esearch on the Relationship between Remittance and Subjective

Well-Bing

Empirical Analysis based on CHIPS Micro Data

Dungang Zang

, Meicen Liu

and Jiahui Chen

Institute of Rural Finance, Sichuan Agricultural University,Huimin Road,Chengdu, China

Department of financial manegement, Sichuan Agricultural University,Huimin road, Chengdu, China

3 Department of financial manegement, Sichuan Agricultural University,Huimin road, Chengdu,China

961082547@qq.com, 386041813@qq.com

Keywords: Rural－Urban Migrants; Subjective Well-Being; Remittance.

Abstract: Using the data from the Rural Urban Migration in China 2008, we estimate the relationship and m

echanism between the remittance and subjective well-being of the rural － urban migrants in China.

And we get some conclusions after robustness tests and use work months of the year as the instru

mental variable to deal with the possible endogenous bias. Our results show that remittances and the

subjective well-being of the immigrants has a weak but significant and negative relationship and the

marginal effect of remittance on happiness is higher than the marginal utility of revenue, even if the effect

of per capita remittances and the relative remittances is eliminated.

1 INTRODUCTION

After China carried out reform and open policy, the

limit of labor mobility has been gradually loosened.

Cheap labor forces and favorable policies have

attracted labor-intensive processing industries

overseas to transfer to eastern coastal China and this

trend brought the “peasant worker rush”

phenomenon. With the implement of OBOR and

other national strategies, new peasant workers will

be the significant force to promote China’s economy

and push our new urbanization forward. Domestic

migration between city and country brings internal

flow of remittance (Feng Hu, Qiwen Wang , 2007).

Aimed at these questions, this text will be based

on theoretical analysis, taking advantage of cross-

sectional data of CHIPS2008, using whether

remitting and remittance as the index of their

remittances, to investigate the influence remittance

makes to their subjective well-being. Finally, we

find that remitting acts of peasant workers will

lightly reduce their subjective well-being.

This text might have made following

innovations: First, studies before mainly focus on

the correlations between income and subjective

well-being, there hasn’t been any study about

peasant workers’ remittances and their subjective

well-being in China yet. So, focusing on this, the

text is to express a glimpse of this issue. Second, this

text tries to overcome the endogenous bias by using

a new instrumental variable, providing valuable

references for similar studies from now on. Third,

this text deepens our understanding of the standard

of human well-being from the view of remittance,

contributing to solve the problem how we increase

peasant workers’ happiness by remittance.

2 MANUSCRIPT PREPARATION

2.1 Literature Review

Rise of studies about subjective well-being benefits

from life-standard research campaign and positive

psychology campaign. Many scholars discuss the

elements that affect subjective well-being from

different angles, such as some micro-variables like

sex, age, schooling, income and marriage. Along

them, income plays an important part in happiness

economics in nowadays economic society:

According to economic rationally hypothesis,

there’s a positive correlation between subjective

Liu M., Chen J. and Zang D.

Research on the Relationship between Remittance and Subjective Well-Bing - Empirical Analysis based on CHIPS Micro Data.

DOI: 10.5220/0006443000190024

In ISME 2016 - Information Science and Management Engineering IV (ISME 2016), pages 19-24

ISBN: 978-989-758-208-0

well-being and income. But in fact, in countries

which have different degrees of development, the

relationship between subjective well-being and

income is totally different. In the opinion from

Easterlin, the correlation between individual

happiness and his income is positive, but the

presence of Critical Point Theory makes it

impossible for a country’s average happiness

increase with the rise of income (Feng Hu , Yupeng

Shi , 2013). Besides, the effect of absolute income

and income anticipation are also remarkable in

subjective well-being (Chuliang Luo ,

2008) ,inequality in income gives negative impacts

to subjective well-being. Furthermore, household

debts also affect subjective well-being a lot (Frey，

B., A. Stutzer.,2002).

Although there are a lot of studies talking about

subjective well-being, faced with a large scale of

remittance from China’s migrant workers, empirical

studies about remittance from migrant workers and

their subjective well-being are only few. Alpaslan

Akay and others find that remittances have positive

relation with migrants’ subjective well-being based

on research theories before, and this is the first time

that academia develops researches on correlations

between migrants’ remittance and their subjective

well-being (Blanchower, D. and A. Oswald ,

2004).To sum up, few foreign learners set out to

analyze it, but no studies related at home. According

to this, this text is based on qualitative and

quantitative analysis, combining with the facts of

our own country, to study from ways like what

peasant workers’ remittance affects their subjective

well-being and how it can increase their subjective

well-being.

