Political Economy of Immigration Policy: Lecture 6

1
Political Economy of Immigration PolicyLecture 6

• Gordon H. Hanson
• UCSD and NBER

2
Introduction

• What weve done so far
• Scale, selection and sorting in international
  migration
• Labor markets impacts of emigration and
  immigration
• Public finances and international labor flows
• Illegal migration
• Final lecture
• Political economy of immigration policy
• Why do receiving countries restrict labor inflows?

3
Immigration Policy Regimes

• Governance of immigration policy
• Regulation of entry
• By quantity (quota on visas) or price (entry fee)
• By ex ante or ex post selection criterion
• Scope of residency rights
• Broad (path to citizenship) or narrow (temporary
  admission)
• Immigration policy regimes
• Legal immigration
• Combination of quantity regulation (quotas), ex
  ante selection criteria (family, skills,
  asylees), broad residency rights
• Illegal immigrants
• Combination of price regulation (implicit entry
  fee), ex post criterion (behavior), minimal
  residency rights

4
Political Economy of Immigration Policy

• Why do countries restrict labor inflows?
• Protect income of median voter or specific
  factors
• Benhabib, de Melo et al
• Facchini and Willmann, Scheve Slaughter, Hanson
  Spilimbergo, Mayda, Facchini and Mayda
• Address distortions due to welfare state
• Wellisch Walz, Sinn et al, Boeri and Brücker,
  Facchini Mayda, Hanson, Scheve Slaughter
• Scholten Thum, Razin Sadka, Poutvaara
• Preserve cultural or ethnic homogeneity
• Huntington, Dustmann and Preston

5
Design of Immigration Policy Regimes

• Actual policy regimes have at least three
  dimensions
• (1) entry regulation, (2) selection criterion,
  (3) scope of residency rights
• Basic tensions in policy design
• National welfare gain from free immigration vs.
  loss in income of favored groups (focus of
  literature to date)
• Providing incentive to assimilate versus
  obtaining information on the desirability of an
  immigrant
• Do bilateral migration treaties address these
  issues?

6
Why are poor consigned to illegal migration?

• Whom do rich countries admit as legal immigrants?
  
• Where do rich countries govern entry through
  quantity regulation, ex ante selection, broad
  rights?
• Use for immigrants whose type (eg, skill, family)
  is observable
• Ex ante screen broad rights strong incentive
  to assimilate
• Whom do rich countries admit as illegal
  immigrants?
• Where do rich countries govern entry thru price
  regulation, ex post selection, narrow residency
  rts?
• Use where type (eg, motivation) is unobservable
• Mix of entry price, ex post screen, narrow rights
  selects desirable types (but reduces incentive to
  assimilate)
• Narrow rights also lower fiscal cost, softening
  opposition

7
Trade versus immigration policy

• Many countries appear to have adopted relatively
  liberal trade policies but relatively illiberal
  immigration policies
• This pattern is replicated in public opinion data
  with support for liberal trade policies being
  stronger than support for liberal immigration
• Why does freer immigration seem to be less
  politically palatable that freer trade?

8
Some Answers

• Non-economic concerns such as attitudes toward
  foreign cultures and minority groups are more
  influential for immigration than trade
• Pat Buchanan, Samuel Huntington (2005)
• Economic concerns about immigration and trade
  differ and may favor trade over immigration as a
  strategy for international economic integration

9
Foreign-born share of the US population is rising
Immigrants in the US, Number and Share of
Population
Millions
Percent of Population
Year
10
Education levels of US immigrants, natives
11
Concerns about immigration

• In the US, immigrants are concentrated at the
  extremes of the skill distribution
• Strongly over-represented among the least
  educated
• Moderately over-represented among the most
  educated
• Economic consequences of immigration
• Changes relative supplies of labor
• Expands US tax base, increases demands on public
  services
• Expands cross-border flows of ideas, capital,
  trade
• Are consequences of immigration reflected in
  individual attitudes toward immigration policy?

