The Benefits of Immigration

Report
The Benefits of Immigration: Some
Implications of Recent Findings
Ethan Lewis
Dartmouth College
The Standard Model
• Gains to immigrants themselves are huge
– Large cross-country income differences persist
– The world would benefit from a much freer
immigration system (e.g., Hamilton and Whalley
1984, Kennan 2013)
• Gains to receiving country are comparatively
small, and contingent on much larger
distributional impacts (e.g., Borjas, 1999)
– Typical estimate of net benefit: Well <1% of GDP.
The Standard Model
• Immigration produces benefits by tilting the wage
structure:
– A country gains the most from immigration when it
admits immigrants with “scarce” skills: those skills
which are rarest in the existing population
– This lowers the wage of those scarce skill, but raises
the wage of everyone else (the commoner skill) and
on average natives benefit
• There are winners and losers
– A skill-balanced inflow produces no long-run benefits
• Perhaps a (very) short run benefit for native “capital owners”
United States, 2000
100
Percent College and Non-College, 1990s Immigrants and Existing Workers
Existing Workers
60
80
1990s Immigrants
55.4
52.0
44.6
0
20
40
48.0
Non-College
Data Source: Docquier, Ozden, and Peri (2010)
College
United States, 2000
50
Percent at Five Education Levels among 1990s Immigrants and Existing Workers
1990s Immigrants
Existing Workers
40
42.0
20
30
30.5
28.0
17.8
19.3
16.3
15.7
13.0
8.0
0
10
9.5
Dropout
HS Grad
Data Source: 2000 Census of Population
1-3 Yrs Coll
4 yrs Coll
Adv. Degree
United Kingdom, 2000
100
Percent College and Non-College, 1990s Immigrants and Existing Workers, 2000
1990s Immigrants
84.2
18.7
15.8
0
20
40
60
80
81.3
Existing Workers
Non-College
Data Source: Docquier, Ozden, and Peri (2010)
College
Percent College and Non-College, 1990s Immigrants and Existing Workers, 2000
College
100
80
0
Non-College
100
Belgium
60
40
0
20
0
College
Non-College
College
100
40
60
80
100
60
40
0
0
20
0
College
College
Portugal
80
100
60
40
Data Source: Docquier, Ozden, and Peri (2010)
Non-College
Spain
20
Non-College
College
80
100
60
40
20
0
Non-College
80
100
80
60
40
20
0
College
Netherlands
Italy
College
40
Non-College
80
100
80
20
40
60
80
60
40
20
0
College
Greece
Non-College
20
0
College
Germany
100
France
Non-College
60
80
60
40
20
Non-College
20
Non-College
Sweden
100
Canada
0
20
40
60
80
100
United Kingdom
0
20
40
60
80
100
United States
Non-College
College
Non-College
College
The standard model: what’s missing?
Factors which make benefits larger:
• Imperfect substitutability between immigrants and
natives in the same observable “skill group”
– ↑s natives’ gains from immigration: adverse distributional
consequences are borne more by immigrants themselves
– Related: gains from product variety
• Productivity spillovers
Factors which (generally) make benefits smaller:
• Long-run adjustments in production technology
• Non-wage impacts: public goods, compositional
amenities
Gains from Variety
• Immigrants increase variety of goods & svcs
– More small firms (Olney, 2013), ethnic diversity in
restaurants (Mazzolari and Neumark, 2012)
though perhaps more big-box retailers
– Hedonic value (Ottaviano and Peri, 2006)
• Another mechanism: scale effects increase the
extent of the product market
– Large effects from scale effects of immigration,
e.g., 5% of GDP in U.S. (di Giovanni et al., 2013)
Productivity Spillovers
• High skill “H1-B” (type of US visa) immigrants may raise
productivity by e.g., generating new ideas
– H1-B immigrants have high patent rates, induce more
patents from native-born: Hunt & Gauthier-Loiselle (2010),
Kerr & Lincoln (2010)
• Link to productivity? HGL say may have ↑d GDP ≈2%
in the 1990s (applying est’s from Furman et al. 2002)
– Direct association between “H1-B induced” increase in
science workers and productivity across U.S. metro areas
(Peri et al., 2013)
• 1990-2000: TFP ↑ 3.8%; higher college, but not noncollege wages
Spillovers, but…
• But Paserman (2013) finds no evidence of
productivity spillovers in Israel following the
influx of former Soviet Union immigrants
– He looks across industries and firms
• Some evidence that high-tech sectors
benefitted
• Misses aggregate effects?
