Migration and Development: Recent Evidence and Some Lessons for Africa International Conference Sustainable Growth and Development in Africa Dakar, November.

Migration and Development: Recent Evidence and Some Lessons for Africa

International Conference Sustainable Growth and Development in Africa

Dakar, November 25, 2013

Christian Ebeke

International Monetary Fund

Washington, D.C., USA

International migration: What’s behind the ramp-up?

Upward trend in migration across destinations …

1990 2000 2010 20130

50,000,000

100,000,000

150,000,000

200,000,000

250,000,000

82 306 645 103 388 690

129 737 280 135 583 436

71 855 339 71 127 043

90 992 020 95 938 779

Less developed regionsMore developed regions

Stock of migrants(by destination regions)

Sources: United Nations.

… with an important share within Sub-Saharan Africa.

1990 2000 2010 20130

20

40

60

80

100

120

62.2 56.347.0 47.8

21.224.2

29.1 28.8

13.0 13.3 15.4 15.2

Oceania North America LAC Asia Europe Africa

World's distribution of African migrants(in percent of total Africa's emigration)

Sources: United Nations; author's estimates.

Host countries’ immigration policies have tightened for migrants from the developing world …

Middle East & North Africa

Europe & Central

Asia

Sub-Sa-haran Africa

South Asia Latin Amer-ica & Car-

ibbean

East Asia & Pacific

0

0.5

1

1.5

2

2.5

2.3 2.3 2.2 2.2

1.81.5

How tight are immigration policies?(tightening faced by migrant-sending regions)

Sources: World Populations Policies; World Bank Bilateral Migration Database; and author's calculations.

1990 2000 2010 20130

20

40

60

80

100

120

32.8 29.4 25.1 25.7

22.1 25.2 30.1 29.6

29.4 26.9 21.9 22.1

15.6 18.5 22.9 22.6

Female outside Africa Female in Africa Male outside AfricaMale in Africa

Gender distribution of African migrants(in percent of total)

Sources: United Nations; and author's estimates.

Female emigration is relatively large

However, there is a large heterogeneity in emigration intensities in Sub-Saharan Africa.

0

5

10

15

20

25

30

35

40

African migrants living abroad in 2013(as a percentage of total population of the country of origin)

Sources: United Nations; and author's estimates.

Is Sub-Saharan Africa loosing its “Brain”?

Skilled emigration from Africa is not necessarily large compared to other regions …

0

10

20

30

40

50

60

Share of skilled migrants in total emigration in the 2000s(in percent)

Sources: E. Artuc, F. Docquier, C. Ozden, Ch. Parsons (2013)

… although in critical sectors such as health, the direct leakage seems important.

19

69

-ea

rlie

r

19

70

-79

19

80

-89

19

90

-99

20

00

-10

19

69

-ea

rlie

r

19

70

-79

19

80

-89

19

90

-99

20

00

-10

19

69

-ea

rlie

r

19

70

-79

19

80

-89

19

90

-99

20

00

-10

SSA-trained US-trained Trained elsewhere

0

500

1,000

1,500

2,000

2,500

3,000

Sub-Saharan African migrant physicians identified in the US physician workforce(Graduation cohorts)

Sources: American Medical Association; Tankwanchi et al. (2013).

Some countries saw a rise in outflows …

0 10 20 30 40 50 60 700

10

20

30

40

50

60

70

4.53.0

19.8

2.2

7.45.6

26.2

11.9

8.8

52.3

4.14.5

Share of Sub-Saharan African-trained international medical graduates in the U.S. physician workforce(in percent of total physicians in source countries)

Sources: American Medical Association; Tankwanchi et al. (2013).

Data as of 2002

Data

as

of

2011

… with young qualified physicians exiting Africa’s labor force.

27 29 31 33 35 373

5

7

9

6.79999999999996

5.40000000000009

4.19999999999982

7.79999999999996

8.79999999999996

6.099999999999915.400000000000

094.79999999999996

7.09999999999991

8

6.90000000000009

7.29999999999996

Characteristics of the SSA born and trained doctors in the U.S.(Data as of 2011)

Sources: American Medical Association; Tankwanchi et al. (2013); and author's calculation.

Estimated mean age at U.S. entry

Years of service before migrating to the U.S.

Is the Sub-Saharan region benefiting from remittances?

Remittance inflows to Sub-Saharan Africa have fully recovered their pre-crisis levels …

2007 2008 2009 2010 2011

100113.0

105.3 110.4118.6

100

113.1123.8 125.1

133.0

Remittance-to-pre-crisis levelAid-to-pre-crisis level

Africa: Remittances and Aid in comparison to pre-crisis levels(2007=100)

Sources: African Development Indicators; author's computation

… and represent an effective insurance mechanism

against disasters events …

-5 -4 -3 -2 -1 0 +1 +2 +3 +4Time

0

0.3

0.600000000000001

0.900000000000001

Remittances inflows around natural disaster events in Sub-Saharan Africa

(median values in percent of GDP; average over 1980-2010)

Sources: EM-DAT CRED; and author's calculations.

