Transcript Behind Great Walls

The Effect of Inter-regional Migration on
Regional Economy Disparity in China
(1990-2005)
Lianqing Peng
South China Normal University
Outline
1. Introduction
2. Survey of Studies
3. The Characteristics of Labor Migration in China
4. The Empirical Study
5. Conclusions
1.Introduction
 Since the implementation of policy of reform and China’s
“opening up” to the outside world, there has been
massive migration from Chinese inland countryside to
coastal cities. This has deeply changed Chinese
economic development and social life, and has
influenced regional economic disparity.
 This study looks at the effect of massive inter-regional
migration on the economic gap across provinces. It asks
if this migration is narrowing or widening regional
economy disparity.
2. Survey of Studies
 The New-Classical Economic theory: (narrowing)
the inter-regional migration →the regional economy convergence
（Braun, 1993; Barro and Sala-i-Martin, 1995）

 The New Economic Geography theory: (could do both)
the inter-regional migration →Widening of the gap →Narrowing
of the gap
（Krugman & Venables,1996; Baldwin, Forslid, Martin,2003）

 Many Chinese Scholars’ empirical studies:

The inter-regional migration has widened the gap in across
China’s regions. Why couldn’t the inter-regional migration lead to
the regional economic convergence? May suggest that this is
because of China’s hukou polices.
Shi Li (2003), ( Du, et al.,2005),Qin (2006), Cai(2007)
Duan(2008), Cheng(2012)
3. The Characteristics of Labor Migration in China
3.1 Five Periods and the scale of Chinese migration

Five Periods of Chinese migration
Year
Stage Description
Period 1
1979-1983
Period 2
1984-1992
Beginning movement(allowed but
controlled )
Allowing movement/loosening
(most of migrants were peasants)
Period 3
1993-1997
Period 4
1998-2005
Period 5
2006-present
Accelerating movement/loosening
(most of migrants were peasants)
Accelerating movement/loosening
(more and more urban labors migrate)
Accelerating movement/loosening
(more movement trends)
3.1 Five Periods and the scale of Chinese migration
 The 1st stage, 1979－1983

Implementation of Chinese reform and opening policies
which encourages the beginning of a labor market. Some
farmers migrate to cities to make a living.
3.1 Five Periods and the scale of Chinese migration
 The 2nd stage, 1984－1992

More and more farmers move to work in local factories in
Eastern China(Jianshu, Zhejiang, Guangdong) because
the rise of non-state-owned companies.
3.1 Five Periods and the scale of Chinese migration
 The 3rd stage, 1993－1997


The market economy had rapidly developed in eastern region.
Mass migrants moved to cities in
Eastern region from Chinese inland
（central and western） countryside.
3.1 Five Periods and the scale of Chinese migration
 The 3rd stage, 1993－1997

The labor supply exceeded the demand in many Eastern cities
contributed to many social problems.
Unpaid wages, the boss
can’t pay the workers
on time
3.1 Five Periods and the scale of Chinese migration
 The 4th stage, 1998－2005


Many state-owned companies began to reform and layoff
workers who also become unemployed in Chinese inland（
central and western) cities.
Many labors moved to cities in Eastern region from Chinese
inland（central and western） cities.
3.1 Five Periods and the scale of Chinese migration
 The 5th stage, 2006－present



The geographic trends of migrating started to change as more
migrants moved to Chinese inland cities
Migrants start to settle down in the cities and urbanization
accelerates.
Eastern cities began to experience labor shortages.
3.1 Five Periods and the scale of Chinese migration