– Israel too far from the technological frontier?
Spillovers, but…
• A concern: “spillover” model may be symmetrically used to
argue that unskilled immigration harms productivity:

ln  =  + 
 + 
(e.g., Moretti, 2004)
But:
– Results sensitive, especially to who is “unskilled”
• Acemoglu and Angrist (2000), Sand(2007), Iranzo&Peri (2009)
– Don’t forget: unskilled immigration is the source of benefits in
standard model (in the US)
– Indeed: in the U.S., immigration is associated with faster wage
growth, despite reducing college share: e.g., Peri(2012)…
Immigration Lowers College Share but Raises Wages
United States Metropolitan Areas, 2000-2010
Change in College Share
.1
-.2
-.05
0
-.1
0
.05
.1
.2
.15
Adjusted Wage Growth, Native-Born*
more
-.02
0
.02
.04
.06
.08
Change in Share Foreign-Born, 2000-2010
-.02
0
.02
.04
.06
.08
Change in Share Foreign-Born, 2000-2010
Data Sources: 2000 Census of Population and 2009-2011 American Community Surveys.
*Change in Mean ln(hourly wages) of native born regression adjusted for experience, education, race and gender separately in 2000 and 2010.
Other factors not in standard model
• Entrepreneurship: immigrants have high rates
– May also have productivity benefits
• Long run adjustments in production techn.
• Public goods
Conclusions
• Standard model: immigration has benefits, but
come with larger distributional consequences
– Benefits largest from immigrants with skill that are
scarce in the native population (low-skill in US)
• Standard model leaves out many things, some
of which may dwarf benefits in stand. model
– Although more research supporting this claim
would be helpful
Bonus Slides
Imperfect Substitutability details
Imperfect Substitutability
• Imms & natives are imperfect substitutes = wage
gap between them responds to their relative #s*
– In U.S. appears tied to English language skills:
• Immigrants w/strong English much closer substitutes for
natives (Lewis, 2011)
– Immigrants and natives specialize in different jobs
• Natives specialize in jobs which require communication (Peri
& Sparber, 2009)
• …and benefit from lower cost of low-skill svcs (Cortes, 2008)
– This increases natives’ net gains from immigration, as
adverse distributional consequences are borne more
by the immigrants themselves
* e.g., Ottaviano and Peri, 2012.
Back
Imperfect Substitutability
ln(Wage Gap) between similar Immigrants and Natives, by Immigrants' English Skills
.5
0
-.5
-1
-1
-.5
0
.5
1
Immigrants with no English
1
Immigrants Fluent in English
-4
-3
-2
-1
0
Ln(Immigrant/Native) Total Hours
1
-4
-3
-2
-1
0
Ln(Immigrant/Native) Total Hours
1
Data sources: 2000 Census of Population and 2007-9 American Community Surveys. Each dot represents a metropolitan area x broad education
(college/non-college) x year cell. Wage gaps computed w/in narrow experience x education cells. Regression adjusted for education x year dummies.
Back to language skills
Back to what’s missing
Aggregate Trends in Israel, Before
and after FSU Immigration
.3
ln GDP per capita*: Israel and Synthetic Control**,
before and after fall of Soviet Union
-.2
-.1
0
.1
.2
Israel
Synthethic Control**
1980
1985
1990
1995
2000
2005
Data souce: World Development Indicators. *In constant U.S. dollars, deviations from 1988 value.
