Migration Policy and Welfare State Policy: A New Political Economy Approach

1
Migration Policy and Welfare State PolicyA New
Political Economy Approach

• Assaf Razin
• June 2008
• Based on Razin, Sadka and Swagell (2002), Sand
  and Razin (2008), Cohen (2008), and Suwankiri
  (2008)

2
Talk Plan

• Presenting basic demographic facts
• Theory (political-economy models)
• The effect of immigration over the welfare state
  policy, given an intra-generational conflicts
  (rich vs. poor)
• Welfare state policy and migration policy under
  inter-generational conflict (old vs. young)
• Welfare state policy and migration policy under
  inter-generational and intra-generational
  conflicts
• Empirical Aspects
• The effect of welfare state policy over
  immigration without immigration restrictions
• The effect of welfare state policy over
  immigration with immigration restrictions

3
An increase in the total net-immigration into
OECD countries
Basic Demographic facts
4
?????? ???????? ???????
Basic Demographic facts
5
?????? ????????? ???????
Basic Demographic facts
6
?????? ???????? ???????
Basic Demographic facts

• Immigrants-Emigrants ratio, by education
  attainment
• (stock data, 2000)

• France, Belgium and Germany have the highest
  immigrants-emigrants ratio among the low-skilled
• Sweden and Switzerland has the highest ratio
  among the medium and high skilled group

7
TFR of immigrants in the U.S. compared to the TFR
in their home countries
?????? ???????? ???????
Basic Demographic facts
TFR in The U.S. TFR in Home Country Country
3.51 2.40 Mexico
2.30 3.22 Philippines
2.26 1.70 China
2.23 3.07 India
1.70 2.32 Vietnam
1.57 1.23 Korea
1.79 1.61 Cuba
2.97 2.88 El Salvador
1.86 1.51 Canada
2.84 1.66 United Kingdom
2.86 2.32 Average
Source Center for Immigration Studies, 2002.
8
Working age rate is higher among immigrants than
natives
?????? ???????? ???????
Basic Demographic facts
Source U.S. Census Bureau, Census 2000 Special
Tabulations (STP-159)
9
?????? ???????? ???????
Basic Demographic facts
10
The Effect of Immigration over the Welfare-State
Policy

• Assume that immigration is set exogenously to the
  welfare-state policy within the host country
• On one hand, higher immigration increases the
  political power of the pro-generous welfare state
  coalition
• With intra-generational conflict between the rich
  and poor, the poor-immigrant coalition become
  stronger, since both these groups benefit from
  redistribution
• This effect assumes immigrants have some
  political power, and that they are low skilled,
  as well as the poor natives group
• Of course, often immigrants has no political power

11
The Effect of Immigration over the Welfare-State
Policy

• On the other hand, the entrance of low-skilled
  immigrants increases the burden upon the host
  countrys welfare system
• The marginal benefit of taxation for natives
  becomes lower, as larger portion of their tax
  proceeds is granted to immigrants, who are
  net-beneficiaries of the tax-welfare programs
  (the fiscal leakage effect)
• Thus the natives would respond by lowering taxes,
  reducing thereby the scope of the welfare state
• This effect exists whether the median voter is a
  domestic low-skilled or high-skilled
• This can explain the surprising negative
  correlation between immigration rates and the
  welfare state, within developed countries

12
Welfare State Policy and Migration Policy in an
Inter-generational Model

• What happens if immigration is not exogenous to
  the host country?
• This question arises especially in light of the
  known argument, that immigrants, who are usually
  in working age, have a positive contribution to
  the social security system Simon (1995).
• On the other hand, other researches have
  concluded that this positive contribution to the
  social security system is negligible Lee and
  Miller (2000) Storesletten (2000)

