulates bilateral migration flows, while GATS replicating provision has a negative effect on bilateral migration flows.[4]

Thus the content of PTAs is a first piece of intuition driving the positive link between PTAs and migration flows. The content of PTAs directly affects the non-monetary cost of migration and may well be used as an instrument for “shadow” migration policy. “Shadow” because migration related provisions are silently included in trade related bilateral agreements, which mainly contain trade in goods related provisions.

On top of the former channel, the signature of a PTA itself might affect migration flows (no matter its contents). Being constant all the other determinants of migration, potential migrant will chose the destination country with the lower information cost (i.e. the one he or she knows better); PTAs by increasing information (familiarity) concerning member countries, might drive potential migration flows towards PTA’s member countries rather than other countries. Econometric analysis in Orefice (2012) shows that having a PTA in common stimulates bilateral migration flows by 17.5%.

Descriptive evidence showed before (together with econometric results in Orefice 2012) supports the idea that PTAs might be used by government to manage bilateral migration flows. But, why this could represent an opportunity for government? And how it places in solving the missing regime for international migration? Wide empirical evidence shows negative attitudes towards immigrants in developed countries among voters (see Mayda 2008), who, in general, are more pro-trade than pro-migration; this makes PTAs easier to put in place than bilateral migration agreements in case of labour shortage (and or population ageing). Finally, managing international migration through bilateral trade agreements does not solve the missing multilateral regime for international migration, but surely it patches up the lack in international government of migration through bilateral agreements.
References : 

This column is strongly based on a recent CEPII working paper, Orefice (2012), to be published.

Hollifield, J.F. (2000) “Migration and the 'New' International order: the missing regime”, in Ghosh., B., (Ed) Managing Migration, Time for a New International Regime. Oxford University Press.

Horn, H., Mavroidis, P.C., Sapir, A. (2010) “Beyond the WTO? An Anatomy of EU and US Preferential Trade Agreements”, The World Economy. 33, 1565-1588.

Mayda, A.M. (2008) “Why are people more pro-trade than pro-migration?”, in Economics Letters. 101, 160-163.

Orefice, G., Rocha, N. (2011) “Deep integration and production networks: an empirical analysis”, in University of Milan Departmental Working paper series Panizzon (2010)

World Trade Report (2011) “The WTO and preferential trade agreements: from co-existance to coherence”, WTO Geneva.

[1] By PTAs I refer here to bilateral (or pluri-lateral, RTAs) agreements on trade which guarantee to member countries easier market access than MFN regime under the WTO.
[2] Tables A1, A2 and A3 in Orefice (2012), show the lists of agreements containing respectively provisions on visa-asylum, labour market and GATS (General Agreement on Trade in Services). These lists are based on a dataset provided by the WTO as an extension of Horn et al. (2010). Further details are reported in Orefice and Rocha (2011). Dataset available here: http://www.wto.org/english/res_e/publications_e/wtr11_dataset_e.htm.
[3] The variable used for the presence of GATS provision keeps also the effect of mode 3 service liberalization, thus the negative coefficient could be due to the substitutability between FDI and migration flows.
Migrations  | Trade & Globalization 
< Back