20-Mar-14

Impact of Immigration on Destination Country: Accounting for More Adjustment

Most of the economic models and empirical studies discussed so far conclude that the overall effects of immigration on destination economies are modestly positive, while the distribution effects could be skewed/biased Traditional labor market models failed to take into account of several facts, such as:
Native workers may move to another region; Native workers may move to alternative jobs/professions; New investment by domestic/foreign capital! to create jobs to take the advantage of newly available labors "nhanced aggregate demand generated by the immigrants etc
1

CECN 640
The Economics of Immigration -Effects of Immigration (II)
(Reading: Chapter 7)

Impact of Immigration on Destination Country: Accounting for More Adjustment

Immigration and Internal Migration

&mmigration to a particular region would increase local labour supply, would therefore e+ert a downward pressure on the wage rate there ,ne possible response of the natives is to move to another region "mpirical evidence however is mi+ed and inclusive

$onger%term effects of immigration:

&nternal migration responses native migration pattern! 'apital adjustments (roduct market responses )djustments in industrial composition Technological choice

Immigration and Internal Migration ( iler! 1""2#

.iler /00#, see 121 pp3 /4#%/4*! employed data from the /056 7383 'ensus to e+amine mobility patterns by natives during a five year period between /09: and /056! for #9# 7383 cities ;ependent variable: native mobility rate &ndependent variable: immigration arrival rate 'ontrol: various characteristics of the area and its labor force .iler Table 53-, ne+t slide!

20-Mar-14

Immigration and Internal Migration ( iler! 1""2! cont$#

1ased on simple OLS, .iler finds:
<igher immigrant arrival rates deter inflows of natives, but $ittle evidence to suggest that out%migration is encouraged

Immigration and Internal Migration ( iler! 1""2! cont$#

1ased on a *8$8 model, the findings:
)mong natives, migration effects much larger than ,$8 estimates /= increase in immigration reduces native net in%migration by *3*-= &ndicates that migratory responses more than offset immigrant arrivals if true, cities receiving large flows of immigrants are likely to shrink in siAe! Mobility response even stronger among low%skilled native workers

&n statistics numbers, /= increase in immigrants would:

reduce native net migration by 63/#=
recall: Net migration > immigrants ? emigrants!

reduces native in%migration by 635*=, but no significant statistical evidence for out%migration

<owever, ,$8 estimator could be biased because location decisions can be endogenous to area characteristics "stimators such as &@/#8$8/*8$8 should be used instead

Immigration and Internal Migration ( iler! 1""2! cont$#

Bhy low%skilled native workers respond more heavilyC
More competition for jobs reDuiring low%level skills because most immigrants are low skilled not like 'anadaC! (rejudice against immigrants and foreign culture particularly among whites!

Immigration and Internal Migration

8trong evidence of native response to the arrival of immigrants by leaving is also found in many others studies .rey /00:!: 'alifornia natives tended to leave in response to more immigrants to the Eolden 8tate, and the out%migration consists of two systems:
an immigration%induced FflightF that sees lower income and less%educated 'alifornians moving primarily to the nearby states of Bashington, ,regon, Nevada, and )riAona more responsive stream of GflightsH! a more conventional migration e+change with the rest of the 7nited 8tates that involves the redistribution of better educated, higher income migrants3 IThis is part of a process which .rey is to call Jdemographic balkaniAationK of 7383 cities3 8ee .rey, Billiam <3, /000, G&mmigration and ;emographic 1alkaniAation,H in Lames B3 <ughes and Loseph L3 8eneca eds! )mericaMs ;emographic Tapestry: 1aseline for the New Millenium, New 1runswick, NutgersO

)s well, .iler finds that in his sample, relatively well to do white workers are also more likely to move, and the reasons:
Ereater access to capital markets 1etter connection hence easier to find another job in alternative areas because white culture dominates in better part of the country!

