Maurizio Bussolo and Denis Medvedev Development Prospects Group

Do remittances have a flip side? A general equilibrium analysis of remittances, labor supply responses and policy options for JamaicaMaurizio Bussolo and Denis Medvedev

Development Prospects Group

Outline

1. What is the “flip side”?

2. Effects of remittances

3. Model setup

4. Description of simulations

5. Main findings

6. Sensitivity analysis

7. Conclusions

1. What is the “flip side”?

2700

2800

2900

3000

3100

3200

1990 1992 1994 1996 1998 2000 2002

US

$ (c

on

stan

t 20

00)

800

850

900

950

1000

1050

1100

1150

1200

Th

ou

san

d

GDP percapita

laborforce

employedworkers

1. What is the “flip side”?

0

500

1000

1500

2000

2500

1995 1996 1997 1998 1999 2000 2001 2002

Average Weekly Real Wage (CPI, 1995=100)

1. What is the “flip side”?

0

5

10

15

20

25

1976 1980 1984 1988 1992 1996 2000

Remittances as a share of GDP

Remittances as a share of household f inal consumption

2. Effects of remittances

Similar to aid flows? (Connell and Brown, 2005)

What about labor supply? Endogeneity problems In Mexico, the sign varies (Amuedo-Dorantes and Pozo,

2005) In Nicaragua and Philippines – negative sign

(Funkhouser, 1992; Rodriguez and Tiongson, 2001) In Jamaica – negative sign (Kim, 2006)

3. Model setup

Single-country CGE model Comparative statics Nested CES production structure Freely mobile factors Savings-driven investment Armington assumption Endogenous labor supply

3. Model setupProduction sectors and Commodities Factors of Production

1 Export Crops 23 Skilled Labor2 Food Crops 24 Unskilled Labor3 Livestock 25 Capital and Land4 Forestry Fishing5 Mining Institutions and other accounts6 Food Products 26 Household7 Processed Sugar 27 Government8 Beverages and Tobacco 28 Investment and Savings9 Textiles and Clothing 29 Indirect taxes

10 Wood Products 30 Tariffs11 Paper and Print 31 USA12 Refined Oil 32 European Union13 Chemicals 33 Rest of the World14 Capital Goods 34 Balance of Payment 15 Electricity and Water16 Construction17 Commerce18 Transport19 Financial and Insurance Services20 Real Est. & Business Services21 Government Services22 Other Services

3. Model setup

Single-country CGE model Comparative statics Nested CES production structure Inter-sectoral factor mobility Savings-driven investment Armington assumption Endogenous labor supply

3. Model setup

Maximize

subject to

)ln(0

N

iiii Cu

N

iii yWTYCP

0

N

iii

i

iii PY

PC

0

)(

3. Model setup

N

iiiPy

WTLS

1

000 )())(1(

00

LS

y

WLS

y

y

LSy

N

iiiW Py

LSWLS

W

W

LS

1

0 )(

4. Description of simulations

Simulation a): 10 percent increase in the level of remittances with respect to their 2002 value

Simulation b): simulation a) plus 25 percent reduction in payroll taxes

Simulation c): simulation b) plus compensatory sales taxes to get a public revenue-neutral policy response

5. Main results: simulation a)Percent change from initial equilibrium

-1.0

0.0

1.0

2.0

3.0

Labor supply Wages

Unskilled Skilled

-3,822 -4,563

5. Main results: simulation a)Percent change from initial equilibrium

-3.0

-2.0

-1.0

0.0

1.0

Real GDP Exports Imports Real exchangerate

5. Main results: simulation b)Percent change from initial equilibrium

-1.0

0.0

1.0

2.0

3.0

4.0

5.0

Labor supply Wages Labor supply Wages

Unskilled Skilled

-3,822 -4,563 -1,061 -2,424

Simulation a Simulation b

5. Main results: simulation b)Percent change from initial equilibrium

-3.0

-2.0

-1.0

0.0

1.0

2.0

3.0

Real GDP Exports Imports Real exchangerate

Simulation a

Simulation b

5. Main results: simulation c)Percent change from initial equilibrium

-1.0

0.0

1.0

2.0

3.0

4.0

5.0

Labor supply Wages Labor supply Wages

Unskilled Skilled

-3,814 -4,552 -808 -2,052

Simulation a Simulation c

5. Main results: simulation c)

Percent change from initial equilibrium

-4.0

-3.0

-2.0

-1.0

0.0

1.0

2.0

Real GDP Exports Imports Real exchangerate

Simulation a

Simulation c

6. Sensitivity analysisHouseholds with remittance

income, percent of total

0

10

20

30

Employed Unemployed Inactive All

Distribution of households with remittance income, percent

0

10

20

30

40

50

60

70

Agriculture Manufacturing Services

Completed primary

Completed secondary or higher

6. Sensitivity analysisDep. Var: Labor income Coefficient Robust Std. Err. P>tIndividual characteristicsYears of schooling 0.196*** 0.015 0.00Experience 0.06*** 0.006 0.00Experience squared -0.001*** 0.000 0.00Male 0.4*** 0.052 0.00Self-employed 0.122 0.126 0.33Geographic dummy variables(Kingston)St. Andrew 1.014*** 0.146 0.00St. Thomas 0.4** 0.171 0.02Portland 0.454*** 0.145 0.00St. Mary 0.177 0.129 0.17St. Ann 0.248* 0.131 0.06Trelawny 1.151*** 0.229 0.00St. James 1.609*** 0.244 0.00Hanover 0.366** 0.150 0.02Westmoreland 0.856*** 0.135 0.00St. Elizabeth 0.525*** 0.127 0.00Manchester 0.238* 0.128 0.06Clarendon 0.704*** 0.155 0.00St. Catherine 0.873*** 0.122 0.00Sectoral dummy variables(Services)Agriculture Sectors -0.339*** 0.094 0.00Manufacturing -0.134 0.082 0.10significance levels: *** 1%; ** 5%; * 10%: no stars insignificantNumber of obs 2461F( 20, 2440) 33.31Prob > F 0.00R-squared 0.21Root MSE 1.25

6. Sensitivity analysisPercent change from initial equilibrium

-4.0

-2.0

0.0

2.0

4.0

6.0

8.0

Labor supply Wages Labor supply Wages

Unskilled Skilled

-11,048 1,848 -6,790 3,200

Simulation a Simulation c

6. Sensitivity analysis

Percent change from initial equilibrium

-4.0

-3.0

-2.0

-1.0

0.0

1.0

2.0

Real GDP Exports Imports Real exchangerate

Simulation a

Simulation c

7. Conclusions

Remittances are not “bad” “Flip side”

Competitiveness Dynamic consequences

Policy may help Tax shift Distributional effects