12
The argument

• In the absence of distortionary tax and spending
  policies, free immigration and free trade would
  often be supported by similar groups thanks to
  similar labor-market effects
• Either tends to depress relative wages for
  low-skilled US natives
• Borjas, Freeman Katz (1997), Feenstra Hanson
  (1999), Scheve Slaughter (2001), Borjas (2003),
  Mayda (2004), Card (2005)
• However, government policies that redistribute
  income alter the distributional consequences (and
  so the politics) of globalization
• Immigrants pay taxes, use public services, and
  vote imports do none of these things
• The net fiscal impact of immigration appears
  negative (at least in short run), especially in
  high-immigration states and for high-income
  citizens who pay most of the tax bill for public
  services to immigrants
• Smith et al. (1997) 200 net transfer per HH in
  US, 1200 in CA
• Smaller fiscal effects are expected as a result
  of international trade

13
Views on fiscal consequences of immigration

• "California's addiction to 'cheap' illegal alien
  labor is bankrupting the state and posing huge
  burdens on the state's middle class tax base."
  Dan Stein, Federation for American
  Immigration Reform
• It uses tax dollars to provide illegal workers
  with educational services, medical care, and
  unemployment compensation for those that are
  injured on the job. They are a drain on the
  taxpayers of Canyon County, the state of Idaho,
  and the US in general. Robert Vasquez,
  Commissioner of Canyon County, Idaho
• "The bottom line is that businesses are
  benefiting from the cheap illegal labor. They're
  making extra profits and they're shifting the
  whole benefits package over on to the
  taxpayer. Rick Oltman, Yuma County,
  Arizona
• Federal immigration policies should ensure that
  new immigrants do not become a public charge to
  federal, state, or local governments.
  National Governors Association

14
A tale of two governors

• In the 1990s, California and Texas both had
  Republican governors
• Governors Bush and Wilson were fiscal
  conservatives and rising stars within the
  national Republican party
• Both supported NAFTA and free trade, but took
  very different approaches toward immigration
• Bush courted the Latino vote (even campaigning in
  Spanish) and tried to appear pro-immigration
• Wilson endorsed Proposition 187, a ballot measure
  to deny access of illegal immigrants to public
  services in California (later blocked by the
  courts)

15
A tale of two governors

• What accounts for these different approaches
  toward immigration?
• California provides generous public benefits to
  immigrants financed by progressive income taxes
• This may have given high-income voters in the
  state an incentive to join the anti-immigration
  coalition
• Texas provides meager public benefits financed by
  a regressive state sales tax
• This may have made high-income voters in the
  state more pro-immigrant and created political
  space to attract socially conservative Latino
  immigrants to the Republican party

16
A simple model of policy preferences

• Let V(p, Ii) be indirect utility received by
  individual i, as a function of commodity prices,
  p, and after-tax income, Ii
• Consider change in indirect utility due to
  immigration
• Expressed in monetary terms this is
• Immigration changes commodity prices (1st term),
  pre-tax income (2nd term), and net fiscal
  transfer (3rd term)

17
Welfare effects of immigration

• Let V(p, yi) be indirect utility received by
  individual i, as a function of commodity prices,
  p, and income, yi
• Change in aggregate welfare due to change in low
  skilled labor supply of dL is (where ai is share
  of group i in the pop. for two groups, skilled
  (H) and unskilled (L))
• If marginal utility of income (?) is constant
  across people
• where

18
Welfare effects of immigration

• Change in welfare as share of GNP (Y) is
• where is the elasticity of income for
  factor type i w.r.t. the supply of low skilled
  labor, ?i is the share of factor i in GNP, and M
  is inflow of L
• y is post-tax income, given by
• where wi is the pre-tax wage for group i, ti is
  the tax rate for group i and gi are govt
  transfers for group i

19
Welfare effects of immigration

• How does post tax income respond to immigration?
• Elasticity of post tax income w.r.t. supply of
  factor i is
• where ?v,I is share of income type v for factor i
  and ?v,iL is the elasticity of income type v
  w.r.t. the supply of low skilled labor for factor
  i
• We expect
• ?w,LL lt 0, ?w,HL gt 0 (immigration of L raises wH
  and lowers wL)
• ?t,LL gt 0, ?t,HL gt 0 (immigration of L raises
  tax rates)
• ?g,LL lt 0, ?g,HL gt 0 (immigration of L lowers
  govt transfers)

