question in description
Theory
A worker is looking at the economic benefits and costs of migrating from Poland to Ireland.
Compares the stream of (net) income would get in Poland with stream of (net) income in Ireland. If Stream of income in Ireland is greater then Migrate.
PV is present value of an income stream and r is the discount rate.
Cost of Migration is M.
Present value of future wage stream in Ireland.
Present value of future wage stream in Poland.
=wage of worker with given age in a given country
Worker moves if net gain is positive.
1. Improved opportunities in the destination increases likelihood of migration.
2. Improved opportunities at home decrease likelihood of migration.
3. An increase in migration costs lowers the gain to migration making it less likely.
Net Gain to migration is
In developing countries migration can be an important way of diversifying risk.
Modelling risk
A raffle ticket where if I win (with probability p pays income 100. If I lose (there are only two states) I get 50
Expected income = p100+(1-p)50
If p=0.5 in the above example a risk neutral person would pay 75 for the raffle ticket
A risk averse person would pay less than 75 and a risk lover would pay more than 75
If workers do not like risk what actions can they take to reduce risk
Look for less risky work
Buy Insurance
Diversify
For workers in poorest country there may not be a ready availability of secure employment that guarantees livelihood.
Credit markets are very limited so insurance option (save in good times, borrow in bad) often not available
Employment and migration patterns may partially reflect an attempt to diversify
Workers may have several occupations (Agriculture, selling goods) If the factors that lead to bad agricultural income are not correlated , or better still negatively correlated with the factors that lead to bad outcomes in the other occupation this is a hedge (form of insurance) against a bad outcome.
One household member migrating to city while others stay on farm may have similar effects
Ireland
Source country
Source country
Ireland
Skills
Skills
S*
S*
Earnings
Earnings
If Ireland has high returns to skill “Roy model” predicts high skill workers migrate to Ireland.
If Ireland has low returns to skill “Roy model” predicts high skill workers migrate from Ireland.
The Roy Model: do low or high skill workers move?
Assumption in Roy model that immigration policy is not selective.
In many cases regulations and restrictions will restrict migration patterns in a way that serves the interests of the native population
For example this model predicts low skill groups will migrate to countries with low income inequality.
High skill groups will do better in countries with higher inequality (other things equal).
This is when inequality driven by high returns to skill
Effect on Native-born Workers
Immigration reduces the wages and employment of similarly-skilled native-born workers, but native-born workers may be able to increase their productivity by specializing in tasks better suited to their skills.
Competing native workers will have lower wages; complementary native workers will have higher wages.
The Short-Run Impact of Immigration When Immigrants and Natives Are Perfect Substitutes
Dollars
Supply
w0
w1
Demand
N0
Employment
E1
N1
As immigrants and natives are perfect substitutes, the two groups are competing in the same labor market. Immigration shifts out the labor supply curve. As a result, the wage falls from w0 to w1, and total employment increases from N0 to E1. At the lower wage, the number of natives who work declines from N0 to N1.
The Short-Run Impact of Immigration when Immigrants and Natives are Complements
w1
w0
Dollars
Supply
Demand
N1
N0
Employment
If immigrants and natives are complements, they do not compete in the same labor market. The labor market here denotes the supply and demand for native workers. Immigration makes natives more productive, shifting out the labor demand curve. This leads to a higher native wage and to an increase in native employment.
The Long-Run Impact of Immigration When Immigrants and Natives Are Perfect Substitutes
Dollars
Supply
w0
w1
Demand
N0
Employment
N0 + Immigrants
Immigration initially shifts out the labor supply curve so the wage falls from w0 to w1. Over time, capital expands as firms take advantage of the cheaper workforce, shifting out the labor demand curve and restoring the original wage and level of native employment..
Arguably in short/medium term high/low skill migration bad for high/low skill natives. May be good for low/high skill natives.
If doctors from other countries allowed to work here might lower doctors wages but improve service and costs to patients.
If low skill migration improves services and lowers costs of hotels, shops restaurants etc. good for high skill workers?
Empirical evidence on impact of migration on native wage
Many studies have found little evidence that sudden increase in low skill migration lowers wages of low skill natives.
Mariel boatlift in Cuba (studied by David Card) is a famous example.
Others (George Borjas) argue that immigrants do impact on the native wage.
