L i Mi ti d R ittLeveraging Migration and Remittances for Developmentp

Dilip RathaMigration and Remittances UnitWorld BankWorld Bankand“Migrating out of Poverty” Research ConsortiumUniversity of SussexUniversity of Sussex

Federal Reserve Bank of AtlantaResearch Conference on Remittances and ImmigrationNovember 5, 2010

OutlineOutline

A St li d f tA. Stylized facts

B D l t i t f i t ti lB. Development impact of international migration

C. Policy implications

A. Stylized FactsA. Stylized Facts

1 O l 3% f ld l ti i t ti l1. Only 3% of world population are international migrants; 97% are not

2. Economic migrants account for 93% of global migrant stock. Economic migration is set tomigrant stock. Economic migration is set to increase in future

In future, migration pressures will increase dramaticallyincrease dramaticallyProjected Change in Labor Force, 2005–50 (millions), ages 15-39

Sub-Saharan Africa 328Middle-East & N. Africa 44Other sending regions 198All developing regions 570g

EU & other Europe -67North America -9China -85East Asia and Pacific -32E Europe & C Asia -23E Europe & C Asia 23Sub-total for these and other receiving regions -216

Source: Shaping the Future : A Long-Term Perspective of People and Job Mobility for the Middle East and North Africa (World Bank 2008)

A. Stylized FactsA. Stylized Facts

1 O l 3% f ld l ti i t ti l1. Only 3% of world population are international migrants; 97% are not

2. Economic migrants account for 93% of global migrant stock. Economic migration is set tomigrant stock. Economic migration is set to increase in future

3. South-South migration is larger than South-North migration

South-South migration is larger than migration from developing countries to high-income p g gOECD countries

Destination of migrants from the South

High-

S th

income non-OECD14%

South44%

High-incomeincome OECD42%

Source: Migration and Remittances Factbook 2011

Migrants' share of 

20%

gsending country 

population

10%

15%

5%

121

41

61

0%

81

101

121 6181

101121Sending country 

ranked by GDP per capita

141

121

41capita

Receiving country ranked by 

GDP  per capitaData source:  Ratha and Shaw (2007)

B Development impact of international migrationB. Development impact of international migration

1 Mi ti b fit ll ti th i t th1. Migration benefits all parties – the migrants, the destination country, and the origin country.

Migration boosts welfare for most householdsMigration boosts welfare for most households

Global income gains of $356 billion (0 6%)

150180

Global income gains of $356 billion (0.6%)

Change in real income in 2025, $ billion

139 143 16290120150

139 143

03060

-88

90-60-30

0

-90Natives, high-

income countriesOld migrants,

high-inc.countries

Residents,developingcountries

New migrants

countries countries.

Source: Global Economic Prospects 2006

Migration benefits all partiesMigration benefits all parties

Global income gains of $356 billion from a 3% (14 million) g $ % ( )increase in labor force of high-income countries (GEP 2006)

Global income gains of $675 billion (Anderson and Winters Global income gains of $675 billion (Anderson and Winters, 2008)

“A conservative estimate of the welfare gain to a moderately skilled worker… moving to the US is PPP$10,000 per worker, per year ” (Clemens Montenegro and Pritchett 2008)per year… (Clemens, Montenegro and Pritchett, 2008)

Dixon and Rimmer (2009) estimate that the difference between the long-run welfare effects for U.S. households of a tighter border policy and a liberalized guest worker program with an optimal visa charge is about $260 billion a yearoptimal visa charge is about $260 billion a year.

B Development impact of international migrationB. Development impact of international migration

1 Mi ti b fit ll ti th i t th1. Migration benefits all parties – the migrants, the destination country, and the origin country.

2. Benefits to countries of origin are mostly through remittances.remittances.

Remittance flows to developing countries remained

600 $ billions

resilient during the crisis

500

300

400 FDI

RecordedRemittances

200

Remittances

100Private debt & port. equity

ODA

0

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

e

port. equity

Remittances will be resilient w r to downturns in host countriesdownturns in host countries

Remittances are sent by the stock (cumulated flows) of Remittances are sent by the stock (cumulated flows) of migrants

Remittances are a small part of migrants’ incomes that can be cushioned against income shocks by migrants

Duration of migration may increase in response to tighter border controls

“Safe haven” factor or “home-bias” -- returnees will take back accumulated savingsback accumulated savings