2.2 Data Sources and Model Building

2.2.1 Data Sources

The data this text uses are from a micro-research of

CHIPS(2008). This research involved 15 cities in 9

provinces, including 5000 samples from urban

families, 8000 samples of rural families and 5000

migrant samples. According to our aim, this text

chooses samples of urban and rural immigrants in

CHIPS (2008) database. These samples are

representative, fully reflecting the basic situation

such as statistical characteristics, income and

outcome, insurance and guarantee, and family

business and life.

2.2.2 Model Building and Equations

In order to study the impact from peasant workers’

remittances to their subjective well-being, we design

following basic metering regression model in the

light of researchers before:

iiii RYXαSWBi

+= 3210

In model 1, explained variable SWBi shows the

control variables set of the individual part of the ith

peasant worker, including: sex (to man as reference

group), age and its square value, quantity of children,

marital status (to married as reference group),

schooling, employment status (to employed as

reference group). Explaining variable Yi shows the

ith worker’s family financial situation, including:

peasant workers’ monthly income and family debts;

, α

are coefficients or coefficient

vectors of solve-for parameter, εi is the random

disturbance term of regression model.

2.3 Representation of Peasant

Workers’ Subjective Well-Being

The measurement of subjective well-being mainly

includes three methods of happiness, life-satisfaction

and mental health, 12-item General Health

Questionnaire (GHQ－12) can effectively evaluate

subjective well-being (Fang Luo , 2007). In

accordance with experience, we adopt quartering as

our method (as 0-1-2-3 point score, its total score

ranges from 0 to 36) to get the index of peasant

workers ’ subjective well-being.Table 1 is the

distribution of 3368 samples’ subjective well-being

index. In general, migrant workers ’ subjective

well-being distributes rather closely, more

subjectively happy they are, more they distribute.

2.4 Representation of Social and

Economic Characteristics of

Migrant Workers

According to previous studies, migrant workers’

individual characteristic variable and family

financial variable affect their subjective well-being

to some extent. Besides, this article adopts the

months that migrant workers go out for work and

ISME 2016 - Information Science and Management Engineering IV

ISME 2016 - International Conference on Information System and Management Engineering

business a year as instrumental variable to cope with

endogenous problems; we add two variables of

average remittance and relative remittance for

robustness examination. We follow different

situations of overall migrant workers, migrant

workers who have remittances and who don’t have

remittances, and make descriptions for these

samples’ social and economic characteristic

variables respectively,table 2 shows as follows.

The remittance of main explaining variable

indicates the monthly remittance standard of migrant

worker’s family. At the same time, we have

controlled other explaining variables which are

proved to possibly affect happiness in previous

studies: (1) Age. Impacts that age makes on

subjective well-being usually shows an U type

characteristic (Yuanping Lu, Tao Wang , 2011). (2)

Marital status. Usually, those who have been

married are more satisfied with their life (Yanhua

Zhang , Yu Jing, Chang Bo , 2005). (3) Quantity of

children. Due to the time cost and money cost they

need to pay, quantity of children has a tight

correlation with their subjective well-being. (4)

Gender. Because men always take more family

responsibility, women usually have higher

subjective well-being. (5) Education year. We can

increase migrant workers’ subjective well-being by

increasing the time for them to have education. (6)

Employment status. Employment status plays an

important part in migrant workers’ subjective well-

being. (7) Labor income. Labor income has positive

correlation with migrant workers’ subjective well-

being. (8) Family debts. Families in higher debts

tend to have better expectation to the future and

higher subjective well-being (Alpaslan Akay , 2012).

2.5 Tables

Table 1: subjective well-being distribution of all

groups.

subjective well-being index Sample size Proportion（%）

0 2 0.06

1-6 2 0.06

7-12 13 0.39

13-18 93 2.76

19-24 579 17.19

25-30 1499 44.51

31-36 1180 35.04

Total 3368 100

Table 2: Descriptive Analyze for Main Variables.

