Immigration, Conflict, and Redistribution

• Author: Ángel Solano García,Santiago Sánchez‐Pagés
• DOI: http://doi.org/10.1111/sjoe.12158
• Date: 01 July 2016
• Published date: 01 July 2016

Scand. J. of Economics 118(3), 557–593, 2016

DOI: 10.1111/sjoe.12158

Immigration, Conflict, and Redistribution∗

Santiago S´

anchez-Pag´

es

University of Barcelona, 08034 Barcelona, Spain

sanchez-pages@gmail.com

´

Angel Solano Garc´

ıa

University of Granada, 18011 Granada, Spain

asolano@ugr.es

Abstract

In this paper, we analyze how the possibility of conflict between natives and immigrants

shapes income redistribution in developed democracies. This possibility can generate in-

come redistribution towards immigrants even if they have no voting rights. We show that

the threat of conflict between natives and immigrants lowers vertical income redistribution

(from the rich to the poor) as the level of immigration increases. The opposite holds for

horizontal income redistribution (from natives to immigrants), which increases with the level

of immigration. Income inequality weakens the negative effect of immigration on vertical

redistribution, but it also reduces horizontal redistribution. These theoretical predictions are

consistent with the results of our empirical analysis on data from 29 European countries:

larger immigrant populations are associated with more redistribution towards immigrants and

lower vertical redistribution.

Keywords: Conflict; immigrants; income redistribution; inequality; natives

JEL classification:D72; D74; F22

I. Introduction

Immigration has become an extremely important issue in many devel-

oped countries. Citizens are very concerned about the consequences of the

large-scale migration that has followed the globalization process. Two main

economic concerns shape natives’ views on immigration. On the one hand,

increasing numbers of immigrants allegedly push down wages in unskilled

jobs and create unemployment among natives. Empirical evidence on this

∗We thank audiences at Alicante, Cardiff, Carlos III Madrid, Strathclyde, SAET 2011, and

the Priorat Workshop on Theoretical Political Science for their valuable comments and

suggestions. We are especially grateful to Elena B´

arcena and to two anonymous referees for

their great help in improving the paper. Solano Garc´

ıa acknowledges financial support from

the Junta de Andalucia, grant number SEJ 0361 (2008-2010).

CThe editors of The Scandinavian Journal of Economics 2015.

558 Immigration, conflict, and redistribution

effect is inconclusive.1The second concern is that immigrants might rely

heavily on welfare benefits, given that they are mostly unskilled and have

lower incomes. The available empirical evidence on this is much more

conclusive, with a broad variety of studies suggesting that low-skilled im-

migrants represent a drain on fiscal coffers (Edmonston and Smith, 1997;

Storesletten, 2000).2

In this paper, we are the first to explore the idea that income redistribu-

tion towards immigrants takes place because of the possibility of a conflict

between natives and immigrants. We show that this mechanism can explain

the positive levels of income redistribution towards immigrants observed

in western democracies. The results of the model are also consistent with

the so-called “anti-solidarity effect” (Lee et al., 2006), that is, the nega-

tive relation between income inequality and income redistribution towards

immigrants observed in developed nations.3

The possibility of conflict between natives and immigrants is very real.

Tensions between the two communities are commonplace. Although this

unrest remains at a low intensity most of the time, several race-related

protests and riots have erupted in the last decade in western countries.4

Our point is that at the heart of all this social unrest, there is compe-

tition between natives and immigrants to capture bigger shares of public

resources.

We build a simple model in order to explore the role of conflict in

the relationship between immigration and redistribution. In this model,

there exists a conflict of interests between natives and immigrants over

public funds that are used to pay group-specific lump-sum transfers. This

conflict of interests is resolved through confrontation: immigrants spark

conflict whenever they can improve their situation by resorting to conflict.

At the same time, there exists a conflict of interest within the native

population; a conflict that is resolved through income taxation. Hence,

taxation constitutes an instrument of vertical redistribution from the rich

to the poor, whereas the share of public funds devoted to immigrants

is an instrument of horizontal redistribution. Both the implemented tax

rate and the share of tax revenues devoted to funding the group-specific

1See, for example, Borjas (2003) and Card and Shleifer (2009) for two different results.

2However, from a general equilibrium analysis, the overall effect of immigration on the

government budget is not so clear. This is because the benefit of capital owners and skilled

workers might be larger than the cost of unskilled workers in terms of income. This would

lead to an increase in tax revenue.

3See McCarthy et al. (2006), Soroka et al. (2006), and Razin et al. (2002) for empirical

evidence on this effect.

4Some examples are the riots that took place in Northern England in 2001, Brussels in

2006, Denmark in 2008, and Italy in 2010.

CThe editors of The Scandinavian Journal of Economics 2015.

S. S´

anchez-Pag´

es and ´

A. Solano Garc´

ıa 559

transfers are chosen by majority vote. We assume that poor natives are

in the majority, so the policy finally implemented is optimal for them.

In addition, immigrants cannot vote and their only chance of altering the

policy outcome is by contesting the share of public funds they receive.

In this simple set-up, the threat of conflict shapes policy outcomes. Im-

migrants always receive a positive share of public expenditures: the mini-

mum share of public funds that avoids conflict. Our main result, though,

is that this threat makes the intensity of vertical redistribution decrease

with the level of immigration. Two effects are at work here. An increase

in the stock of immigrants makes natives prefer higher tax rates because

of the now bigger tax base. We call this the tax base effect. However, as

the immigrant group becomes larger, immigrants become more effective in

the event of confrontation, and this raises the share of public expenditures

they can obtain. This makes natives prefer lower levels of public spending

and, consequently, reduces the levels of income redistribution. We refer to

this as the conflict effect. When the stock of immigrants is sufficiently

large, the latter effect dominates and the majority of natives prefer less

redistribution. This result is thus consistent with the empirical evidence on

the anti-solidarity effect.

Razin et al. (2002) and Lee et al. (2006) also found that redistribution

decreases with the size of the immigrant population, but the mechanisms

they explored are different from ours. Razin et al. (2002) assumed that

immigrants are entitled to benefits in the host country. This exogenous

leakage induces natives to vote for less redistribution. In contrast, our model

endogenizes immigrants’ access to benefits. Lee et al. (2006) proposed

a model of political competition where natives are xenophobic towards

immigrants. This makes poorer natives vote for lower public spending. In

our model, we do not resort to a primordial distaste towards immigrants in

order to obtain this result.5

Our second main result is that the negative effect of immigration on

redistribution becomes weaker with income inequality. As the income gap

increases, poor natives demand higher levels of vertical redistribution. More

inequality makes natives, who are on average richer than immigrants, enjoy

“deeper pockets” so they can mobilize more resources in the event of

conflict. This makes the conflict effect relatively less important. As a result,

tax rates are higher and the degree of horizontal redistribution decreases.

We next show that the welfare of both rich and poor natives increases with

the stock of immigrants. The reason for this is that the resulting increase

5Roemer et al. (2007) discuss the existence of a policy-bundle effect, which allows conserva-

tive parties to obtain considerable support from poor and working-class natives by proposing

low redistribution and tough stances on the immigration issue.

CThe editors of The Scandinavian Journal of Economics 2015.