Sectoral shifts – and fiscal stimulus packages – may help some migrantssome migrants

However Anti-immigrant sentiment isHowever, Anti-immigrant sentiment is rising in major migrant-destination

t icountries

B. Development impact of international migrationp p g

1. Migration benefits all parties – the migrants, the destination country, and the origin country.

2 Benefits to countries of origin are mostly through2. Benefits to countries of origin are mostly through remittances.

3. Emigration of skilled people may be a problem in small countries

Brain drain is a small country problem if at allBrain drain is a small country problem, if at all

Share of developing country population (%)

75%80%

p g y p p ( )

60%

19%

40%

19%

3% 3%%

20%

0%< 10% 10%-20% 20%-30% > 30%

High-skilled emigration rateg g

Source: Docquier and Marfouk (2004)

B. Development impact of international migrationp p g

1. Migration benefits all parties – the migrants, the destination country, and the origin country.

2 Benefits to countries of origin are mostly through2. Benefits to countries of origin are mostly through remittances.

3. Emigration of skilled people may be a problem in small countries

4. Diasporas also provide business contact, trade network, technology, and capital to the origin country.

Diaspora bonds can be used to tap the wealth

15 Percent

of the diaspora, often with “patriotic” discount15 Percent

11

13 US Treasury 10-year

11

13

9

11

9

11

5

7

5

7

1

3Israel DCI bond1

3

1953

1958

1963

1968

1973

1978

1983

1988

1993

1998

2003

I l d I di h i d $35 billi

1953

1958

1963

1968

1973

1978

1983

1988

1993

1998

2003

Israel and India have raised over $35 billion via diaspora bonds

OutlineOutline

A S t li d f tA. Some stylized facts

B D l t i t f i t ti lB. Development impact of international migration

C. Policy implications

C Policy implicationsC. Policy implications

1 Th i t ti l itt d1. The international remittances agenda

1 Monitoring1. Monitoring, analysis, projection

International 3. Financial access for

4. Capital market access

RemittancesAgenda

access for households

market access for institutions

g

2. Retail payment systemssystems

International remittances Agenda1. Improve monitoring, analysis, projection (MAPping)

2 I t il t t2. Improve retail payment systems:

Reduce remittance costs

Improve competition in remittance industry

Share networks - avoid exclusivity contractsShare networks avoid exclusivity contracts

Avoid overregulation of remittance industry

I t d t h l Introduce new technology

3. Leverage remittances for financial access for households

4. Leverage remittances for improving access to capital markets for institutions/countries

International remittances Agenda1. US Wall Street Reform Bill

2 US BRIDGE i iti ti2. US BRIDGE initiative

C. Policy implications

1. The international remittances agenda

2. Know your migrants/diaspora

3 Help potential migrants acquire globally marketable3. Help potential migrants acquire globally marketable skills

4 P i t b d t d d ff t4. Point-based systems can produce adverse effects on developing countries –

5. But ethical recruitment policies may be ineffective, and unethical

6. Improve transparency in recruitment of migrants

7 Border control policies should be revisited7. Border control policies should be revisited

Do border controls stop migration?

Increase in border control seems to have little (perverse?) effect on illegal migrant stock

ThousandsShare of US population

(perverse?) effect on illegal migrant stock

204.0%

p p

173 6%

3.8%Illegal migrant stock

143 4%

3.6%

113.2%

3.4%

Border patrol agents,

83.0%

Southern border (right scale)

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Source: Passel and Cohn (2010), Department of Homeland Security

Employment opportunities in the US appear to be a dominant pull factor for immigration

PercentShare of population

be a dominant pull factor for immigration

10

4 0%

4.2%p p

Illegal migrant stock

83.8%

4.0%

3.6%

6

3 2%

3.4% US unemployment rate (right scale)

43.0%

3.2%

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Source: Passel and Cohn (2010), Department of Homeland Security

Tighter border controls increase coyote fees

ThousandsUS$

Tighter border controls increase coyote fees

20

24

2,800

3,200$

162,000

2,400

12

1 200

1,600

C t f

4

8800

1,200 Coyote fees

Border patrol agents, Southern border

0

4

0

400

1984 1989 1994 1999 2004 2009

(right scale)

1984 1989 1994 1999 2004 2009

Source: DHS, MMP

Apprehensions in the US-Mexico border have not declined significantly even though US