20
Estimating equation

• Assume pre-tax income and the net fiscal transfer
  are linear functions of individual
  characteristics
• (where X is defined such that ßgt0)
• Assume ?ß/?M?, in which case we expect
• With low-skilled immigration
• ?yi/?MXi??ei/?M gt 0 for high-skilled labor
• ?yi/?MXi??ei/?M lt 0 for low-skilled labor
• Assume ??/?M?
• ? will vary across US states according to their
  spending and tax policies and the size of their
  immigrant populations

21
Estimating equation

• Putting this together, we obtain
• Preferences on immigration policy depend on
  individual characteristics, state
  characteristics, and their interactions (use
  categorical variables to allow for
  non-linearities)
• All else equal,
• Overall, the low-skilled will be more opposed to
  immigration (than the high-skilled) and their
  opposition will be stronger in states with larger
  immigrant populations
• The opposition of the high-skilled to immigration
  will be stronger in states that (a) make larger
  fiscal transfers to immigrants, or (b) have
  more-skilled immigrant populations
• Policies that weaken redistribution will also
  weaken high-skilled opposition to immigration

22
Identification

• Empirical estimation
• We estimate the correlation between opposition to
  immigration (or trade) and schooling and how this
  varies with size of state immigrant population,
  skill mix of state immigrant population, and
  state fiscal exposure to immigration
• Given controls for state fixed effects, our focus
  is on how slope coefficient on schooling varies
  across states identification is based on
  cross-state variation in how opposition to
  immigration responds to schooling (and whether
  this is conditional on the fiscal regime of the
  state)

23
Identification

• Estimation issues (and partial corrections)
• Endogeneity of state of residence, state welfare
  policies
• Issue is possible correlation between unobserved
  determinants of individual attitudes toward
  immigration and size of state immigrant
  population (or nature of state welfare policies)
• Would likely cause us to understate cross-state
  variation in opposition to immigration (since
  those more opposed to immigration or to generous
  welfare policies would be more likely to settle
  in low-immigration, low-welfare states)
• Over sample period, few individuals change states
  and ranking of states in terms of welfare
  generosity to immigrants is stable issue may be
  ameliorated by controlling for state fixed
  effects
• Measurement error in preferences (Bertrand
  Mullainathan)
• Issue is that measurement error in stated
  preferences may be correlated with individual
  characteristics (eg, education)
• Exploit information on other opinions (including
  those seemingly irrelevant to immigration) to
  absorb idiosyncratic error terms

24
Data

• Examine public opinion across US states, for
  which there there is variation in fiscal costs of
  immigration
• Define US states as tax jurisdiction (states set
  income, sales, and property taxes, provide many
  public benefits)
• Sources
• National Election Studies Surveys, 1992 and 2000
• Census of Population and Housing, Census of
  Governments, CPS

25
Immigrants are geographically concentrated
26
But recent immigrants follow job growth
27
Fiscal costs of immigration vary by state
28
Since 1996, immigrant-native differences in
welfare use have declined for all categories
except Medicaid
29
Differences remain in welfare uptake across US
states ( of households receiving
assistance by type)

A. Overall Welfare Participation Rates A. Overall Welfare Participation Rates A. Overall Welfare Participation Rates A. Overall Welfare Participation Rates A. Overall Welfare Participation Rates A. Overall Welfare Participation Rates A. Overall Welfare Participation Rates B. Cash Program Participation Rates B. Cash Program Participation Rates B. Cash Program Participation Rates B. Cash Program Participation Rates
  Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State
  Less Less More Less Less More Less More Less More
  Natives Natives Natives Immigrants Immigrants Immigrants Natives Natives Immigrants Immigrants
1994 16.7 16.7 13.9 22.8 22.8 25.2 7.7 7.6 8.2 14.3
1998 13.9 13.9 12.6 16.9 16.9 21.9 5.7 6.0 5.6 10.5
2002 16.0 16.0 13.7 20.7 20.7 25.5 4.9 4.9 3.4 8.3
C. Medicaid Participation Rates C. Medicaid Participation Rates C. Medicaid Participation Rates C. Medicaid Participation Rates C. Medicaid Participation Rates C. Medicaid Participation Rates C. Medicaid Participation Rates D. Food Stamp Participation Rates D. Food Stamp Participation Rates D. Food Stamp Participation Rates D. Food Stamp Participation Rates
  Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State Welfare Generosity of State
  Less Less More Less Less More Less More Less More
  Natives Natives Natives Immigrants Immigrants Immigrants Natives Natives Immigrants Immigrants
1994 14.1 14.1 12.4 18.6 18.6 23.6 9.6 7.3 14.2 13.6
1998 12.2 12.2 11.6 15.4 15.4 20.7 6.4 5.2 5.7 9.1
2002 14.5 14.5 12.8 19.4 19.4 24.7 6.3 4.6 5.8 6.8
30
Percent of school children with immigrant
mothers, 2000
31
Expenditures and revenues by foreign-born and
native households in California (1996)
32
State generosity in welfare benefits
           
