Journal of Integrative Agriculture 2021, 20(4): 998–1011

Available online at [Link]

ScienceDirect

RESEARCH ARTICLE

Poverty alleviation through e-commerce: Village involvement and

demonstration policies in rural China

PENG Chao1*, MA Biao2*, ZHANG Chen3

1
Administration and Management Institute, Ministry of Agriculture and Rural Affairs, Beijing 102208, [Link]
2
School of Agricultural Economics and Rural Development, Renmin University of China, Beijing 100872, [Link]
3
Institute of Population and Labor Economics, Chinese Academy of Social Sciences, Beijing 100006, [Link]

Abstract
The diffusion of e-commerce has played a significant role in recent rural economic development in China. E-commerce
is also considered as an efficient channel to alleviate poverty in rural China. Voluminous studies have investigated the
contribution of e-commerce to agricultural development, yet it is lacking empirical evidence as to the effects of e-commerce
on rural poverty alleviation. Since the year of 2014, in order to develop rural e-commerce, Chinese government launched
the National Rural E-commerce Comprehensive Demonstration Project. This gradual involvement policy offered a natural
experiment for evaluation of e-commerce. Based on village-level survey data from rural China and Heckit method, our study
finds that rural e-commerce has a significantly positive effect on rural income. Moreover, the effect is inverted U-shaped for
the relative-poverty villages. The estimation of the propensity scores matching model confirms that the results are robust.
The following policy recommendations are proposed: (1) policy support to rural e-commerce should prioritize the poverty-
stricken villages. By doing so, the marginal income effects of e-commerce will be maximized. (2) Investment in internet
infrastructure and establishment of human resources for e-commerce in rural areas will have spillover effects, increasing
rural income through the “digital dividend”.

Keywords: poverty alleviation, income, National Rural E-commerce Comprehensive Demonstration Project, Heckit method

achieved the Millennium Development Goal for poverty

alleviation set by the United Nations (Liu et al. 2019).
1. Introduction Although the Chinese government has prioritized policies
that target at improving rural residents’ welfare, poverty
Since 1978, China has lifted more than 700 million people remains a critical problem with 16.6 million people estimated
out of poverty. China was the first developing country that to live in poverty and an poverty incidence rate of 1.7% in
2018 (NBSC 2019). Poverty alleviation remains a massive
task for the Chinese government, which is determined to
Received 2 June, 2020 Accepted 16 September, 2020 eradicate absolute poverty by 2020.
Correspondence PENG Chao, E-mail: 15801698196@[Link]; Thanks to the development of information technology and
MA Biao, E-mail: mabiao1992@[Link]
* increasing use of smartphones in rural areas, e-commerce
These authors contributed equally to this study.
penetration in the rural economy has been on the rise in
© 2021 CAAS. Published by Elsevier B.V. This is an open
recent years. Rural e-commerce is a direct approach to take
access article under the CC BY-NC-ND license (http://
[Link]/licenses/by-nc-nd/4.0/). advantage of information technology and achieve higher
doi: 10.1016/S2095-3119(20)63422-0 profits. It adapts to the requirements of rural industrial
 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011 999