13
Welfare State Policy and Migration Policy in an
Inter-generational Model

• Target
• Examining the influence of immigration over the
  pension system (PAYG), specifically in light of
  population aging
• Namely, how do immigrants effect the conflict of
  interests between old and young regarding
  inter-generational transfers
• Sand and Razin (2008) use a dynamic model to
  explore how do two policy decisions are adopted,
  under median voter mechanism the tax rate which
  funds the PAYG pension system and immigration
  quotas
• The work relies on political-economy models of
  pension systems Cooley and Soares (1999),
  Boldrin and Rusthicini (2000), Froni (2007)

14
Welfare State Policy and Migration Policy in an
Inter-generational Model

• The assumption is that immigrants do not effect
  the fiscal-system only economically, as a working
  force
• They also effect by their ability to vote and
  influence future policy
• Voters are rational thus take this element into
  account, and choose the policy strategically
• Another assumption immigrants fertility rate is
  higher than for the aging native population

15
Welfare State Policy and Migration Policy in an
Inter-generational Model

• Equilibrium
• Immigration policy with restriction is opted for
  by a young median voter
• This strategy replaces the identity of the median
  voter in the following period, from young to old
• Thus, when the current median young voter becomes
  old and in need of pension benefits he will be
  the median voter

16
Welfare State Policy and Migration Policy in an
Inter-generational Model

• Main results
• In comparison to no migration model, in periods
  with young majority, immigration can pose a
  threat to the stability of the pension system.
  This despite the fact that immigration increases
  the total pension payments per worker
• Explanation the young majority is interested in
  an old median voter to support the continuation
  of the pension system in the future, when he
  becomes pension receiver. Young immigrants might
  support cutting down their due pension payments

17
Welfare State Policy and Migration Policy in an
Inter-generational Model

• As expected, population aging increases openness
  to immigrants and reduces the total pension
  payments per worker, since immigration increases
  the total pension payments per worker
• Unexpectedly, despite that the wage of the young
  median voter is adversely affected by the
  increased labor force due to immigration, he is
  not cutting down migration to zero. The voter is
  forward looking to the period when will need
  pension benefits, in which he could use the
  larger labor force he approves today

18
Skilled and unskilled labor
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model
19
Immigration
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model
20
Tax
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model
when
Reflects transfers per worker as a function of
immigration
Tax revenues as a function of immigration
Note higher immigration increases the burden
upon working population
21
Majority voting
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model

• Young skilled are majority
• Young unskilled are majority
• Old are majority

22
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model

• Sincere voting equilibrium, namely, when people
  vote according to their true preferences, without
  strategic considerations

23
Strategies profile in a sub-perfect equilibrium
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model
24
Tax rate chosen by unskilled
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model
25
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model

• Fiscal leakage effect
• More skilled immigrants enable higher
  inter-generational redistribution
• When wages are fixed, more skilled immigrants
  increases inequality, thus the demand for income
  redistribution is higher
• More unskilled immigrants increases the fiscal
  burden, thus the majority of unskilled voters
  reduces current welfare state policy below the
  Lauffer point

26
Wage complementarity between skilled and
unskilled
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model
Skilled is majority
Unskilled is majority
Old is majority
Young is majority, s is small
Skilled is majority, s is big
old is majority
27
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model

• Incentives driving preferences are from two
  channelswage and transfer.
• Transfer channel similar to fixed wage model.
• Wage channel unskilled benefit from
  complementarity with skilled unskilled young
  prefer even more skilled immigrants.
• Due to the wage channel, Skilled young prefer
  unskilled immigrants due to complementarity and
  shun skilled due to substitution.

28
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model

• Skilled young prefer unskilled immigrants
  because of the wage effect, but prefer skilled
  immigrants whose children (inheriting their
  skill) help support the welfare state.
• In addition, young cohort try strategically to
  place its older self as the majority next period.

29
Revised fiscal leakage
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model

• Take immigration quota and immigration
  composition as given.