  4 IL MO, NE CA, MA, MD, ME, RI WA
3 FL OR CT, MN, NJ, NY, PA, WI HI
Public Assistance Availability to Immigrants 2 DC, DE, KY, MT, NV, NC, TN, VA, WY AK, AZ, CO, GA, IA, KA, NM, ND, UT MI NH
  1 AL, AR, ID, LA, MS, OH, TX, WV IN, OK, SC, SD    
    1 2 3 4
    Public Assistance Generosity Public Assistance Generosity Public Assistance Generosity Public Assistance Generosity
33
Ranking of US states in terms of generosity
toward immigrants is stable over time
34
Data

• Measuring fiscal exposure to immigration (start
  with 4 categories per measure, test down to 2)
• Immigration fiscal exposure measures
• Fiscal Exposure 1 Equal to 1 if state welfare
  spending per native is above 90 national mean
  and if high-immigration state (immigration
  population share is above 90 national mean), 0
  o/w
• Fiscal Exposure 2 Equal to 1 if ratio of state
  immigrant HHs on welfare to number of native HHs
  above 90 national mean, 0 o/w
• Other fiscal exposure measures (for trade policy
  analysis)
• Fiscal Exposure 3 Equal to 1 if state welfare
  spending per native is above 90 national mean
• Fiscal Exposure 4 Equal to 1 if state welfare
  spending as share of state GDP is above 90
  national mean
• Interactions with schooling (No High School, High
  School, Some College, College) or income quartile

35
High immigration and generous welfare states

• High immigration states
• 1992 AZ, CA, CT, FL, IL, MA, NJ, NY, TX
• 2000 92 states plus CO, DC, MD, NV, NM, OR,
  VA, WA
• States with high immigrant uptake of welfare
• 1992 CA, CT, FL, MA, NJ, NY
• 2000 92 states plus WA
• States with high welfare spending
• Per native household
• 1992 CA, CT, NH, IL, LA, MD, MA, MI, MN, NJ,
  NY, OH, PA, TN, WA, WV, WI
• 2000 CA, CT, DC, IL, KY, ME, MD, MA, MN, MS,
  NH, NJ, NM, NY, ND, OR, PA, SC, TN, WA, WV
• As share of state GDP
• 1992 AZ, AK, CA, CT, LA, MA, MI, MN, NH, NJ,
  NY, OH, PA, TN, WV, WI
• 2000 AL, AK, CA, IA, KY, LA, ME, MA, MN, MS,
  NM, NY, ND, OH, OR, PA, SC, TN, WV

36
Data

• NES immigration policy question
• Do you think the number of immigrants from
  foreign countries who are permitted to come to
  the US to live should be increased a little,
  increased a lot, decreased a little, decreased a
  lot, or left the same as it is now?
• Immigration Opinion is coded a 1 for those
  individuals favoring immigration be decreased a
  little or a lot and 0 for those individuals
  favoring immigration be maintained or increased.

37
Data

• NES trade policy question
• Some people have suggested placing new limits on
  foreign imports in order to protect American
  jobs. Others say that such limits would raise
  consumer prices and hurt American exports. Do
  you favor or oppose placing new limits on
  imports, or haven't you thought much about this?
• Trade Opinion is coded a 1 for those individuals
  favoring protection and 0 for those opposing it

38
Native support for decreased immigration by
education and fiscal exposure to immigrants, 2000
Fiscal Exposure Measure Fiscal Exposure No High School High School Some College College Grad

Fiscal Exposure 1 Low 0.537 0.571 0.507 0.362
High 0.625 0.558 0.461 0.420

Fiscal Exposure 2 Low 0.498 0.572 0.516 0.361
  High 0.658 0.557 0.464 0.435
39
Native support for trade restrictions by
education and fiscal exposure to immigrants, 2000
Fiscal Exposure Measure Fiscal Exposure No High School High School Some College College Grad