Variables all remitting non-remitting

AVE SD AVE SD AVE SD

Age 31.27 10.54 31.22 10.43 31.35 10.72

Sq-age 1088.9 761.6 1083.4 749.0 1097.9 782.15

Gender 0.62 0.49 0.63 0.48 0.61 0.49

Marital 0.58 0.49 0.58 0.49 0.57 0.49

Edu-year 7.97 3.98 7.93 3.99 8.04 3.95

Employ 0.96 0.20 0.96 0.19 0.96 0.20

Kid 0.88 0.96 0.88 0.96 0.87 0.96

Income 2415.5 1560.7 2679.4 1737.5 1981.2 1083.9

debts 285.2 1923.0 122.6 880.1 168.4 1095.9

Annotation：①Gender:0 for women，1 for men；

②Marital status:0 for single ， 1 for married ；

③Employment status:0 for unemployed ， 1 for

employed ；④0 for non-remittance ， 1 for

remittance ；⑤Labor income, family debts and

remittance are monthly amount ；⑥Average

remittance uses amendatory OECD equivalence

scale: remittance/ （ 1+0.5*adults+0.3*juveniles ，

relative remittance=（remittance/labor income）

3 ANALYZE OF EMPIRICAL

RESULTS

3.1 The Benchmark Regression Results

Because subjective well-being is a non-negative

continuous variable, we first adopt Tobit method to

evaluate econometric model. The benchmark

regression results are in Table3.The first line only

considers the influence remittance make to migrant

workers subjective well-being. Without bringing in

control variable, remittances reduce migrant

workers’ subjective well-being dramatically but with

low extent. Regression equation (2) considers all

variables including remittance and dramatic but

weak negative effects that remittances bring to

migrant workers’ subjective well-being. Therefore,

whether bringing in control variables or not,

remittance standard can make prominent effects to

migrant workers’ subjective well-being.

The third line doesn’t think about family

remittances, studying about what individual

characteristics of migrant workers and family

financial characteristic variable bring to subjective

well-being. In terms of individual characteristics of

migrant workers: Regression coefficients of age and

square of age are dramatically respectively negative

Research on the Relationship between Remittance and Subjective Well-Bing - Empirical Analysis based on CHIPS Micro

Data

Research on the Relationship between Remittance and Subjective Well-Bing - Empirical Analysis based on CHIPS Micro Data

and positive under 1% statistical level, indicating

that migrant workers’ subjective well-being firstly

descend and then rise up with age in our country;

Men’s regression coefficient is dramatically

positive, showing that men are happier than women;

Results in marital status show that those who have

been married are significantly happier than those

who haven’t got married; the unemployed migrant

workers dramatically unhappier than the employed.

Whether there are any children in the family makes

no difference to migrant workers’ subjective well-

being; Under 5% statistical level, the regression

coefficient of education year is significant negative,

indicating that longer they are educated, lower their

happiness is. In terms of family: Family debts make

weak but significant positive effect to subjective

well-being. The effects labor income makes are

positive, but the marginal effects brought by

remittance are more than those brought by income.

Furthermore, the forth, fifth and sixth lines all

adopt Ologit model, their results are similar to those

using Tobit, remittance variable is still significant.

Table 3: Empirical results of how remittances affect

subjective well-being.

Var (1) (2) (3) (4) (5) (6) (7) (8) (9)

Tobit Tobit Tobit OLogit OLogit OLogit IVTobit 1st stage 2SLS

Remit -0.000**-0.000** -0.000** -0.017 -0.017

tance

(0.000) (0.000) （0.000）（0.000）（0.013）（0.012）

Income 0.000*** 0.000*** 0.000** 0.000** 0.000 -0.003 0.000

(0.000) (0.000) (0.000) (0.000) (0.000) (0.003) (0.000)