8ee .rey, Billiam <3 /00:3 F&mmigration and &nternal Migration F.lightF: ) 'alifornia 'ase 8tudy3F Population and Environment, /4 -!: *:*%9:3 Neprint No3 --03

/6

%rig&t! et al' (1""(# )&e lin*age +et,een immigration and internal migration in large metropolitan areas in t&e -nited .tates
Bright et al3 /009! re%e+amined .reyHs specifications and conclude that native outflows from large metropolitan areas are unrelated to immigrant inflows3 Nather, Bright et al3 using the /056 and /006 78 census micro% samples, and controlling for the siAe of the areas, find that large metropolitan areas suffer net losses of internal migrants for reasons other than the flow of immigrants to these localities:
net migration of the native born for metropolitan areas is either positively related or unrelated to immigration net migration loss of unskilled native workers from metropolitan areas is probably a function of those citiesM population siAe rather than immigrant flow to them conclude that the net migration loss of native born low%skilled workers from large metropolitan areas is more likely the result of industrial restructuring than of competition with immigrants )s well, net migration of natives with high levels of education, is statistically significantly positively related to immigration flows

/orjas! reeman! 0 1at2! (1""(#: 3o, Muc& Do Immigration and )rade Affect 4a+or Mar*et 5utcomes
1orjas et al3 build another model, taking into account of the fact that the native population of various regions could grow at various rate in the absence of immigration 7sing a Gdouble differenceH techniDue!3 Main findings:
immigration does not decrease the native population, but decrease the native labor force so immigration causes natives to commute to neighboring labor markets! &mmigration does not have a consistent, discernible effect on area economic outcomes There are other regional factors dominate the ups and downs of area economies!, but: &mmigration has a marked adverse impact on the economic status of the least skilled 7383 workers high school dropouts and those in the bottom #6 percent of the wage distribution!3
8ee: 1orjas, E3, .reeman, N3, 2 PatA, $3 /009!3 <ow much do immigration and trade affect labor market outcomes3 1rookings (apers on "conomic )ctivity, /, /?493

'ause of this difference: Model specification

//

/#

20-Mar-14

/orjas (2007#! 8ati9e Internal Migration and t&e 4a+or Mar*et Impact of Immigration
1orjas #664!, using data from the /046?#666 decennial censuses, concludes that immigration is associated with:
lower in%migration rates, higher out%migration rates, and decline in the growth rate of the native workforce3 The native migration response attenuates the measured impact of immigration on wages in a local labor market by -6 to 46 percent, depending on whether the labor market is defined at the state or metropolitan area level3

/orjas (2007#! 8ati9e Internal Migration and t&e 4a+or Mar*et Impact of Immigration
1orjas also directly estimated the impact of the immigrant share on the native workforce and migration rate3 .indings:
the impact of immigration on the native workforce diminished with the siAe of the area3 )t the city level, for every ten immigrants that arrive, slightly over : natives leave3 )t the state level, around # natives will leave the workforce for every ten immigrants that enter3 )t the 'ensus division level, the estimated effect was found to be unstable, varying between nil and appro+imately # natives3

http://jhr3uwpress3org/content/Q$&/#/##/3full3pdfRhtml

/*

/-

3atton! 0 )ani! (200:#' Immigration and Inter-;egional Mo+ility in t&e -1! 1"<2=2000
<atton and Tani #66:! in a study of the local labour market response to immigration in the 7P, find: Most studies that look across local labour markets have found the effects of immigration to be benign3 ,ne possibility is that
immigrants to a specific area do indeed push non%immigrants onwards elsewhere, thereby diffusing the negative labour market wage effects The effects are not uniform across the 7P stronger in the southern regions where immigrants are more heavily concentrated! 'onclusion: internal migration is indeed one of the mechanisms through which regional labour markets adjust to immigration shocks3