Fiscal Exposure 1 Low 0.748 0.719 0.590 0.383
High 0.662 0.674 0.560 0.418

Fiscal Exposure 2 Low 0.728 0.719 0.606 0.392
  High 0.707 0.674 0.548 0.397
40
Probit results on native opposition to immigration

• Estimation
• Sample is native respondents
• Regressors
• Dummies for schooling category alone and
  interacted with size of state immigrant
  population, state fiscal exposure to immigration
• Other controls gender, quadratic in age, race,
  ethnicity, state unemployment rate, state dummy
  variables, year dummy variable
• Additional controls Isolationist sentiment,
  union membership, public sector employment,
  partisanship, ideology
• Standard errors clustered by state

41
Table 2 Fiscal Exposure 1 Fiscal Exposure 1 Fiscal Exposure 2 Fiscal Exposure 2
High School 0.182 0.168 0.211 0.206
(0.101) (0.107) (0.095) (0.103)
Some College 0.047 0.027 0.102 0.092
(0.097) (0.098) (0.100) (0.104)
College -0.347 -0.356 -0.298 -0.295
(0.119) (0.130) (0.116) (0.127)
No High School Immigration 0.445 0.474 0.365 0.667
(0.141) (0.173) (0.257) (0.270)
High School Immigration -0.016 0.034 -0.012 0.248
(0.117) (0.136) (0.122) (0.140)
Some College Immigration -0.098 -0.033 -0.143 0.116
(0.094) (0.097) (0.088) (0.151)
College Immigration -0.078 -0.016 -0.183 0.043
(0.111) (0.128) (0.115) (0.165)
No High School Fiscal Exposure -0.090 0.246 0.102 0.189
(0.200) (0.169) (0.264) (0.259)
High School Fiscal Exposure -0.008 0.303 -0.011 0.144
(0.141) (0.130) (0.137) (0.136)
Some College Fiscal Exposure -0.036 0.237 -0.003 0.138
(0.089) (0.109) (0.082) (0.110)
College Fiscal Exposure 0.219 0.490 0.347 0.526
(0.118) (0.145) (0.127) (0.117)
State Fixed Effects No Yes No Yes
42
Impact of education on attitude toward
immigration (Facchini Mayda, 2006)
43
Impact of income on attitude toward immigration
(Facchini Mayda, 2006)
44
Trade policy preferences
Fiscal Exposure Measure Fiscal Exposure Measure Fiscal Exposure Measure Fiscal Exposure Measure
Table 5 1 2 3 4
High School -0.111 -0.020 -0.074 -0.033
(0.121) (0.123) (0.151) (0.161)
Some College -0.390 -0.267 -0.268 -0.263
(0.133) (0.146) (0.138) (0.143)
College -0.971 -0.849 -0.908 -0.946
(0.132) (0.128) (0.163) (0.143)
No High School Fiscal Exposure -0.367 0.232 0.004 0.245
(0.264) (0.360) (0.306) (0.275)
High School Fiscal Exposure -0.228 0.132 -0.021 0.160
(0.132) (0.196) (0.238) (0.189)
Some College Fiscal Exposure -0.221 0.111 -0.173 0.064
(0.148) (0.188) (0.236) (0.179)
College Fiscal Exposure -0.046 0.223 0.045 0.376
(0.161) (0.193) (0.259) (0.164)
45
Additional results

• Results are robust to
• Including interactions between education and
  skill mix of state-immigrant population (measured
  continuously or discretely)
• Including state share of Latinos in immigrant
  population, individual attitudes toward
  minorities, other individual opinions
• Substituting schooling with reported income,
  measured by quartile
• Dropping individual high-immigration states (CA,
  NY, FL, TX, IL, NJ)
• Dropping those who self identify as very
  conservative or very liberal
• Using continuous measures of immigrant
  population, fiscal exposure
• Re-estimating using ordered probit instead of
  bivariate probit

46
Summary

• Puzzle of why we tend to see free trade but
  restrictionist immigration policy is usually
  posed as if their distributional politics are the
  same
• While this may be true before taxes, transfers,
  it is unlikely to be true after taxes, transfers
• The distributional consequences and thus politics
  of immigration and trade appear to differ