**Combination of OECD countries matched on GDP/cap in 1978-1989, education, and openness.
Largest weights on Belgium (37%), New Zealand (35%), Finland(13%), and US(8%).
back
-11.8
Ln(Patents/Capita)*: Israel and Synthetic Control**,
before and after fall of Soviet Union
-12.8
-12.6
-12.4
-12.2
-12
Israel
Synthethic Control**
1980
1985
1990
1995
2000
2005
Data souce: World Development Indicators. *ln(Patents/Working Age Population)
**Combination of OECD countries matched on patents/Capita 1978,84,89; GDP/capita in 1989;
and education. Largest weights are: US(41%), UK(32%), Korea(15%), New Zealand(7%).
Investment/GDP: Israel and Synthetic Control*,
before and after fall of Soviet Union
18
20
22
24
26
28
Israel
Synthethic Control*
1980
back
1985
1990
1995
2000
Data souce: Penn World Tables and World Development Indicators.
*Combination of OECD countries matched on investment/GDP 1980-89, GDP/cap in 1989,
openness, and real interest rate. Largest weights are: Ireland(39%), US(39%), Chile(21%).
2005
.3
ln GDP per capita*: Israel and OECD, before and after fall of Soviet Union
-.2
-.1
0
.1
.2
Israel
OECD
1980
1985
1990
year
1995
2000
2005
Data souce: World Development Indicators. *In constant U.S. dollars, deviations from 1988 value
back
More Native Wage Growth
Results
Immigration Lowers College Share but Raises Wages
United States Metropolitan Areas, 2000-2010, Versus Ethnic Enclave Instrument
Change in College Share
.1
-.2
-.05
0
-.1
0
.05
.1
.2
.15
Adjusted Wage Growth, Native-Born*
0
.05
.1
.15
.2
0
.05
.1
Ethnic Enclave Instrument: Change in Predicted Share Foreign-Born, 2000-2010
.15
.2
Back
Data Sources: 2000 Census of Population and 2009-2011 American Community Surveys.
*Change in Mean ln(hourly wages) of native born regression adjusted for experience, education, race and gender separately in 2000 and 2010.
First Stage: Change in Immigrant Share vs. Ethnic Enclave Instrument
-.02
0
.02
.04
.06
.08
United States Metropolitan Areas, 2000-2010
0
.05
.1
.15
.2
Ethnic Enclave Instrument: Change in Predicted Share Foreign-Born, 2000-2010
Data Sources: 2000 Census of Population and 2009-2011 American Community Surveys.
Adjusted Wage Growth, Native-Born, by College/Non-College*
United States Metropolitan Areas, 2000-2010, Versus Ethnic Enclave Instrument
.1
0
-.1
-.2
-.2
-.1
0
.1
.2
Native-Born Non-College-Educated
.2
Native-Born College-Educated
0
.05
.1
.15
.2
0
.05
.1
.15
Ethnic Enclave Instrument: Change in Predicted Share Foreign-Born, 2000-2010
Back
Data Sources: 2000 Census of Population and 2009-2011 American Community Surveys.
*Change in Mean ln(hourly wages) of native born regression adjusted for experience, education, race and gender separately in 2000 and 2010.
.2
Immigrant Entrepreneurship
Immigrant Entrepreneurship
• Entrepreneurial ability – or willingness to take
risks – may be another “scarce” skill that
immigrants bring, with productivity benefits
– Imms more likely to start businesses (eg, Hunt 2011)
• Hunt studied college educated imms, but imms have high
entrepreneurship rates at all education levels
– Raises productivity? Immigrant-owned businesses
have 12% more revenue/worker than native-owned
businesses (Garcia-Perez, 2008)
• Also, business turnover accounts for a large fraction of
productivity growth (e.g., Haltiwanger 2009)
Self-Employment Rates by Nativity and Education
United States, 2010-12
10
9.86
Immigrants
Native-born
8.87
8
7.62
7.51
6.72
6.22
6.69
6.68
6
6.09
0
2
4
5.17
Dropout
HS Grad
1-3 Yrs Coll
4 yrs Coll
Data Source: 2010-12 March Current Population Surveys.