transformation and upgrade. By doing so, it provides new e-commerce affecting income? According to the literature,
drivers for the revitalization of rural industry. In recent rural producers could overcome the geographical limitation
years, “empowerment” to players of rural e-commerce has of the tangible market through e-commerce (Baorakis et al.
also attracted much attention (Ogutu et al. 2014; Leong 2002; Shimamoto et al. 2015; Goldfarb and Tucker 2019).
et al. 2016; Zeng et al. 2018). In these literature, rural Traditional sales channel has been transformed via rural
e-commerce was considered as an efficient tool to enable e-commerce by integrating agricultural products’ supply and
people to develop modern agriculture and commerce. marketing information. Especially, it offers a “bridge” for poor
Especially, e-commerce promoted agricultural value chain people who used to be excluded from the marketing channel,
development and commodity circulation (Baorakis et al. but now could make their production and sales decisions
2002; Mintert 2003; Jalali et al. 2011; Parker et al. 2016; according to market demand. By doing so, poor people
Song et al. 2019). E-commerce, unlike traditional commerce could gain bargaining power. Moreover, rural e-commerce
in which employees need to be present in stores, enables its can improve poor people’s human capital (Clark and Gorski
staff to work from home. It is a highly accessible form of off- 2002; Fafchamps and Minten 2012; Chan 2015). In general,
farm work for those constrained in rural areas due to either rural e-commerce practitioners need to be trained. Some
agricultural production or caring responsibilities. In some talented poor people can use an e-commerce platform to
rural areas, e-commerce has become a potential engine for operate their online stores and more farmers can engage in
farmers’ income growth. As an innovative solution, poverty e-commerce related works after being trained. Thirdly, rural
alleviation through e-commerce began to emerge. e-commerce can motivate poor people to fight out of poverty.
A large body of literature has confirmed the positive role of Poor people can access information from the world outside
rural e-commerce in achieving online commodity exchanges via the internet (Leong et al. 2016; Leng et al. 2020). Some
and increasing rural household income (Akca et al. 2007; rural residents from remote and poor villages have become
Tao and Zhao 2012; Jha et al. 2016; Leong et al. 2016; Gao internet celebrities by selling the products or showcasing the
et al. 2018; Li et al. 2019). China’s practical experience cultures of their hometowns in recent years.
has also proved that rural e-commerce has a positive impact This paper aims to examine the impact of rural
on farmers’ income. In 2018, the amount of online retail e-commerce on household income. We employ National
sales in state-level poverty-stricken counties reached 69.79 Fixed-site Rural Survey (NFRS) village-level data conducted
billion CNY, a year-on-year increase of 36.4%. The growth by the Ministry of Agriculture and Rural Affairs of China in
rate was 6% higher than that of overall rural online retail 2018. The paper extends the previous literature on rural
sales and 12.5% higher than that of national online retail e-commerce and farmers’ income in three main areas.
sales.1 More than 28 million people were employed in rural First, we provide one of the first studies examining the
e-commerce in 2018 and nearly three million “filed-on-card” effect of rural e-commerce on household income by using
poor households benefited from rural e-commerce (Ministry village-level data in rural China. Second, considering
of Commerce of China 2019).2 Given the solid evidence of the heterogeneity among different types of villages, we
the positive effects of e-commerce on rural industries and analyze the impact in relative poverty villages. Third,
rural employment, we assume that rural e-commerce as unlike previous studies focusing on the access to internet
well plays a significant role in improving the income of poor or smartphones, we measure e-commerce participation of
rural households. villages as their enrolment in the National Rural E-commerce
In this paper, we investigate whether rural people’s Comprehensive Demonstration Project (NRECDP).
involvement in e-commerce could improve their income.
Access to information can increase income in rural areas, 2. Background
especially poor villages (Acker and Mbiti 2010; Hartje
and Hübler 2017). Rural e-commerce theoretically could 2.1. E-commerce in rural China
play a positive role in poverty alleviation by introducing
new income sources. What is the mechanism of rural E-commerce is becoming one of the most mesmerizing

1
In order to support poverty-stricken areas, Chinese government has set up a national standard for poverty-stricken counties. Those
state-level poverty-stricken counties have been recognized by the State Council Leading Group Office of Poverty Alleviation and
Development.
2
Chinese government has filed rural households in absolute poverty on card. The national rural poverty alleviation standard of 2 736 CNY
per capita net income of farmers in 2013 is taken as the identification standard. On the basis of ensuring the completion of the task
of identifying the national rural poverty alleviation standards, the provincial governments carry out the work of identifying the poor
households according to the provincial standards, and incorporate it into the national poverty alleviation information network system for
unified management.
1000 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011