… and during political conflicts.

-4 -3 -2 -1 0 +1 +2 +3 +4Time

0

0.5

1

1.5

2

2.5

3

3.5Battle deaths > 100 casualties

Remittances inflows around major conflict events into Africa(median values in percent of GDP; average over 1980-2011)

Sources: African Development Indicators; and author's calcula-tions.

Remittance inflows also help mitigate food price shocks …

18.05

8.44

9.81

05

10

15

20

Rem

ittan

ce s

hock

(in

pe

rcen

t)

High Intermediate Low

Combes, Ebeke, Ntsama, and Yogo (2013). World Development.

Remittances and Food price shocks

Range of vulnerability to food price shocks

Remittance inflows are strongly associated with booms in the construction sector in SSA…

-4 -3 -2 -1 0 +1 +2 +3 +4Time

0

1

2

3

Remittances inflows around booms in the construc-tion sector

in Sub-Saharan Africa(median values in percent of GDP; average over 1980-2012)

Sources: African Dev. Indicators; and author's calculations.

… thereby reducing the prevalence of working poverty.

ALB

ALBALBALB

ALBARGARGARGARGARGARGARG

ARM

ARMARM

ARMARMARM

ARM

AZE AZE

AZEBDI

BENBFABFA

BFA

BGD

BGDBGD

BGD

BIHBIH

BIH

BLR

BLRBLZBLZ

BOLBOL

BOL BOLBRABRABRABRABRABRABRABRABRABRABRABRABRABRABRA

BWA

CAFCHLCHLCHLCHL

CIVCIVCIVCIVCIV

CMR

CMR

COG

COLCOLCOLCOLCOLCOLCOL

CPV

CRICRICRICRICRICRICRICRICRICRI

DJI

DOMDOMDOM

DOM DOMDOMDOMDOM

ECUECUECUECU

ECUECU

EGYEGYEGYEGYETH

ETH

ETH

FJI

GAB

GEOGEOGEO

GEOGHA

GHAGHA

GINGIN

GIN

GMB

GNB

GNBGTMGTM GTM

GUYGUY

HND

HND

HNDHND

HNDHND

HND

IDNIDNIDN

IDN

IDN

IND

IND

IRNIRN

IRN

JAMJAMJAM JAM

JOR JORJOR

JORKAZ

KAZKAZKAZ

KENKEN

KENKEN

KGZ

KGZKGZ

KGZ

KGZ

KHMKHM

LAOLAOLAOLAO

LBR

LKALKALKA

LKA

MAR MARMARMDA

MDA

MDA

MDA

MDAMDA

MDGMDGMDGMDGMDG

MDV

MDV

MEXMEXMEXMEXMEXMEXMEXMEX

MEXMEX

MKD

MKD

MKDMKD

MKD

MLI

MLIMLI

MNG

MOZ

MOZMOZMRT

MRT

MWIMYS

MYSMYSMYSMYS

NAM

NERNERNER

NGA

NGA

NGA

NICNIC NIC

NIC

NPL

NPL

PAKPAK

PAK

PAKPAK

PANPANPAN

PANPANPANPAN

PAN

PANPAN

PERPERPERPERPERPER

PHLPHL

PHLPHL PHLPHL

PNG

PRY

PRY

PRYPRY

PRYPRYPRY

RWARWA

SENSENSEN

SEN

SLB

SLE SLVSLVSLVSLV

SLVSLVSLV

SLVSLVSRB

SUR

SWZ

SWZ

SYR

TGO

THATHATHATHATHATHATHA

TJK

TJKTUNTUNTUR

TURTURTUR

TZATZA

UGAUGAUGA

UGA

UKRUKR

URYURY

VENVENVENVEN

VEN

VNM

VNMVNM

VNM

YEM

YEM

ZAFZAF

ZAFZAF

ZMBZMB

-40

-20

02

04

0W

ork

ing

po

vert

y re

sid

ua

ls

-10 0 10 20 30Remittance-to-GDP residuals

Data sources: ILO KILM 7th Edition dataset, The World Bank; and Ebeke et al. (2013).

Remittances reduce working poverty

Now, let’s discuss some new empirical evidences:

Political and Institutional Effects of Migration

Migration and Transfers of Norms

• Return migration and: – Voter turnout rates in Mali: (Chauvet and Mercier, 2013)– Convergence in fertility choices in Egypt: (Bertoli and

Marchetta, 2013)– Cross-country study of the convergence in fertility rates:

Beine et al. (2013).

• Migration and governance back home– In Cape Verde: Batista and Vicente (2012)

• Emigration and democracy:– Cross-country work: Docquier et al. (2011)

Remittance inflows are associated with lower political accountability in Sub-Saharan Africa …

Ebeke and Yogo (2013).

AGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGOAGO

BDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBDIBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBENBEN

BFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFABFA

BWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWABWACAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAFCAF

CIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIVCIV

CMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMRCMR

COGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOGCOG

COMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOMCOM

ETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETHETH

GABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGABGAB

GHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGHAGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGINGIN

GMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMBGMB

GNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNBGNB

KENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKENKEN

LSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOLSOMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDGMDG

MLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMLIMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZMOZ

MRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRTMRT

MWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWIMWI

NAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAMNAM

NERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNERNGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGANGA

RWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWARWA

SDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDNSDN

SENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSENSEN

SLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLESLE

TCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCDTCD

TGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGOTGO

TZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZATZA

UGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGAUGA

ZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMBZMB

ZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWEZWE

20

40

60

80

100

Vote

r tu

rnout ra

te (

%)

0 2 4 6

Remittances per capita (log)

… therefore reducing the political business cycle.Combes, Ebeke and Maurel (2013).

0.75

0.26

0.17 0.19

0.2

.4.6

.8E

lect

ora

lly-m

otiv

ate

d fis

cal i

mp

uls

e (

in p

erc

en

t o

f G

DP

)

[0, Q1] [Q1, Q2] [Q2, Q3] [Q3, max]Remittance-to-GDP range

How best to maximize the benefits of migration in Sub-Saharan Africa?

More dual-citizenship laws?

Remittances are significantly larger in countries allowing holding multiple-citizenships …

Ebeke (2013).

40.1

142.9

050

100

150

Rem

itta

nces p

er

ca

pita

(in

US

D)

0 1No dual-citizenship Dual-citizenship

3.9

5.2

01

23

45

Rem

itta

nces in

perc

en

t of G

DP

0 1No dual-ctizenship Dual-citizenship

… and so is the sensitivity of private investment to remittances. (Ebeke, 2013).

0.17

0.320

.1.2

.3C

orre

latio

n (I

nves

tmen

t ; R

emitt

ance

s)

0 1No dual-citizenship Dual-citizenship

Barriers to people mobility exist, even in the sending countries! Look at passport costs …

0

10

20

30

40

50

60

70

60.15

50.91

41.45

31.126.3

19.8217.7513.4113.3612.711.49.958.98.78.598.536.676.035.194.854.194.14.023.733.63.583.43.012.92.72.62.52.432.292.232.11.971.911.91.811.711.541.51.491.431.361.291.211.171.11.11.11.031

0.760000000000001

0.700000000000001

0.670000000000001

0.6500000000000010.510.50.50.470.430.410.40.360.320.310.290.270.160.120.1

Passport costs around the developing world(as percentage of GDP per capita)

Sources: McKenzie, D. (2005); The World Bank Group.

Passport costs are especially high in Sub-Saharan Africa

How about remittance transaction costs?

Sending money to Africa is more costly …

05

10

15

20

Cos

t of s

end

ing

US

$ 20

0

EAP ECA LAC MENA SA SSA

Cost of sending US$ 200 (in percent)

Source: World Bank; author's estimates.

… and more so when considering within-Africa corridors.

51

01

52

02

5C

ost o

f sen

din

g U

S$

200

Outside SSA SSA

Is sending remittances from SSA more expensive?

Source: World Bank; and author's estimates.

Sending money to Africa also takes much more time to arrive…

02

46

Spe

ed o

f tra

nsfe

rs

EAP ECA LAC MENA SA SSA

Remittances: Median speed of transfers

Source: World Bank; and authors' estimates.

… and even more within Sub-Saharan Africa’s corridors.

12

34

56

Spe

ed

Outside SSA SSA

Is sending money from Africa less rapid?

Source: World Bank; and author's estimates.

Costs are too high across operators, …

51

01

52

0C

ost o

f sen

din

g U

S$

200

Bank MTO Non-Bank FI Post office

(in percent of amount transfered)Cost of sending US$ 200 to Sub-Saharan Africa (SSA)

Source: World Bank; and author's estimates.

0.5

11

.52

Inde

x of

com

petit

ion

EAP ECA LAC MENA SA SSA

Competition is lower regarding remittances sent to SSARemittance inflows: How competitive are the markets?

Source: World Bank; author's estimates

… , and sending money to Sub-Saharan Africa is achieved from a less competitive market structure …

0.2

.4.6

Inde

x of

com

petit

ion

Outside SSA SSA

Are within-SSA corridors less competitive?

Source: World Bank; and author's estimates.

… especially in within-Africa corridors.

Unresolved issues in the migration-development literature

I. Long-term growth and development? Despite strong evidence that migration and remittances lower poverty, no strong evidence that they promote long-term growth. Why?

– Does the size matter?– Brain drain?– Remittances fully consumed rather than re-invested? – Are remittances an engine of informal sector growth?– Public moral hazard: Migration/Remittances increase corruption (Abdih et al., 2011;

Ahmed, 2013), reduce public social spending (Ebeke, 2012). – Private sector’s moral hazard: reduce accountability and voter turnout (Ebeke and Yogo,

2013), remittances and lower labor force supply.

II. Harnessing the diasporas: – How to maximize the re-investment of migrant remittances? – and how to fully internalize the benefits of returnees?

Thank you for your attention.