Labor Market Consequences of Immigration Econ/Demog C175 Economic Demography Prof. Goldstein Spring 2018, UC Berkeley 1
Agenda Is international migration good or bad for the United States? Last time, fiscal effects This time: Labor market effects Simple one-sector model Empirical evidence What could be wrong with our simple analysis? 2
Trade-offs in considering costs and benefits of immigration immigrants vs. natives skilled vs. unskilled natives consumers of market goods vs. consumers of public goods increased tax revenue vs. increased expenses federal finances vs. local finances national labor market vs. local labor market what else? 3
A simple model One good One kind of worker Downward sloping demand curve reflecting marginal product Competitive market à all works paid product of marginal worker 4
Our picture rents to capital owners supply of native workers D L wage w rate for labor wage bill number of workers 5
The effect of immigration (1) S L D L S' L w w' wage rate immigrant labor number of workers 6
The effect of immigration (on migrants) S L w w' wage rate D L Immigrant income S' L What is the gain to migrants? (Not shown) immigrant labor number of workers 7
The effect of immigration (redistribution among natives) S L D L S' L w w' wage rate Redistribution from native workers to native owners Immigrant income immigrant labor number of workers 8
The effect of immigration ("net gain" to natives) S L D L S' L w w' wage rate Redistribution from native workers to native owners Immigrant income "Net gain" from immigration -- who gets this? -- how big is it? immigrant labor number of workers 9
The effect of immigration ("net gain" to natives) S L w w' wage rate D L Redistribution from native workers to native owners Immigrant income S' L How big is net gain? Say we have 10 natives for every 1 migrant and w = (w w' Then, redistribution = 10* w How much is gain? Hint: area of triangle is ½ * base * height. immigrant labor number of workers 10
Political economy comment (if the simple model is right,...) Net gains are small compared to workers losses. Net gains are also small compared to capitalists gains. A good explanation why "net gains" story might not be convincing. 11
So, immigration appears to Create a net gain for native-born But reduce salaries of workers Gain goes to capitalists complementary workers (the skilled) consumers (from lower prices due to cheaper labor) Good to be a complement; not good (at least from wage perspective) to be a substitute. Gain depends on steepness of demand curve for unskilled labor and amount of immigration 12
Empirical evidence Wage effects Research design cross-sectional studies of small areas (not so good) natural experiments (e.g., Mariel boatlift and Miami wages; influx of meat processors to Alabama.) Small effect near zero for all kinds of labor some evidence that latest immigrants do get hurt (e.g., Grossman finds 10% increase in immigrants reduces wages of earlier immigrants by 2% Card finds same increase reduces earlier immigrant wages by 4% in Miami) 13
Elasticity of Native wages Study Impact on: Elasticity Altonji and Less skilled natives +.01 Card Bean, et al. Black men ~ 0 Borjas White native men -.01 Black native men +.02 Grossman All natives -.02 LaLonde and Young black natives -.06 Topel Note: an elasticy of -.06 means that a 10% increase in immigrants will reduce wages by only (.1)(.06) = 0.006 = 0.6% 14
Alabama case study 15
Set-up In the small city of Albertville, a wave of thousands of Mexican migrants, driven by recession in Mexico and jobs in U.S. starts in mid-1990s This study looks at what happened to wages and unemployment in Albertville, compared to other similar places. I highly recommend the whole audio show! 16
Synthetic control We then identify within Alabama the labor markets which, prior to 1990, looked similar to CZ 6100 in terms of immigrant share, wage, and employment rates of the less educated but which did not experience large inflows of immigrants after 1990. 17
Checking to make there is a difference in "treatment" from control. 18
Focus on < HS 19
Effect on wages Minimum wage à only effects in unemployment? 20
No effect on unemployment either 21
Peri and Wiltshire conclude 22
Are we convinced? 23
Why is there so little effect? Demand curve may have almost no slope Demand curve may shift upwards as a result of more consumers? Labor markets are not localized, so people who are hurt might move away and not be observed (a potential problem of study design) 24
The effect of immigration (revisited) With no demand effect With demand effect S L D' L S L D L S' L D L S' L w w w' w' immigrant labor number of workers immigrant labor number of workers Who are winners and losers if both SL and DL shift? 25
Advantages of the simple model Explains why (historically) workers tend to oppose more immigration and capitalists tend to favor more immigration Net gain (like comparative advantage) explains why economists tend to favor open borders Magnitude of redistribution explains political conflict 26
Problems with the simple model We don't actually see much or any decline in wages Could be that all studies are wrong... Or it could be that simple model is wrong... immigrant and native workers are complements immigrants increase demand... 27
My conclusion Simple model probably has too much influence ("econ101-ism") Effects of migration on native workers appear to be very small Perhaps anti-immigration sentiment largely driven by psychology and cultural issues, not economics Next week effects of migration on immigrants themselves! 28