industries in China. Since the first decade of the 21st a year-on-year increase of 33.8%, 9.9 percentage points
century, China’s internet economy has made vast progress higher than the national online retail sales, accounting for
in technology, business form, application and cross- 21.1% of the national online retail sales of rural material
border integration. Rural e-commerce has experienced goods. Among them, the retail sales of leisure food, tea,
profound changes. Since 2014, China’s overarching act and nutrients ranked the top 3, accounting for 24.2, 12.5,
on agriculture, rural areas and farmers, known as No. 1 and 12.0% of the total online retail sales of agricultural
Central Document, has produced specific policies for the products, respectively. Fruits, seasoning, and bean products
development of rural e-commerce. In 2019, the No. 1 ranked the top 3 most rapidly growing commodities, with a
document clearly stated that China will implement the rate of 45.6, 41.9 and 41.3%, respectively (Fig. 2). A recent
“Digital Rural Strategy”. As a result, e-commerce has been successful story of rural e-commerce development was that
introduced to broader rural areas. Agricultural products during the Chinese Farmers’ Harvest Festival (CFHF)4 on
have increasingly been sold out of towns and villages September 23, 2018, sales of agricultural products on major
through e-commerce. By the end of 2019, 1 231 counties e-commerce platforms exceeded 20 billion CNY.
in 28 provinces/autonomous regions have been funded to E-commerce has played an important role in poverty
enroll in the NRECDP. Among those NRECDP member alleviation in poverty-stricken counties as agricultural
counties, 875 are national poverty-stricken counties and products could be introduced to urban consumers directly
189 are severely impoverished ones. by rural e-commerce. Other non-agricultural products could
It is suggested that national rural online retail sales also be promoted by e-commerce. In 2017, products worthy
reached 1.37 trillion CNY in 20183, a year-on-year increase of 120.79 billion from poverty-stricken counties were sold
of 30.4 and 6.5 percentage points higher than the growth online. It increased by 52.1% year-on-year and was 13%
rate of the national online retail, accounting for 15.2% of the higher than rural online retail. In the first 11 months of 2019,
national online retail (Fig. 1). the online retail sales of 832 poverty-stricken counties was
Specifically, the online retail volume of rural material 216.61 billion CNY. The year-on-year increase was 31.7%.
goods was 1.09 trillion CNY, a year-on-year increase of The trend of online retail from the national poverty-stricken
30.9%, accounting for 80% of the national rural online retail counties was shown in Fig. 3.
sales. The online retail volume of rural non-material goods
was 273.31 billion CNY in 2018, a year-on-year increase 2.2. Anti-poverty in China
of 28.4%, accounting for 20% of the national rural online
retail sales. E-commerce also shows a rapid growth in Over the past 40 years, China has made significant
agricultural product retail. The national online retail sales progress towards its poverty alleviation goals. Most rural
of agricultural products reached 230.5 billion CNY in 2018, households have gradually fought out of absolute poverty.

Rural network retail sales Year-on-year growth rate

16 000 160.0
153.4%
Year-on-year growth rate (%)
Rural network retail sales

13 679.4
14 000 12 448.8 140.0
(100 million CNY)

12 000 120.0
10 000 96.1% 8 945.4 100.0
8 000 80.0
6 000 60.0
3 530 39.1%
4 000 40.0
1 800 30.4%
2 000 20.0
0 0
2014 2015 2016 2017 2018

Fig. 1 National rural online retail sale from 2014 to 2018. Source: Ministry of Commerce of China, 2019.

3
In 2018, the Ministry of Commerce of China adjusted the statistical monitoring objects, scope and methods of online retail of rural and
agricultural products, and the year-on-year growth rate was calculated according to the comparable caliber.
4
Chinese Farmers’ Harvest Festival, the first in China created especially for farmers, reflects the high priority the central government
places on agriculture, rural areas and rural people, as well as its determination to see a thriving agricultural industry, beautiful
countryside and well-off farmers.
 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011 1001

Proportion of retail sales (%) Year-on-year growth rate (%)

(%)
50
45 45.6 41.9
40 39.6 33.9 41.3
30.5 38.3
35 32.5
30 29.0 34.3 33.3
25 24.2
20 12.5 12.0
15 10.2 9.6 9.1 8.4
10 6.0
3.2 3.1 1.7
5
0
od

od

ts

gs

ts
ilk
oi
ui

le
Te

in
uc

uc
M
Fr

eg
fo

fo

ab
on
od

an

od
e

et
as

nd
ur

ris

pr

pr
in

g
Se

ta
is

Ve
ou

ra
tic

an
Le

ea
G
N

ua

Be
M
Aq

Fig. 2 The proportion of retail sales and year-on-year growth rate of agricultural products in 2018. Source: Ministry of Commerce
of China, 2019.