Card 0 Di8ardo (2000#! Do immigrant inflo,s lead to nati9e outflo,s

'ard 2 ;iNardo #666!, using data from the /096, /056, and /006 'ensuses, investigate the e+tent to which skill%group%specific changes in the immigrant population across various M8)Hs has led to flights of similarly skilled native%born individuals
'ontrary to the demographic%balkaniAation hypothesis .rey, /00:!, 'ard and ;iNardoHs results suggest that, if anything, increases in immigrant population in specific skill groups lead to small increases in the population of native%born individuals of the same skill group3 This pattern also suggests that systematic out%migration by native%born individuals is not very likely to provide an e+planation of the small measured effects of immigration on the labor%market outcomes of the native%born population found in most Jarea analyses3K .ind that immigration has had Duite significant impacts on the skill distribution of various M8)Hs3 'onclusion: local labor market impacts of unskilled immigration are mitigated by other avenues of adjustment, such as endogenous shifts in industry structure, rather than by rapid adjustments in the native population3

/:

/4

Migration and t&e >stimates of Immigration$s %age >ffect

"ffect of immigration on native wage rates is another touchy issue facing the public, policy makers, and economists )gain, results, theoretical and empirical, are inconclusive

/orjas (2007#: 8ati9e internal migration and t&e la+or mar*et impact of immigration
1orjas #664! asked how much of the difference between the estimated wage effects of immigration obtained from national studies and those obtained from local studies can be e+plained by the diffusing effects of internal migration3 Main findings ne+t slide!:
Bhen sample of the labor market under consideration includes both men and women, a /6= increase in immigration share will:
Neduce weekly earning by :3*#= at the national level3 Neduce weekly earning by *3:= at the 'ensus division level Neduce weekly earning by #39= at the state level, and Neduce weekly earning by between 6= and 634= at city level

The estimated wage effect thus nearly evaporates when the labor market is disaggregated down to the local level3

/9

/5

20-Mar-14

Impact of Immigration on Destination Country: Accounting for More Adjustment

/orjas (2007#: 8ati9e internal migration and t&e la+or mar*et impact of immigration
)s well, large differences in the wage effects of immigration across geographic definitions of the labor market could be accounted for by native labor mobility3 assuming that the national labor market appro+imates a closed economy, so that the estimated wage effects of immigration at the national level reflect the true elasticity of the native born wage with respect to immigration3! Main findings:
native mobility responses account for about -6= of the gap between state% level wage effects and national%level effects; and

mobility responses account for as much as 46=of the difference between city wage effects and national effects3 ,verall, these results all suggest that internal migration is an important albeit secondary! adjustment process3
/0 #6

Immigration and Industry .tructure

NybcAynsky theorem: )t constant relative goods prices, an increase in the endowment of one factor of production will lead to a more than proportional e+pansion of the output in the sector which uses that factor more intensively, as well as an absolute decline of the output of the other good3 )ccording to the theorem, immigration to a region should lead to a relative e+pansion of those industries employing immigrants more intensively

Immigration and Industry .tructure

&f large number of immigrants arriving are highly skilled, the theorem would suggest that the destination region should e+perience an e+pansion in industries that use high skilled labour more intensively, and vice versa Most studies, in the 7383 conte+t, however, find that evidence is not very supportive of the theorem Nelative wages and employment rates in different local labor markets of the 78 are surprisingly unaffected by local factor supplies, in particular, immigrant workers

#/

##

Immigration and Industry .tructure

$ewis #66*!, in studying the puAAle between the NybcAynsky theory and evidence, evaluated two possible e+planations:
78 cities are not closed economies, but are better described as <eckscher%,hlin open economies, which can trade away local skill imbalances by specialiAing in production; and &ndustries adapt production technology to better utiliAe the local skill mi+ of workers3

Immigration and Industry .tructure

$ewis #66*! contH! Main findings:
'hanges in labor supply indeed have some influence on the mi+ of industries for e+ample, apparel tends to grow with the supply of less% educated labor!, but the changes are typically small in comparison to the supply changes )cross the /90 metropolitan areas in the 78, cities that received more of one type of labor had a little more tendency than other cities to JgrowK industries that tend to employ it3 <owever, the skill intensity of most industries is highly responsive to city%specific changes in skill mi+: )n increase in the relative supply of some type of labor leads to an increased intensity of its use across a wide range of industries that is, on average, around 56 percent as large as the supply change
"3g3 sectors like finance and professional services took on a relatively greater share of dropout labor where it was locally abundant

wages are relatively unresponsive to local supply, there seems to be accommodating changes in production technology
#* #-