More unskilled composition of Immigrants create
more desire for redistribution
30
Welfare State Policy and Migration Policy in An
Inter/Intra-generational Model

• Fixed Transfer Level T
• Exogenous wages
• Increasing the skilled/unskilled ratio of
  immigration lower the tax pressure on native
  workers.
• If immigrants net contribution is higher thenper
  capita benefits, then immigrants lower tax
  burden otherwise they raise fiscal burden.
• a fall in native birth rate, put strains on
  financing of T. this raises the tax on working
  young.

31
Empirical Support in the Absence of Immigration
Restrictions

• Immigration policy (other than quotas) can be
  hard to measure
• Nevertheless, in some environments, neglecting
  migration policy as a determinant of migration in
  equilibrium, is reasonable
• For instance, within the EU-155 old member
  states. Since the Paris treaty (1951), gradually,
  including Norway and others by the European
  Economic Area agreement and Switzerland by its
  bilateral treaties with the EU.
• Among all those countries, it is agreed that
  people are free to relocate, immigrate and work,
  with specific equal treatment clause, prohibiting
  discrimination between natives and immigrants in
  the labor market
• In which case, immigration equilibrium is supply
  determined

32
Empirical Support in the Absence of Immigration
Restrictions

• What is the effect of the welfare-state over
  immigration?
• We need to separate unskilled immigrants from
  skilled immigrants
• Unskilled immigrants are drawn to countries with
  generous welfare benefits (the welfare magnet
  effect).
• The reason is obvious welfare state policy
  reflects redistribution of income, benefiting the
  low income individuals

33
Empirical Support in the Absence of Immigration
Restrictions

• There are evidences for welfare migration
• Gramlich and Laren (1984)
• Blank (1988)
• Enchautegui (1997)
• Borjas (1997)
• Meyer (1998)
• And there are evidence to the contrary
• Levine and Zimmerman (1995)
• Walker (1994)

34
Empirical Support in the Absence of Immigration
Restrictions

• High-skilled immigrants, on the other hand, who
  usually enjoy high wages, are deterred by
  generous welfare states, since they are the ones
  who finance it
• Thus welfare state policy induces negative self
  selection among immigrants
• There are no conclusive evidence for that,
  however, in the literature

35
Empirical Support in the Absence of Immigration
Restrictions

• The model
• The sample includes bilateral immigration among
  the U.K, Austria, Belgium, Denmark, France,
  Germany, Italy, Netherlands, Norway, Sweden,
  Switzerland, Finland, Greece, Ireland, Portugal
  and Spain
• The dependent variable the skill-differences of
  the immigrants stock rate in 2000
• We control for per worker GDP in the source and
  host countries, distance, common language and the
  dependent variable in 1990.
• We use lagged average values of 1974-1990 to
  capture aggregate social benefits per capita
• This skill differences model neutralizes the
  effect of omitted variables whose impact on
  migration is skill invariant, as well as time
  invariant variables

36
Empirical Support in the Absence of Immigration
Restrictions

• Welfare benefits per capita has a negative effect
  over the skill composition of immigrants
• This result is robust to different measure of
  welfare

37
Empirical Support in the Absence of Immigration
Restrictions

• This result reoccur also using IV (coverage rate
  of collective bargaining)
• Although, the effect here is more substantial,
  which may suggest a possible upward bias in the
  former regression due to the fiscal leakage effect

38
Empirical Support in the Absence of Immigration
Restrictions

• This result is also based upon a different
  estimation approach, using the micro-economic
  decision to emigrate
• We estimate it using a conditional logit model,
  which enables the individuals to decide whether
  to emigrate and if so, where to, based on the
  possible utilities in each country

39
Empirical Support with Immigration Restrictions

• The challenge estimating policy rules of
  migration and welfare, as a function of
  macro-economic and demographic variables,
  according to the above theory of political
  economy
• This estimation assumes that immigration reflects
  a demand side determined equilibrium, as set by
  quotas, and not according to the choice of the
  immigrants
• This requires simultaneous estimation of both
  policy rules as endogenous