(Billion CNY) 331 major national aid-the-poor counties. After Comrade

250.00 Deng Xiaoping’s speech on the Southern Patrol in 1992,
216.61
200.00
China’s economy entered a new era of high-speed growth.
164.47 Economic growth became the main driving force of poverty
150.00 120.79 reduction in this stage of economic development. However,
100.00 79.41 rapid economic growth inevitably brought about unbalanced
regional development, and the poor population further
50.00
congregated in the underdeveloped areas. In response,
0 China began to implement the Seven-Year Priority Poverty
2016 2017 Jan.–Nov. Jan.–Nov.
2018 2019 Alleviation Program (1994–2000) in 1994, prioritizing the
problems of food and clothing for the poor in underdeveloped
Fig. 3 The online retail from national poverty-stricken counties.
areas. Six years after implementing the Program, China’s
Source: Ministry of Commerce of China, 2019.
absolute poverty-stricken population dropped to 32 million.
Despite these remarkable achievements, a new challenge
Under the current poverty line, the rural poor population has emerged: many of the remaining poor are distributed
has decreased by 739.9 million (Liu et al. 2019). The in sparse counties that are not supported by the previous
World Bank acknowledges that China has significantly anti-poverty programs.
progressed towards inclusive and sustainable development In response to the new challenge, the government has
that alleviates poverty. An extensive review of China’s great issued the Outline for China’s Rural Poverty Alleviation and
achievements in anti-poverty was provided below. Development (2001–2010), which relaxed the restrictions
Before 1978, the size of China’s poor population was on using poverty alleviation funds, extending to cover poor
about 250 million. The large scale of poverty was due to low people not living in the national poverty-stricken counties. In
rural productivity. With the reform of the basic household 2011, the government issued the Outline for Development-
management system and the commodity circulation system oriented Poverty Alleviation for China’s Rural Areas (2011–
in rural areas, rural productivity has been increasing 2020), which identified 11 consistently poor areas and three
considerably. Agricultural outputs and farmers’ income main battlegrounds for poverty alleviation, including Tibet,
have rapidly increased. Accordingly, the size of rural the Tibetan area in four provinces and southern Xinjiang.
poor population has greatly decreased. The remaining From 2014, the government started to identify the villages in
poor population usually geographically concentrate in the poverty and integrate them into the unified management of
“old, small, border and poor” areas. In those areas, it is the national poverty alleviation information network system.
difficult for the poor to get out of poverty effectively without According to the poverty standard in 2010, the number of
substantial economic growth. The Chinese government people in rural poverty decreased from 661.0 million in
has started a series of large-scale organized rural poverty- 1985 to 16.6 million in 2018, and the incidence of poverty
alleviation projects since 1986. These projects formulated decreased from 78.3 to 1.7%. China is on track to meet
a “county-level targeting” mechanism and established its goal of eradicating absolute poverty by the end of 2020.
1002 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011

During the new stage, e-commerce could contribute to Commerce, and other ministries jointly issued an act on the
poverty alleviation in three aspects. First, poverty alleviation NRECDP, in which the proportion of national poverty-stricken
through e-commerce has been ready in both infrastructure counties must not be less than 50% of the enrolled counties.
and business type, as e-commerce has developed rapidly Table  1 reports the integrated demonstration project on
in urban areas and extended to rural areas in some poverty e-commerce in rural areas from 2014 to 2019. The funding
regions. Second, e-commerce platforms can standardize on rural e-commerce has continued to increase since 2017.
the agricultural products from the specific poverty-stricken In 2017, the No. 1 Central Document set up a special request
areas that are often unique and small in volume, and then for local governments on “promoting the development of
transform them into hot sales. This helps to target poverty rural e-commerce”, which encouraged the development of
issues aggregated in some specific areas. Third, the new e-commerce industry parks in local communities.
poverty-alleviation strategy seeks a transition from “blood According to the documents issued by the Ministry
transfusion” to “hematopoiesis”, underpinning by the of Finance, the Ministry of Commerce, and the Poverty
philosophy of development-oriented poverty alleviation. Alleviation Office of the State Council in 2018, the focus
In this regard, e-commerce is considered an innovative of the NRECDP is to strengthen the ability to establish
industry to promote hematopoiesis through building human e-commerce for poverty-stricken villages and households
capital and igniting people’s enthusiasm of anti-poverty. in rural areas. Especially, the deeply poverty-stricken areas
will be supported by transferring payment from the central
2.3. E-commerce as anti-poverty toolkit government. Other specific measures include expanding
the marketing channels of rural products, encouraging
Chinese government’s policies have promoted rural more rural residents to participate in rural e-commerce,
e-commerce as an important means of targeted poverty and facilitating poor villages in taking advantage of their
alleviation. In recent years, the rural industrial value chain resource endowment potential. One of the measurable
has been promoted by rural e-commerce. E-commerce is objectives of the NRECDP is that the absolute poverty-
an optimal channel for the circulation of rural commodity. stricken villages’ e-commerce service coverage rate
As mentioned above, rural e-commerce also provides a new reaches 50%. The overall administrative villages are also
form of employment for rural people. According to various required to achieve this goal. Every county involved in the
policy documents, the key areas supported by e-commerce NRECDP would receive 20 million CNY from the Ministry
poverty alleviation are poor counties. Since the Ministry of Finance. The transfer payment should be invested in
of Finance and the Ministry of Commerce jointly launched specific projects that promote rural e-commerce, including
the NRECDP in 2014, 875 state-level poor counties have promoting rural product marketing, improving rural public
been successively listed as NRECDP member counties. By service system, and training rural e-business personnel.
2019, the NRECDP will achieve a coverage of all state-level To introduce rural products to urban areas, the NRECDP
poor counties. supports the production, grading, packaging and marketing
Early focus of the NRECDP was agricultural product of rural products. The NRECDP supports the construction of
sales. It did not explicitly refer to e-commerce poverty e-commerce public service centers and sites. Agglomeration
alleviation. But the project in the pilot stage led to the effect may be achieved by concentrating logistics, training
local e-commerce infrastructure construction and service and poverty alleviation means in rural areas. With the
improvement. In the NRECDP counties, a group of rural support from the NRECDP, farmers who have the intention
e-commerce leaders have been trained. In 2016, the to engage in e-commerce would be trained.
State Council Leading Group Office of Poverty Alleviation A text analysis of government documents summarize
and Development, the Ministry of Finance, the Ministry of the following ways in which e-commerce contributes to