The Scale of Migration
Other estimation
ten thousands
Census and sample survey
Year
1983
1987
1989
1990
1993
1994
1995
1996
1997
1998
1999
2000
2005
2010
Inter-towns
Inter-counties
Inter-provinces
Inter-towns
200
3000
6200
8000
7223
3890
4936
5204
-
4300
5200
4487
2602
3218
3622
-
700
2050
2400
2364
1488
1872
2125
-
3053
3413
5400
14439
14735
22143
Inter-provinces
632
1183
2500
4242
4779
8588
3.2 Flows of migration
inflow areas
To eastern
1987
1990
1995
2000
2005
To central
1987
1990
1995
2000
2005
To western
1987
1990
1995
2000
2005
Sum
outflow areas
From eastern From central
From western
5252.0%
.0%
54.6%
63.1 %
75.0 %
84.6 %
84.6%
49.7%
56.0%
63.5%
64.4%
78.3%
61.7%
59.0%
71.8%
84.3%
89.8%
44.2%
49.3%
56.5%
68.3%
80.1%
24.6%
24.0%
18.8%
9.8%
5.5%
31.3%
28.4%
20.5%
19.7%
10.9%
21.8%
23.5%
12.7%
7.1%
4.4%
21.2%
20.4%
13.4%
7.9%
4.4%
23.3%
21.4%
18.1%
15.3%
9.9%
18.9%
15.6%
16.1%
15.9%
10.8%
16.6%
17.5%
15.5%
8.6%
5.8%
34.6%
30.3%
30.2%
23.9%
15.5%
3.2 Flows of migration
 Migration occurred mainly from the central and western
regions to the eastern region which is more developed.
Main outflow areas
Main inflow areas
3.3 The main characteristics of labor migration
 (1) The synchronization : inter-regional migration and
inter-sector migration.
Outflow areas
Inflow areas
Year
From
countryside to
countryside
city
countryside
city
1982-1987
68.0%
32.0%
23.6%
76.4%
74.4%
1985-1990
62.5%
37.5%
17.3%
82.7%
78.5%
1990-1995
59.8%
40.3%
28.6%
71.4%
60.2%
1995-2000
58.7%
41.3%
11.9%
88.2%
69.0%
2000-2005
61.3%
38.7%
15.6%
84.4%
80.3%
city
3.3 The main characteristics of labor migration
 (1) The synchronization : inter-regional migration and
inter-sector migration.

The structure of employment of inter-regions labors
industry
%
eastern
central
western
sum
The primary industry
5.68
31.94
22.07
8.27
The secondary industry
74.85
33.27
32.26
69.56
Manufacturing
66.41
17.16
16.58
60.19
Construction
7.33
9.57
12.37
7.81
Other
1.12
6.54
3.3
1.57
The tertiary industry
19.47
34.79
45.68
22.17
3.3 The main characteristics of labor migration
 (2) "circular migrants“
Many migrants had dual identity: Farmer working in the city
 Job in the city and family in the countryside
 Regular movement
The structure of population in some eastern cities(2005)
thousand

City
T-POP
Have hukou
No hukou
Ratio
(no hukou)
Beijing
15380
11841
3539
23.01%
Shanghai
17780
13603
4177
23.49%
Guangzhou
9942
7029
2913
29.30%
Shenzhen
7009
1223
5786
82.55%
Dongguan
6446
1544
4901
76.04%
Fushai
5338
3284
2054
38.48%
3.3 The main characteristics of labor migration
 (3) Selective migration hasn’t emerged in the eastern
region.


The average years of education of migrants in eastern region
was less than that in central and western regions.
The average years of education of migrants leaving eastern
region was more than that in central and western regions.
The average years of education of immigrants
year
Flow out
From eastern From central From western sum
Flow in
To eastern
-
9.01
8.76
8.91
To central
10.23
-
9.38
9.88
To western
9.77
9.01
-
9.32
sum
10.02
9.01
8.82
4. The Empirical Study
4.1 The Evolution of Regional Economy Disparity in China
 The gap of regional economy had narrowed from 1978 to
1990, but widened from 1991 to 2005.
The Coefficient of Variance (COV) of per capita GDP each
province in China
1.2
1
0.8
0.6
0.4
0.2
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
1978
0
4.1 The Evolution of Regional Economy Disparity in China
 The gap of regional economy had narrowed from 1978 to
1990, but widened from 1991 to 2005.
the contrast of per capita GDP in three regions
Year
1990
1995
2000
2005
Eastern
100
100
100
100
Central
60.47
51.98
49.22
47.56
Western
52.25
43.17
38.56
39.82
* The per capita GDP in eastern region was assumed “100”.
4.1 The Evolution of Regional Economy Disparity in China
 The gap of regional economy had narrowed from 1978 to
1990, but widened from 1991 to 2005.
Had the inter-regional migration caused the Regional Economy
the contrast of per capita GDP in three regions
Disparity?(which verifies the hypothesis of New Economics)
Year
1990
1995
2000
2005
Geography).
It is difficult to solve the problem by the Conventional
Eastern
100
method
of econometrics.
100
100
100
Central the question,
60.47 Taylor
51.98
49.22
47.56
To answer
and Williamson（1997）
Proposed
ask another:
Westerna new method.
52.25 They
43.17
38.56what would
39.82
* Thethe
per measured
capita GDP in convergence
eastern region washad
assumed
“100”.
have been
there
been no
(net) migration?
4.2 The Method of Study
 Laborer is producer and consumer.


on the one hand, more migrants means more input
factors, which is beneficial for the output;
on the other hand, more migrants may influence the per
capita GDP.
 So, the impact of migration to per capita GDP in
inflow area and outflow area depends on the
comparation of the migration elasticity of output
and the migration elasticity of population.
 Taylor and Williamson Proposed a new method to
estimate the two elasticity.