242 0Millions Thousands

not declined significantly even though US border controls have increased

20

24

1 6

2.0

Apprehensions

161.2

1.6

12

0.8

1.2

4

8

0.4B d t l t

0

4

0.0

Border patrol agents(right scale)

1969 1973 1977 1981 1985 1989 1993 1997 2001 2005 2009

Source: DHS, CBP

Migration rises and then falls as border controls increase (the Migration Curve)

Apprehensions along US-Mexico border (millions)

increase (the Migration Curve)

1.9Apprehensions along US Mexico border (millions)

1.5

1.1

0.7

0.30 5 10 15 200 5 10 15 20

Border patrol agents (thousands)

Mi t t k t dMigrant stock today

= Existing migrant stock Existing migrant stock

– Return migration

+ New migration

OrOrMt = Mt-1 - Rt + ∆Mt

Return migration and border controlsReturn migration and border controls

Duration of Mexican migration has increased –return rate has declined – as controls have

Th d

return rate has declined as controls have been tightened at the US-Mexico border

1116Thousands months

1012

Border Patrol agents

8

98

7

8

4Migration duration

60

4 6 8 0 2 4 6 8 0 2 4 6

Migration duration (right scale)

198

198

198

199

199

199

199

199

200

200

200

200

Source: DHS, MMP

New migration and border controls

Willingness to migrate is a function of developmental

g

gaps

Ability to migrate is a function of border controls Ability to migrate is a function of border controls

Border controls increase the segmentation of labor gmarkets and increase developmental gaps

Border controls also affect income differencesBorder controls also affect income differences

CIncome difference

Maximum0Border controlBorder control

Break-up of the Soviet Union saw a divergence in income levels of FSU countries

Coefficient of variation of GDP per capita (2000 US$)

divergence in income levels of FSU countries

0.9

0.8

0.7

0.61987 1989 1991 1993 1995 1997 1999 2001 2003 2005 20071987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007

Source: DECPG

Creation of borders in South Asia led to divergence of income levels

GDP per worker (2000 US$), thousands

divergence of income levels

8

Pakistan6

Pakistan

4India

2Bangladesh

0Creation of Bangladesh

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000

Source: Penn World Tables 6.2. India and Pakistan were partitioned in 1947.

Accession of Portugal to EU in 1986 allowed it t t h ith F ’ i l lto catch up with France’s income level

Portugal GDP as % of France GDP

66%

70%

60%60%

56%

50%

40%1981 85 1986 1990 1991 19951981-85 1986-1990 1991-1995

Source: Penn World Table 6.2

Income gaps narrowed between East and West

80East Germany as % of West German value

g pGermany after unification

80

60Nominal wages

Real GDP per worker

401970 1975 1980 1985 1990 1995 2000 20051970 1975 1980 1985 1990 1995 2000 2005

Source: Penn World Tables 5.6 and Burda (2008)

New EU accession countries also saw their income levels converge

Coefficient of variation, GDP per capita (constant 2000 $)

income levels converge

0.6

0.5

0.41996 1998 2000 2002 2004 20061996 1998 2000 2002 2004 2006

Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta, Poland, Slovak Republic, and Slovenia joined the EU in 2004. Source: WDI

Migration CurveMigration Curve

Migrationg

Maximum0 Border control

Migration CurveMigration Curve

MigrationC

g

Maximum0 Border control

Migration rises and then falls as border controls increase

Apprehensions along US-Mexico border (millions)

increase

1.9Apprehensions along US Mexico border (millions)

1.5

1.1

0.7

0.30 5 10 15 200 5 10 15 20

Border patrol agents (thousands)

Migration CurveMigration CurveOn the left of the curve, shifting from border controls to development aid could be very effective

MigrationC

controls to development aid could be very effective

On which side of this curve are Mexico-US and gBangladesh-India corridors?

What is a border? Does it have to be the same for international trade, foreign investment, economic , g ,nationality, political sovereignty, cultural identity?

Maximum0 Border control

C. Policy implications

1. The international remittances agenda

2. Know your migrants/diaspora

3 Help potential migrants acquire globally marketable3. Help potential migrants acquire globally marketable skills

4 Ethi l it t li i b i ff ti d4. Ethical recruitment policies may be ineffective, and unethical –

5. Improve transparency in recruitment of migrants

6 Border control policies should be revisited6. Border control policies should be revisited

7. Migration is not a substitute for employment creation at homecreation at home