Table 1 E-commerce into rural comprehensive demonstration counties from 2014 to 2019
Number of national poverty-stricken Proportion of national poverty-stricken
Year Number of demonstration counties
counties counties (%)
2014 56 14 25.00
2015 200 90 45.00
2016 240 158 65.83
2017 260 237 91.15
2018 260 238 91.54
2019 215 138 64.19
Data source: China Electronic Commerce Report (2014–2019), Ministry of Commerce of China ([Link]
 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011 1003

poverty alleviation. First, e-commerce is an effective anti- variables have been transformed and updated according to
poverty toolkit. This is because e-commerce overcomes the latest development in rural China in recent years. As
the geographical constraints encountered in the uplink a result, new variables related to e-commerce have been
of traditional agricultural products and promotes the included in the new survey. However, although NFRS is a
production and marketing of rural products in poor areas. longitudinal survey, the new survey in 2018 only provided
The government is also increasingly paying attention to the cross-sectional information as of 2018, covering 371 villages
role of e-commerce as a new toolkit in public services such nationwide. Sample distribution by province and time of
as rural education and medical care. It is predicted that participation in the NRECDP are shown in Table 2.
e-commerce poverty alleviation will shift from commodity
circulation to social and livelihood services. Moreover, 3.2. Key variables
e-commerce poverty alleviation is regarded as an essential
part of the country’s “Internet Plus” action. China’s “Internet The dependent variable of this study is per capita disposable
Plus” operation has already achieved world-renowned income at the village level and the key independent variable
achievement in urban areas. E-commerce empowerment is the implementation time of the NRECDP in the village.
is equally important to rural people, especially those in Definition of all the variables and descriptive statistics are
poverty. The e-commerce poverty alleviation policy is shown in Table 3.
oriented towards “paving the way for rural poor to share The effect of the NRECDP on income may be lagged and
the “digital dividend” and living convenience brought by carried out well into the future. For example, the villages
the Internet”. Last but not the least, e-commerce poverty may show a more obvious income increase in the second
alleviation is a part of agricultural and rural modernization. or third year of participation, and the trend may continue or
An important aspect of the e-commerce poverty alleviation reverse in the following years. So, the variable “NRECDP_
policy is extending the information infrastructure to the time” is defined as a dummy variable, with one referring to
vast rural areas, which will help lay out the foundation that a village has participated in the NRECDP for one year
and application of modern information technology all over and zero otherwise. The disposable income per capita in
China. the villages that were involved in the NRECDP is higher
than that of the non-participating villages. The population
3. Data and methods in labor force accounted for approximately 59% of the total
sample. Over 20% of the labor force had at least 9 years
3.1. Data of schooling or received professional skills training. Labor
burden coefficient refers to the ratio of the non-working-
Our dataset is derived from NFRS conducted by the Ministry age populations to the working-age populations. Our data
of Agriculture and Rural Affairs of China in 2018. NFRS has shows that, on average, the labor burden ratio is 1.22. The
been carried out annually since 1986. It covers 31 provinces, mean household cultivated land was 2.82 mu (equivalent to
municipalities and autonomous regions of Mainland China. 0.18 ha). The average agricultural subsidy per mu is 140 CNY.
In the beginning, stratified sampling was used to select In our sample, about 19% of the villages were close to the
villages in each province based on their topography, urban suburbs. The average distance from the villages to
economic development, agricultural type (agricultural area, the nearest city’s downtown was 1.82 km. The average
forest area, and pastoral area), and geographic location. As hardened road ratio was 84%, and less than 50% of villages
time passed by, some villages left the survey, because they had cold storages.
either became urban areas or their village commissions did For participation in e-commerce, we find a U-shape trend
not support the survey. When a village exited, another village with respect to the NRECDP duration. In the first year of
with similar characteristics in the same county would be the participating in the NRECDP, the average number of labor
replacement. Households were randomly selected within engaged in e-commerce were 14.89 per village. In the
villages. Besides its household-level dataset, NFRS has second year, the number decreased to 7.30.5 “Down in
accumulated a village-level dataset on production practices, the dumps” emerged in the third year, with only 3.32 labors
economics of household businesses and industries in rural involved, even less than the number in the non-NRECDP
areas. In 2018, NFRS launched a new survey that added villages. Bouncing back from the bottom, the number of
a wide range of new variables. Especially, the village-level e-commerce labor increased to 9.26 in the fourth year and