20-Mar-14

Immigration and Industry .tructure

$ewis #66*! contH! 'onclusion:
changes in the mi+ of industries accommodate only a small part of the changes in the composition of local labor supply in 78 cities3 &n low% skill immigration cities and in 78 cities in general changes in industry mi+ did not tend to favor employment of newly arriving workers during the /056s standard <, model is not a very good description of how local labor markets adjust to labor mi+ shocks changes in industry mi+ are not a major source of adjustment to labor supply shocks3 8till, vast majority of industries are highly responsive to the composition of local labor supply in hiring3 industries tend to choose production technology to complement local factor supply mi+
"3g3: computer use grew more rapidly in areas with a faster relative growth in the supply of educated labor

Immigration and Demand

&mmigrants will increase labour supply in the labour market, creating a downward pressure on wages &mmigrants will also consume, thus generate higher demand, and shift the labour demand @M(! schedule up to the right, which would at least partially offset the downward pressure on wages To what e+tend immigrants will generate enough demand in their host economy, for their own labour, as well as the nativesHC

Nesults consistent to theory of endogenous technological change3

#: #4

Bage Nate e

Immigration and Demand

8 L6 8 L1 s m t p D t s ;$/ ;$6 ) 1 ' ; $abor r

d B6! f B#! c B/!

b a 6

To what e+tend immigrants will generate enough demand in their host economy for their own labour and their hostHsC &f immigrants spend at least part of their earnings in the destination economy on goods and services produced locally, immigration will trigger changes in the derived demand for more labor, including their own3 Bages and employment rates in the destination economy would therefore be affected3 'hanges in immigrant remittance rates and changes in public spending on goods and services for immigrants could further shift derived demand3 )s well, if new immigrants indeed drive down wages in the host economy, the resulting lower labor costs will shift the product supply curve and generate lower costs for consumers3 &mmigration is therefore likely to change product prices directly through consumer demand or indirectly through product supply channels3
#9 #5

SuiA: Bhat is the Gimmigration surplus H in this caseC

Immigration and Demand: 3erco,it2 and ?as&i9 (2002#

<ercowitA and Tashiv #66#! predicts that immigrants will delay entry to the destination economyHs labor market finding job takes time!, but buying in the goods market immediately food, housing, etc!3 )dverse labor market effects of immigration should be delayed sometime after the arrival of immigrants3 ;estination price level should be e+pected to rise first, then go down, and finally settle at some eDuilibrium level Bage rates would probably remain unchanged or even go up! for a while then could go down as immigrants join the work force

Immigration and Demand: 3erco,it2 and ?as&i9 (2002#

<ercowitA and Tashiv #66#! tested their theory using Duarterly data on the arrival and assimilation of Lewish Nussian immigrants to &srael over the years /006?/000 Main .indings:
(roduct prices go up first though not very significantly!, then go down significantly!, largely as predicted )fter a year following an immigration shock, native employment e+perienced some significant negative effects

#0

*6

20-Mar-14

iscal Implication of Immigration

(olicymakers, ta+payers and thus economists! in destination countries are concerned about how government costs are affected by immigration:
&n most developed economies, public services and government transfers are a large chunk of E;( 8mall changes in the amount of ta+es paid by immigrants or services used by immigrants could be much greater than the small estimated overall net economic effects of immigration on the destination country3

iscal Implication of Immigration

&n the 78, however, evidence does not support the belief that immigrants are a fiscal burden to the state
e+cept for refugees and the elderly, immigrants actually use less government services than natives &f refugees are included who are indeed heavy users of government services and recipients of government transfers!, immigrants as a group still use government services only slightly more often and receive only slightly more welfare payments than natives )ccording to the /006 census, 0= of immigrant families received welfare payments, while 93-= of 7383%born families received the same (re%/05# legal immigrants to the 7nited 8tates living in the si+ states with the largest immigrant populations were actually paying more in total ta+es than they received in government%provided benefits3