NOTE: Lawyers, doctors, taxi drivers, and construcion workers not counted as self-employed in this figure.
back
Adv. Degree
Details: Characteristics of
Immigrant-Owned Businesses
30
Percent in Each Revenue Category
by Nativity of Business Owner, 1992
$0
-
$4
,9
99
$5
,0
00
-$
$1
9,
99
0,
00
9
0$2
$2
4,
99
5,
00
9
0$4
$5
9,
0,
99
00
9
0
-$
$1
99
00
,9
,0
99
00
-$
$2
19
00
9,
99
,0
00
9
$2
$2
49
50
,9
,0
99
00
-$
$5
49
00
9,
99
,0
00
9
-$
99
9,
99
>$
9
1,
00
0,
00
0
0
10
20
Foreign-Born Owner
Native-Born Owner
Data Source: US Department of Commerce, Bureau of the Census. Characteristics of Business Owners.
CBO92-1, Washington, DC: US Government Printing Office. September 1997. Table 6b.
80
Percent in Each Employment Category at Employer
Businesses, by Nativity of Business Owner, 1992
60
Foreign-Born Owner
Native-Born Owner
>1
00
Em
p
p
50
-
99
Em
p
49
20
-
19
10
-
Em
p
Em
p
Em
59
14
Em
p
0
20
40
Foreign-Born: 82.4% Non-Employer
Native-Born : 84.7% Non-Employer
Data Source: US Department of Commerce, Bureau of the Census. Characteristics of Business Owners.
CBO92-1, Washington, DC: US Government Printing Office. September 1997. Table 6c.
back
Other Factors Not in Standard
Model
The Long Run
• In the long-run, adjustments in production
tech may diminish adverse distributional
impacts of immigration (Lewis, 2013)
– In response to unskilled imm, shift to (Beaudry &
Green 2003, 2005) or develop (Acemoglu 1998,
2002) more unskilled production technology
• Other mechanisms: capital adjustments under capitalskill complementarity; adjustments in product mix
– Also implies that benefits will diminish over time
back
Public goods
• Likely animates much of the opposition to
immigration, rather than labor market impacts
– Natives reveal a strong distaste for living in
neighborhoods (Saez & Wachter, 2011), sending
children to school w/imms (Cascio & Lewis, 2012)
– Also, some benefits (social security solvency)
– Needs more research
• Natives may have an exaggeratedly negative view of
immigrants’ impact on public goods
back
Sources Cited
Many of the ideas in this article derive from
Lewis E. 2013. “Immigration and Production Technology.” Annual Review of Economics 5.
The standard model is well described in:
Borjas, GJ. 1999. “The Economic Analysis of Immigration.” In Handbook of Labor Economics Volume 3A, ed. O.
Ashenfelter and D. Card, pp. 1697-1760. Amsterdam: Elsevier.
Other sources include:
Acemoglu D. 1998. “Why do new technologies complement skills? Directed technical change and wage inequality.” Q. J.
Econ. 113:1055–89
Acemoglu D. 2002. “Technical change, inequality and the labor market.” J. Econ. Lit. 40:7–72
Acemoglu D, Angrist J. 2000. “How large are human capital externalities? Evidence from compulsory schooling laws.” In
NBER Macroeconomics Annual, ed. BS Bernanke, K Rogoff, pp. 9–74. Cambridge, MA: MIT Press
Beaudry P, Green DA. 2003. “Wages and employment in the United States and Germany: What explains the
differences?” Am. Econ. Rev. 93:573–602
Beaudry P, Green DA. 2005. “Changes in U.S. wages, 1976–2000: ongoing skill bias or major technological change?” J.
Labor Econ. 23:609–48
Cascio, EU and Lewis, EG. 2012. “Cracks In the Melting Pot: Immigration, School Choice, and Segregation.” Am. Econ J.
