Migration and institutions: Evidence from internal EU mobility

• Date: 01 January 2018
• Published date: 01 January 2018
• Author: Silvia Migali
• DOI: http://doi.org/10.1111/twec.12525

INVITED REVIEW

Migration and institutions: Evidence from internal

EU mobility

Silvia Migali

Fern Universit€

at in Hagen, Hagen, Germany

1

|

INTRODUCTION

The establishment of the Single European Labour Market has guaranteed free labour mobility within

the EU. Despite the removal of the formal impediments to the free movement and the attempt to foster

workers’mobility, other aspects may de facto deter or ease internal migration. One possible factor

could be the institutional setting of the European labour markets. This paper investigates the existence

of the relationship between labour market institutions and internal European migration. Indeed, the

ease of entry into a foreign labour market may depend on the institutional characteristics of the labour

market of both the origin and the destination country. For instance, institutions such as the degree of

employment protection may create barriers of entry for particular kinds of jobs. Moreover, relatively

high minimum wages abroad compared to in the home country may make it difficult for immigrants

to access the foreign labour market. This could be the case in countries with flatter wage structures

and relatively high union power. Besides the labour market institutions, the origin and destination

countries’business cycle situation influences the ease of entry into the foreign labour market.

The paper estimates a panel data gravity model using annual data of bilateral migration flows

of European citizens to EU countries, from 2001 to 2011. Indeed, the formal steps towards the

completion of the Single European Labour Market have been reached in the 2000s. Most impor-

tantly, the chosen time window allows capture of the changes in institutions. The variation of the

institutions observed in the EU labour markets within countries and within macro-regions is

exploited for the empirical analysis.

The evidence on the determinants, the impediments and the motivation of the European migra-

tions is scant (see, for instance, Zaiceva & Zimmermann, 2008). The analysis of the internal Euro-

pean mobility is complex due to the scarcity of disaggregated data on intra-EU immigrants.

However, the heterogeneity of the welfare systems (De Giorgi & Pellizzari, 2013; Giulietti, Guzi,

Kahanec, & Zimmermann, 2013; Skupnik, 2014) and the different arrangements of the housing

markets across European states are commonly recognised as informal factors influencing internal

EU movements (Bonin et al., 2008). Furthermore, the imperfect transferability of pension rights

and the difficulty to harmonise the recognition of academic qualifications are included among the

frictions for the EU mobility (Boeri & Br€

ucker, 2005). Indeed, the cost of the internal mobility

can increase due to the inadequate coordination of the regulation among the EU Member States.

Data limitations also constrain the analysis of labour market institutions. Both the definition of

institutions and the reliability of their measurement are controversial. However, indicators of the

DOI: 10.1111/twec.12525

World Econ. 2018;41:29–58. wileyonlinelibrary.com/journal/twec ©2017 John Wiley & Sons Ltd

|

29

country’s institutional labour market setting and of its rigidity are available and commonly used as

proxies for institutions. While the debate and the assessment of the labour market institutions’indi-

cators go beyond the scope of this paper (for a review, see Eichhorst, Feil, & Braun, 2008), the

difficulty of measuring them and the cautiousness to interpret the empirical results will be high-

lighted in the rest of the paper. In particular, two dimensions of the labour market institutions,

namely institutions related to the wage formation and employment protection legislation, can be

measured and analysed. Among the former, trade union density, commonly used to proxy trade

union power, is introduced in the empirical analysis. Trade union density may influence the wage

formation and the country’s earnings structure. The ease of entry into the foreign labour market

may also be affected by the earnings distribution. Whether this is a possible channel of influence

on migration patterns will also be explored. The other institutions related to the wage formation,

such as the bargaining coverage, the level where collective contracts are negotiated and the degree

of coordination of the bargaining are stable over the time period considered for most of the EU

countries. This constrains their use in the empirical analysis.

The other institutional dimension considered, namely the employment protection legislation, is

expected to affect the job finding probability abroad, hence to influence the migration patterns. As

a proxy for employment protection legislation, an indicator of the degree of protection for tempo-

rary employment forms and contracts is used: when taking the migration decision, individuals may

perceive that they will be employed in a temporary employment relationship once in the destina-

tion country. Hence, an index for the strictness of regulation of the temporary employment forms

is used.

The estimation results indicate that the trade union density tends to influence the migration

patterns. Ceteris paribus, an increase in trade union density in the destination country is associ-

ated with lower migration inflows. In contrast to previous contributions, clear evidence of the

effect of the employment protection legislation on migration flows is not found for the EU.

The remainder of this paper is organised as follows: Section 2 reviews the related literature.

Section 3 briefly describes the intra-EU migration patterns and the institutional setting of the Euro-

pean labour market. Section 4 shows the data, and Section 5 discusses the empirical specification.

Section 6 presents the estimation results, and Section 7 discusses some alternative specifications.

Section 8 concludes.

2

|

LABOUR MARKET INSTITUTIONS AND MIGRATION

While the question about the effects of labour market institutions on immigrants’and natives’

employment outcomes is largely debated (see, for instance, Angrist & Kugler, 2003; Br€

ucker, Jahn,

Hauptmann, & Upward, 2014; D’Amuri & Peri, 2014; Lundborg, 2013; S

a, 2011), less debated is

the question, tackled by this paper, about the effects of labour market institutions in shaping migra-

tion patterns. Gravity-type models using macro-data of migration flows or stocks are suitable to

investigate this issue. For instance, in a study for OECD countries, Cigagna and Sulis (2015) find

a sizeable and positive impact of employment protection of the destination country on the migra-

tion inflows to that destination. They also document a positive effect of union coverage of the des-

tination country on migration movements. The relationship between employment protection and

immigration patterns in OECD countries is contrasting in the literature. Indeed, there is also empir-

ical evidence that immigrants tend to move where the employment protection legislation is similar

to their origin country and that they are not attracted by highly protected labour markets (Bazillier

& Moullan, 2012). While the above-mentioned studies use macro-data to analyse the relationship

30

|

MIGALI