Upload
mariah-perkins
View
216
Download
0
Embed Size (px)
Citation preview
Income volatility, labor mobility, and fiscal policy in Russian regions
Goohoon Kwon Antonio Spilimbergo
IMF IMF, CEPR, WDI
Outline
• Brief Description of Regions (GDP. Income shocks. Welfare. Integration)
• Tale of three adjustments• Consequence for Unemployment • Source of regional volatility• Fiscal Response to oil shocks• Transfers• Conclusions
Data
• Russia (89 regions)
• US (50 states + DC)
• China (30 provinces)
• Canada (11 provinces)
• Europe 15 (178 regions)
Regional Income Differences
.2.3
.4.5
.6S
tand
ard
Dev
iatio
n of
Reg
iona
l Inc
ome
1975 1980 1985 1990 1995 2000year
Russia ChinaUS Canada
Europe
Figure 1: Regional Income Dispersion
Regional Income Shocks
0.0
5.1
.15
Sta
ndar
d D
evia
tion
of R
egio
nal G
row
th
1975 1980 1985 1990 1995 2000year
Russia ChinaUS Canada
Europe
Figure 2: Size of Regional Shocks
Regional shocks (controlling for regional trends)
0.0
2.0
4.0
6.0
8.1
Sta
ndar
d D
evia
tion
of d
etre
nded
Reg
iona
l Gro
wth
1975 1980 1985 1990 1995 2000year
Russia ChinaUS Canada
Europe
Figure 3: Size of Reg. Shocks (controlling for regional trends)
Why are Russia regions subject to large idiosyncratic shocks?
• Natural Resources are concentrated in few regions
• Soviet faith in economy to scale has created “excessive” specialization
How to measure shocks?
• Oil shock = (oil price) * (size energy sector)
• Industrial shock = rer * (size manufacturing sector)
Source of Shocks
Table 1. Regional growth and regional shocksFixed effects Arellano-Bond
oil shock 1.64***industrial shock -0.11**Lag D.gdpgrowth -0.04D.oil shock 2.28 *** D.industrial shock -0.23*** Constant 2.85 1.98***
Number of samples 760 608Number of Regions 76 76** p<.05; *** p<.01
Tale of three adjustments
How do economies react to regional shocks?
Do people stay in distressed areas or do they move?
To answer this question we need to make an econometric detour
(Small) Econometric Detour
itj
jitjj
jitjit
itj
jitjj
jitjit
dummiesregionaldummiesyearpopulationincomepopulation
dummiesregionaldummiesyearpopulationincomeincome
2
3
12
3
12
1
3
11
3
11
lnlnln
lnlnln
Income Shocks: response of Income
0
.02
.04
.06
0 2 4 6 8 0 2 4 6 8 0 2 4 6 8
Europe, lminc, lminc Russia, lminc, lminc US, lminc, lminc
90% CI orthogonalized irf
Inco
me
Re
spo
nse
Years
Graphs by irfname, impulse variable, and response variable
Income Shocks: response of population
-.002
0
.002
.004
0 2 4 6 8 0 2 4 6 8 0 2 4 6 8
Europe, lminc, lpop Russia, lminc, lpop US, lminc, lpop
90% CI orthogonalized irf
Po
pu
latio
n r
esp
on
se
Years
Graphs by irfname, impulse variable, and response variable
Consequence on Regional Unemployment
.2.3
.4.5
.6R
ussi
a/U
S/E
urop
e
1975 1980 1985 1990 1995 2000year
Russia US
Europe
Figure 7: Coefficient of Variation of Regional Unemployment
Adjustment - American Style
• Flexible Labor and housing markets.
• Limited unemployment benefits.
• Labor is highly mobile.
• Variation in regional unemployment is limited.
Adjustment – European Style
• Rigid labor and housing markets
• Unemployment subsidies.
• Regional fiscal transfers.
• Scarce labor mobility.
• Large variation in regional unemployment.
• Persistency in regional unemployment.
Adjustment – Russian style
• Large shocks
• Scarce labor mobility
• Rigid housing market (housing subsidies)
• Fiscal transfers not necessarily targeted to compensate shocks
Adjustment – Latin American style
• Large shocks
• Small role for fiscal policy (procyclical)
• Population concentrates in large cities (Argentina, Chile, Uruguay among the highest urban concentration)