&mmigrants do not seem to make settlement decisions based on the availability of welfare and social services; they settle where jobs are and where their close family are3

*/

*#

iscal Implication of Immigration

&n "urope where fiscal issues of immigration are more contentious, studies suggest that the ta+%transfers ratio is not as burdensome as is often feared
governments adjust both ta+es and transfer programs to improve the balance for native workers3 NaAin, 8adka, and 8wagel #66#! found that for // "uropean countries, both ta+es on workers and transfers to the poor were reduced as immigrants came to represent a higher percentage of the population3

>@ternalities of Immigration
&mmigration may generate various kinds of e+ternalities, positive or negative:
Naising productivity throughout the economy, as a result of the increasing siAe of the market, and level of competition3 Naising level of technology as a result of new ideas, new products and new production methods that may come with the immigrants, which could raise the productivity of all factors in the economy )s well, because immigration increases the total siAe of the economy, immigrants allow further e+ploitation of the scale economies3 )s well, on the negative side: increased crime, pollution, and the destruction of traditional culture Necently, Nobert (utnam suggests that immigration reduces social solidarity and social capital, which could also undermine productivity
*-

'anadaHs newly inaugurated Gsuper visaH for parents and grandparents serve the same purpose

**

Immigration and >conomies of .cale

Nosenberg /00-!: rapid economic growth of the 7nited 8tates in the /566s was the result of GGrapid growth in demand and circumstances conducive to a high degree of product standardiAation3HH The 7nited 8tates was able to e+ploit economies of scale because its domestic markets grew very rapidly and, The rapid growth of the markets was in turn the result of Jthe e+tremely rapid rate of population growth 3 3 3 with immigration assuming a role of some significance in the /5-6s3HH IThe same logic applies to 'anada and )ustralia, apparently, if only referring to different periods of timeO

Immigration and >@ternality

&mmigration and (ositive "+ternality in the ;estination 'ountry

&f positive e+ternality prevail, wage rate will not fall but rise to w*!3 &n this situation, native workers will gain the checkered area between w/ and w*, which is eDual to shaded area, a3 &mmigrants gain additional income eDual to b, and other factors capital, etc3! gain area c3
*: *4

20-Mar-14

Immigration and >@ternality

&n a very recent study, GThe "ffect ,f &mmigration ,n (roductivity: "vidence .rom 7383 8tatesH, (eri has three findings to offer:
.irst, there is no evidence that immigrants crowd out employment of or hours worked by! natives3 8econd, immigration is significantly associated with total factor productivity growth3 Third, such efficiency gains are unskilled biasedUlarger, that is, for less educated workers3

Immigration and >@ternality

Negative "+ternality in the destination country could also result from immigration of course, because of, for instance, increased crime, pollution, and the destruction of traditional culture

(eri claims that their findings are robust to including several control variables individually such as N2; spending, technological adoption, sector composition, openness to international trade, or sector composition!, and they are not e+plained by productivity convergence across states or driven by a few states or particular period of time3 (eri conjectures that at least a part of the positive productivity effects are due to an efficient specialiAation of immigrants and natives in manual%intensive and communication%intensive tasks, respectively in which each group has a comparative advantage!, resulting in a gain in overall efficiency3 Net inflow of immigrants, even those driven by their historical location and pro+imity to the border Me+ican%)merican border!, is associated with significant productivity gains for the receiving states3
*9

&f negative e+ternality prevail, wage rate will fall even more heavily from w/ to w# then to w*!3 &n this situation, even the usually positive Gimmigration surplusH may be wiped out, if, as shown in the above figure, area f is greater then g fVg!
*5