Family 0.000** 0.000** 0.000* 0.000* 0.000* 0.013* 0.000*

debts

（0.000） (0.000) （0.000）（0.000）（0.000）（0.008）（0.000）

Age -0.040*** -0.040*** -0.018 -0.019 0.039 6.725 0.047

(0.011) (0.011) （0.019）（0.019）（0.130）（4.829）（0.129）

Square 0.001*** 0.001*** 0.000 0.000 -0.001 -0.092 -0.001

of age (0.000) (0.000) （0.000）（0.000）（0.002）（0.0621) （0.002）

Gender 1.431*** 1.428*** 0.524*** 0.523*** 1.568*** 7.805 1.511***

(0.181) (0.181) （0.064）（0.064） (0.335) （15.679) （0.323）

Quantity -0.111 -0.120 -0.048 -0.051 0.126 2.529 0.119

of children

(0.159) (0.159) （0.055）（0.055） (0.163) （7.852）（0.157）

Educati -0.050** -0.050** -0.013* -0.012 -0.133* -5.135*** -0.129*

on year

(0.022) (0.022) （0.008）（0.008） (0.077) （1.878）（0.074）

Employ 1.578*** 1.626*** 0.439** 0.457** -0.383 -116.758*** -0.372

Status

(0.52) (0.519) （0.192）（0.191） (1.753) （44.776）（1.693）

Marital 0.402* 0.394* 0.187* 0.188* 0.555 4.085 0.505

status

(0.225) (0.225) （0.101）（0.101） (0.451) （21.964）（0.436）

Inter 28.48** -27.25*** 27.08*** 33.05*** 21.78*** 32.64***

cept

（0.104） (0.622) (0.616) (4.506) （91.074）（4.352）

R2 0.000 0.005 0.005 0.000 0.006 0.005 0.007 0.000

AR 0.004***

Wald test 35.43*** 34.47***

N 3368 3368 3368 3368 3368 3368 3297 3297 3297

Annotation: ①* 、 ** 、 ***respectively show

significant level of 10%, 5% and 1%②Wald test

indicates exogenous exemplary inspection.

3.2 The Endogenous Processing

Accessing equation (1) by using Tobit model will

lead to key parameter’s biased estimation. In order

to amend endogenous errors, we decide to use

instrumental variable LM, it shows the total amount

of months migrant workers going out for work and

business last year. It will straightly affect the

frequency they remit and then affect their amount of

remittances, but make no difference to their

subjective well-being.

Using this instrumental variable, we adopt the

IVTobit model, its results are in Table3, Line 7; at

the same time, we choose 2SLS accessing model and

its report of results is in Table3, line 8 and line 9.

Line 7 and line 9 in Table3 show that all Wald

exogenous exemplary inspections have rejected to

use remittance as exogenous variable ’ s original

assumption, that is to say the remittance is internal

variable. Line 8 in Table3 shows that robustness

examinations of weak instrumental variable have

turned the original assumption down and indicates

that there’s no problem about weak instrumental

variable. Therefore, there is neither problem of weak

instrumental variable nor endogenous problem in

instrumental variable we use. It turns out to be that

remittance still performs significantly in subjective

well-being.

3.3 Robustness Problems

In order to examine the reliability of these results,

we conduct following robustness examinations.

Firstly, we use alternative instrumental variables of

migrant workers’ remittances to make regression

for benchmark model again; Secondly, we consider

whether bringing into new variables will make any

difference to our results. The examination results

will be sorted out and then reported in Table4.

(1)Seeking for alternative instrumental variables

for migrant workers’ remittances. In the robustness

examination, we respectively adopt regression of

average remittance of migrant workers, per-capita

remittance and a dummy variable of whether there is

a remittance as alternative variable for migrant

workers ’ remittance. In Table4, when we

respectively use remittance, average remittance and

relative remittance to perform regression, the rise of

relative remittance affects migrant workers’

ISME 2016 - Information Science and Management Engineering IV

ISME 2016 - International Conference on Information System and Management Engineering

happiness much more than absolute remittance and

per-capita remittance. Overall, remittance level

really makes a significant difference to migrant

workers’ subjective well-being, the major

conclusions are still correct.

(2)Bringing into the medical expenditure control

variables. We bring into the medical expenditure as

control variables to regress to benchmark model

again and its results are reported in Table4, line 2.

It’s observed that migrant workers’ remittance will

make weak and steady negative effects on their

subjective well-being, and that medical expenditure

variables make no difference to their remittance.

Therefore, it’s plain to see, our model is robust.

Table 3: Explanation results of robustness

Variables (1)

Ologit

(2)

Ologit

Remittance

Medical

expenditure

Other

variables

Pseudo R2

Per-capita

remittance

Medical

expenditure

Other

variables

Pseudo R2

Relative

remittance

Medical

expenditure

Other

variables

Pseudo R2

-0.0001

(0.0000)

Already controlled

3368

0.0002

-0.0003**

(0.0001)

Already controlled

3368

0.0002

-0.2838***

(0.0819)

Already controlled

3368

0.0006

-0.0001

(0.0001)

-0.0000

(0.0000)

Already controlled

3368

0.0013

-0.0003*

(0.0001)

-0.0000*

(0.0000)

Already controlled

3368

0.0013

-0.2718***

(0.0833)

-0.0000*

(0.0000)

Already controlled

3368

0.0017

Annotation: * 、 ** 、 ***respectively show

significant level of 10%, 5% and 1%.