5
Labor engaged in e-commerce mainly refers to those who acquire income through e-commerce. The income should accounted for
10% or more of total household business income.
1004 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011

Table 2 The year of villages’ involvement in the National Rural E-commerce Comprehensive Demonstration Project (NRECDP)
Province, municipality Involved
All the villages Not involved
and autonomous region 2018 2017 2016 2015 2014
Beijing 3 3 0 0 0 0 0
Tianjin 6 6 0 0 0 0 0
Hebei 11 7 1 1 0 2 0
Shanxi 12 7 1 2 2 0 0
Inner Mongolia 4 4 0 0 0 0 0
Liaoning 12 12 0 0 0 0 0
Jilin 20 15 0 0 2 3 0
Heilongjiang 14 8 1 1 0 2 2
Shanghai 4 4 0 0 0 0 0
Jiangsu 11 10 0 0 0 0 1
Zhejiang 10 10 0 0 0 0 0
Anhui 17 10 0 1 2 3 1
Fujian 11 8 0 0 0 3 0
Jiangxi 11 10 0 0 0 1 0
Shandong 26 24 1 1 0 0 0
Henan 16 9 2 1 0 4 0
Hubei 15 12 1 0 0 1 1
Hunan 11 8 0 2 0 1 0
Guangdong 32 32 0 0 0 0 0
Guangxi 14 13 1 0 0 0 0
Hainan 2 2 0 0 0 0 0
Sichuan 22 17 2 0 0 3 0
Guizhou 9 4 1 1 3 0 0
Yunnan 10 8 1 1 0 0 0
Tibet 1 1 0 0 0 0 0
Chongqing 8 4 0 0 3 1 0
Shaanxi 25 12 4 2 6 1 0
Gansu 13 6 3 3 0 1 0
Qinghai 4 1 0 1 2 0 0
Ningxia 7 3 0 0 2 2 0
Xinjiang 9 5 2 1 1 0 0
Data source: The State Council Leading Group Office of Poverty Alleviation and Development, [Link].