Econ. Pol. 4: 91-117
Sources Cited (2)
Cortes P. 2008. “The effect of low-skilled immigration on US prices: evidence from CPI data.” J. Polit. Econ. 116:381–422
di Giovanni J, Levchenko A, Ortega F. 2013. “A Global View of Cross-border Migration.” Unpublished Manuscript,
International Monetary Fund
Dustmann C, Frattini T, and Preston I. 2013. The Effect of Immigration along the Distribution of Wages.” Rev. Econ.
Stud., forthcoming.
Eaton J, Kortum S. 1996. “Trade in ideas: patenting and productivity in the OECD.” J. Int. Econ. 40:251–78
Furman JL, Porter ME, Stern S. 2002. “The determinants of national innovative capacity.” Res. Policy 31:899–933
Garcia-Perez M. 2008. “Does it matter who I work for and who I work with? The impact of owners and coworkers on
hiring and wages.” Unpublished manuscript, Univ. Maryland, College Park
Haltiwanger, J. 2009. “Entrepreneurship and Job Growth.” In Entrepreneurship, Growth and Public Policy, ed. ZJ Acs, DB
Audretsch, RJ Strom, pp. 119-145. Cambridge: Cambridge Univ. Press
Hamilton B, Whalley J. 1984. “Efficiency and Distributional Implications of Global Restrictions on Labour Mobility:
Calculations and Policy Implications.” J. Dev. Econ. 14: 61-75.
Hunt J. 2011. “Which immigrants are most innovative and entrepreneurial? Distinctions by entry visa.” J. Labor Econ.
29:417–57
Hunt J, Gauthier-Loiselle M. 2010. “How much does immigration boost innovation?” Am. Econ. J.Macroecon. 2:31–56
Iranzo S, Peri G. 2009. “Schooling externalities, technology, and productivity: theory and evidence from U.S. states.”
Rev. Econ. Stat. 91:420–31
Kennan, J. 2013. “Open Borders.” Rev. Econ. Dyn., forthcoming.
Sources Cited (3)
Lewis E. 2011. “Immigrant-native substitutability: the role of language ability.” NBER Work. Pap. 17609
Mazzolari F, Neumark D. 2012. “Immigration and Product Diversity.” J. Pop. Econ. 25: 1107-1137.
Moretti E. 2004. “Estimating the social return to higher education: evidence from longitudinal and repeated crosssectional data.” J. Econom. 121:175–212
Olney W. 2013. “Immigration and Firm Expansion.” J. Reg. Science 53: 142-157
Ottaviano G, Peri G. 2006. “The Economic Value of Cultural Diversity: Evidence from U.S. Cities.” J. Econ. Geography 6:
9-44.
Ottaviano G, Peri G. 2012. “Rethinking the Effects of Immigration on Wages.” J. Eur. Econ. Assoc. 10: 152-197
Paserman DM. 2013. “Do high-skill immigrants raise productivity? Evidence from Israeli manufacturing firms, 1990–
1999.” Unpublished manuscript, Boston Univ.
Peri G. 2012. “The effect of immigration on productivity: evidence from U.S. states.” Rev. Econ. Stat. 94:348–58
Peri G, Shih K, Sparber CS. 2013. “STEM Workers, H1B Visas, and Productivity in US Cities.” Unpublished manuscript,
University of California Davis
Peri G, Sparber CS. 2009. “Task specialization, immigration, and wages.” Am. Econ. J. Appl. Econ. 1:135–69
Saez A, Wachter S. 2011. “Immigration and the Neighborhood.” Am Econ. J. Econ. Pol. 3:169-188
Sand B. 2007. “Has there been a structural change in the labor market? Evidence from U.S. cities.” Unpublished
manuscript, Univ. British Columbia, Vancouver
Sources Cited (4)
US Department of Commerce, Bureau of the Census. 1997. Characteristics of Business Owners. CBO92-1. Washington,
DC: United States Government Printing Office.

similar documents