The method of mathematical derivation in the appendix
4.3 Analysis of the results
 The results:
 Table 1990
 Table 2000
 Table 2005
The result shows that the inter-regional migration
narrowed the gap of regional economy. In other
words, the inter-regional migration caused regional
economy convergence.
4.4 An explanation:
the income of migrants transferred
 Most of migrants’ income was earned in the city but
transferred to their families in central and western
regions.
The amount of income transferred from immigration in eastern region
Year
Per worker
Income
transferred
（$）
Labors(flew in eastern
region), million
From central From western
Income transferred(from
eastern), $ billion
To central
To western
1990
205.28
0.96
1.24
0.29
0.38
2000
546.80
14.48
7.93
11.87
6.51
2005
618.49
18.36
10.70
17.03
9.93
5. Conclusions
1.The China’s inter-regional migration led to more efficient allocation of
labors across regions, which was helpful in national economic
growth.
2.The eastern region was a net in-migration region, and labor migration
had promoted economic development in the eastern region; At the
same time, the wages (or income) of migrants had been raised.
3.The central and western regions were the net emigration region,
where many rural families of emigration benefits through income
transfers from the eastern region. As a result, the growth rate of percapita consumption and per-capita income had been greatly
increasing, which also promoted regional development.
4. The China’s inter-regional migration has caused convergence not
divergence of regional economies.
Thank you
4.2 The Method of Study
 Assume a aggregate production function for output,
Y=F(L, K, R)
where Y is total output, L is labor input, K is capital and
R is land.
 we assume the producer price P is equal to one.
taking K and R as fixed.
Y  F ( L, ...)
dY  FL ( L, ...)dL
(1)
4.2 The Method of Study
Y  F ( L, ...)
dY  FL ( L, ...)dL
(1)
 Under long-run full employment conditions, competitive
wages are equal to labor’s marginal product, where w is
wage.
w  FL (L, ...)
(2)
 Solving (1) and (2), produces an equation (3), .
dY  w dL
(3)
4.2 The Method of Study
dY  w dL
(3)
 Divide two sides of equation (3) By Y , produces an new
equation(4).
dY w
wL dL
 dL 
Y
Y
Y L
dY
dL
we assume Y＊＝ ，L＊ 
Y
L
wL ＊
wL
Y＊ 
L  L＊
( 
)
Y
Y
(4)
(5)
 In equation(5) , θ is the ratio of wages in output,
approximately equal to the migration elasticity of output.
4.2 The Method of Study
 we assume:
M is cumulative net migration rate, POP*= M;
a share αM of its migrant stream is active in the labor force,
while its total population has an active share αP.
 Moreover, assume that migrants have an effective-worker
(or worker-quality) ratio of μ with respect to the total labor
force.
 the labor content of the population is:
L =αP POP
 and the labor content of the migrant flow is:
dL = μαM M POP
4.2 The Method of Study
 the labor content of the population is:
L =αP POP
 and the labor content of the migrant flow is:
dL = μαM M POP
 Migrant streams of population measured by M can be
converted into labor supply shocks L* .
dL  M
L 

M  M
L
P
＊
 M
( 
)
P
(6)
 Defining γ = (the migrant-to-population ratio of labor-force
participation rates)
4.2 The Method of Study
wL ＊
Y 
L  L＊
Y
＊
wL
( 
)
Y
(5)
dL  M
 M
L 

M  M
( 
)
(6)
L
P
P
 We can now derive the simulation equations used to
calculate the impact of migration on per capita GDP:
＊
(Y / POP)＊  Y＊  POP＊  L＊  M  (  1)M
(7)
 If we estimate M, θ and γ , we can calculate the impact of
migration on per capita GDP. Then we can estimate per capita
GDP without migration.
 So we can appraise the change of the gap of per capita GDP
between outflow area and inflow area, which approximatively
measures the Effect of Inter-regional Migration to Regional
Economy Disparity.
Data and estimation of parameters
 This paper calculated M, θ by analyzing the data of
1990,2000 Population Census and 2005 Population Sample
Survey.
 This paper estimated γ with the ratio of Compensation of
Employees in GDP each province. Data were from The
Chinese Statistical Yearbook.
return