3.4 Equations

iiii RYXαSWBi

++++= 3210

(1)

4 COPYRIGHT FORM

Sichuan Agricultural University , Dungang Zang

5 CONCLUSIONS AND POLICY

PROPOSALS

Studies find that remittances from migrant workers

make negative effects to their subjective well-being.

Remittances will shortly reduce their subjective

well-being; but the marginal effect to happiness of

their remittances is far greater than income effect,

the marginal effect to their happiness of relative

remittance is larger than the marginal effect of per-

capita remittance. Therefore, we think remittances

especially relative remittances are is an important

way to improve migrant workers’ welfare. Relative

results are still correct when considering endogenous

problems.

After the financial crisis in 2008, it’s a long term

for us to adjust our imbalance of economic structure

and transition of urbanization. To some degree, after

we lowered our annual economic growth target, the

pressure of employment increased among migrant

workers. In order to protect the interests of migrant

workers, we should start in reducing remittances and

increasing income to promote increasing in

happiness, and this will bring them more positive

sentiments. (De Neve et al., 2013 ，Gielen et al.,

2014 ， ShuLi ， 2015)We have following policy

proposals for this:

First, we should properly organize the rural

system of social security. The rural guarantee system

construction in our country occurs to be imbalanced.

Because of altruism motivation, migrant workers

will send more income back home to keep family’s

basic life. Therefore, while social security system is

all-covered, government also needs to design a

system based on facts from different areas at the

same time to ensure we can really make “cover the

basic”, “multi layers” come true.

Second, we should provide service sources for

migrant workers to go back to start a business. On

one hand, the government should contribute to

realizing the equality of urban and rural fundamental

public services; on the other hand, government

should positively build incubator for migrant

workers to make a business.

REFERENCES

Alpaslan Akay,Corrado Giulietti, D.Robalino, Klaus

F.Zimmermann. , J., 2012. In IZA DISCUSSION ,

Remittances and Well-Being among Rural-to-Urban

Migrants in China.ELSEVIERPRESS.

Research on the Relationship between Remittance and Subjective Well-Bing - Empirical Analysis based on CHIPS Micro

Data

Research on the Relationship between Remittance and Subjective Well-Bing - Empirical Analysis based on CHIPS Micro Data

Blanchower, D. and A. Oswald, J., 2004. In JOURNAL

OF PUBLIC ECONOMICS,Well-being over time in

Britain and the USA.ELSEVIERPRESS.

Chuliang Luo, J., 2008. In JOURNAL OF FINANCE AND

ECONOMIC, Absolute income, relative income and

subjective well-being: Empirical analysis of China’s

urban and rural household survey data from

China .FINANCE AND ECONOMIC PRESS.

Fang Luo, J., 2007.In CHINA’S RURAL ECONOMY,

Analisis of factors and incentive about immigrant’s

remittance:Take WuHan as example.SICHUAN

SOCIETY AND SCIENCE PRESS.

Feng Hu , Yupeng Shi, J., 2013. In WORLD ECONOMIC,

Migrant Workers’ Remittances and Economic

Development in Exporting areas:Analysis of Factors

Based on Using of Migrant Workers’

Remittances.SOCIETY AND SCIENCE PRESS.

FengHu, Qiwen Wang, J., 2007. In STATISTIC

RESEARCH, Analysis of Factors that Affecting

Remittances of Migrant Workers in China: A Use of

Section Regression Model. NATIONAL STATISTIC

PRESS.

Frey ， B., A. Stutzer., J., 2002. In JOURNAL OF

ECONOMIC LITERATURE, What can economists

learn from happiness research? ECONOMICPRESS.

Shu Li , Gang Chen, J., 2015.In ECONOMIC RESEARCH,

Employment effects of subjective well-beining:The

empirical research on happiness, employment and

implicit re-employment. ECONOMIC RESEARCH

PRESS.

Yuanping Lu, Tao Wang, J., 2011. In ECONOMICS.

Social crimes and Social well-being----Empirical

evidence from China. ECONOMICPRESS.

Yanhua Zhang , Yu Jing, Chang Bo, J., 2005. In

JOURNAL OF JISHOU UNIVERSITY, Analysis of

factors that Affect Urban People’s subjective well-

being---- Take Shenyang Urban People for Instance.

JISHOU UNIVERSITY PRESS.

ISME 2016 - Information Science and Management Engineering IV

ISME 2016 - International Conference on Information System and Management Engineering