The Native Labor Market’s Response to Immigration
Dollars
PPT
w0
Demand
(b) Pittsburgh
Employment
S0
S3
w*
Dollars
w0
PLA
wLA
(a) Los Angeles
Employment
S0
S1
S2
Demand
w*
Originally, both markets pay equilibrium wages of w0. After immigration into Los Angeles, both markets eventually converge to a new equilibrium wage at w*, which is less than w0.
Manacarda et. al look at native/migrant wage gap in U.K. between late 70s-2000.
Ottavania and Perri for US
These papers develop a model where migrants and natives imperfect substitutes
They argue that natives and migrants are different.
Occupational segregation even within skill groups.
Selection of migrants
higher skill, income and education typically of working age.
Used to be men were more likely to migrate but this has changed.
Since 60’s increased income divergence, population growth rates concentrated in poor countries but higher average income meant poorer countries moved up along the “migration hump”. Transport costs fell.
These factors countered by increasing administrative barriers especially for low skill migrants from poorer countries.
The gains from migration
Clemens suggests that there are large gains from migration: 66%-147% increase in global GDP from removing barriers
Small easing of barriers would have substantial change
40% of adults in poorest quartile would like to move countries permanently
Enormous potential benefits
Why are these benefits not realised?
10
Urban wage
Rural wage
50
Rural demand
Urban demand
15
5
25
Last urban worker produced 15, last rural worker 5
If workers are the same in both regions the efficiency gain is the area between the demand curves for the last 25 rural workers employed
Net Gain from Migration
Domestic workers lose
Migrant workers gain
Owners of capital also gain
Four issues
What are the externalities associated with skilled migrants leaving developing countries (Brain drain)
This is difficult to measure but what evidence we have is that they are not that large. Spillover effects would have to be huge to substantially effect the sending country GDP relative to migration gains
Research shows that return migrants may have positive spillover effects
Spillover effects on destination countries may be positive or negative
Remittances from migrants represent a substantial part of GDP in sending countries
2. What is the elasticity of labour demand in origin and destination countries (How much will wages change and how large are the gains from migration)
Elastic demand means small gains from migration but also small effects on wages
Most estimates find small effects of migration on wages but this possibly not associated with inelastic demand, rather how similar migrants and natives are etc. Most estimates suggest demand is relatively inelastic
In long run increased investment would reduce any fall in wages associated with migration. This effect (this may happen fairly quickly)
3. How much of the international differences in productivity depend on worker traits
Differences in estimates of gains mostly reflects assumptions on this
In theory it makes sense that migrants are positively selected from source country but there may be country/cultural specific traits that lower productivity in the destination country
Migrants typically earn less than natives after controlling for skill and job characteristics
Evidence suggests these differences are relatively small
4. What future level is feasible
Political economy issues
There are winners and losers in migration
There is a literature that tries to devise side payments to compensate the losers
Should we tax inward migrants to compensate natives for short term loss
Remember working migrants will pay tax in any case so this may not involve higher tax for migrants/natives (especially if migrants have low dependency rates)
For Ireland Barrett et al (2013) show migrants less likely to claim welfare than natives in years before crisis.
While there was a surge in migrant claims during the great recession this possibly to reflect migrants higher probability of job loss
Dustmann and Frattinni look at migrants to the UK between 1995-2011
Analyse whether were net contributors to government receipts in terms of tax-welfare-cost of government services
Found positive effect especially in recent years
OECD find effects small over a range of countries
Loss in South output is C, Gain in North is A+B (shaded Area is gain in output)
Borjas (2015) suggests a small negative spillover from excessive migration may substantially reduce these gains
The wage equals the marginal product of labour and in a Cobb-Douglas example
North (n) is rich country and South(s) is poor:
Take logs to calculate elasticities:
Borjas captures negative spillover effect by letting intercept of North be a weighted average of the North and South.
He suggests that spillovers can dramatically reduce and even reverse the surplus.
Conclusion
Evidence that enormous welfare benefits associated with easing restrictions on migration
The barriers to easing these possibly depends on a range of political economy issues
.
2122
20
2
....
1(1)
irlirl
irlirl
ww
PVw
rr
=++
++
2122
20
2
....
1(1)
polpol
polpol
ww
PVw
rr
=++
++
country
age
w
irlpol
PVPVM
--
nnnssn
wlwl
hh
aa
==
logloglog
logloglog
nnn
sss
wl
wl
ah
ah
=+
=+
*
(1)
nsn
alala
=+-