Table 3 Definition and descriptive statistics

Variables Obs. Definition Mean (S.D.)2)
Dependent variables
ln_avrincome 353 ln(1+per capita disposable income) 9.35 (0.48)
Independent variables
NRECDP_time 371 Implementation time of the National Rural E-commerce Comprehensive Demonstration 0.72 (1.38)
Project (NRECDP) (year)
eagent 371 1 if village has rural e-commerce agents, 0 otherwise 0.30 (0.46)
edu9ratio 371 Ratio of labor force over 9 years of compulsory education 0.23 (0.16)
proration 371 Ratio of labor force with professional skills training 0.21 (0.23)
dependency 371 Labor burden coefficient 1.22 (1.55)
avrsubsidy 371 Average agricultural subsidy per mu (million CNY mu–1)1) 0.01 (0.92)
partyratio 371 Ratio of party household 0.04 (0.04)
avrmobile 371 Number of mobile phones per capita 0.88 (0.53)
ifsuburbs 371 1 if villages are nearby city, 0 otherwise 0.19 (0.39)
hardroadrate 371 Ratio of hardened road 0.84 (0.22)
freezer 371 Number of cold storages in the village 0.46 (1.72)
poverty village 371 1 if village in national poverty country, 0 otherwise 0.18 (0.39)
1)
1 ha=15 mu.
2)
S.D.=standard deviation.
 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011 1005

9.96 in the fifth year. By examining the change in the village- as a positive driver of income. This paper adopts the time
level income, it appears that e-commerce increases farmers’ when the villages were involved in the NRECDP, rather
income (Table 4), although in the first year of the NRECDP than the experience about rural e-commerce, as the key
participation, the NRECDP villages had lower income than independent variable. Many factors affect household
non-NRECDP villages. income, including the characteristics of rural household,
Summary statistics for disposable income per capita by land resources, and the economic development of their
different groups are shown in Table 5. We take the World villages. To identify the impact of rural e-commerce on
Bank’s standard as a reference to define the “relative poverty household income while controlling other factors, we used an
villages”. According to the World Bank’s standard, living on ordinary least square (OLS) regression approach to estimate
less than 3.2 USD and 5.5 USD per capita per day reflects eq. (1) directly. The variable “avrincome” is the per capita
poverty lines in low and middle-income countries and disposable income in the village; NRECDP_time represents
high-income countries, respectively. The current poverty the time when villages were involved in the NRECDP; eagent
standard in China is between 3.2 USD and 5.5 USD per represents the emerging of e-commerce agents in the
capita per day, so this paper defines relative poverty line as village. And X1 is a vector of exogenous variables that might
the daily cost of living being less than 5.50 USD. influence the outcome variable. Among them, edu9ratio,
According to Table 5, rural e-commerce could play a more proration and partyratio represent the village’s human
important role in the relative poverty villages than in those not capital characteristics; the variable “dependcy” represents
in poverty. This offers a clue that the “NRECDP may be more the old-age dependency ratio; the variable “avrsubsidy”
effective in relative poverty villages. The survey showed that represents the amount of agricultural subsidies; “ifsubsurbs”,
the per capita disposable income (PCDI in the table) of the “hardroadrate” and “freezer” represent the location and
rural e-commerce experiment villages was higher than that of public facilities of the villages. α and β are parameters to
non-experiments villages. The per capita disposable income be estimated; and ε is an error term.
of the NRECDP villages and the non-NRECDP villages were ln_avrincome=α+β1NRECDP_time+β2eagent+γX1+ε(1)
10 043.97 and 13 708.83 CNY, respectively. The per capita However, our data shows that only 96 out of the 238
disposable income for the NRECDP villages in relative sample villages involved in the NRECDP. This data structure
poverty was 8 773.34 CNY, and that of the non-NRECDP may lead to serious sample selection errors, resulting in the
villages in relative poverty was 9 161.74 CNY. endogeneity of the regression model not being controlled.
We then used the Heckit method to address the sample
3.3. Methods selection issue (Heckman 1979; Heckman and Robb 1985;
Semykina and Wooldridge 2010). Given the NFRS sample
E-commerce, theoretically and empirically, has been taken villages are considered as typical villages of each county, it

Table 4 Labor engaged in e-commerce

Village category1) Year involved in NRECDP1) Labor engaged in e-commerce Income from e-commerce (CNY)
Non-NRECDP – 5.67 56 000
NRECDP Yr 2018 14.89 32 889
Yr 2017 7.30 71 500
Yr 2016 3.32 129 776
Yr 2015 9.26 129 727
Yr 2014 9.96 111 080
1)
NRECDP, the National Rural E-commerce Comprehensive Demonstration Project.
Source: Calculated from National Fixed-site Rural Survey.

Table 5 Income distribution by different category villages1)

Relative-poverty villages Total villages
Category2)
PCDI (CNY)3) Proportion (%) PCDI (CNY)2) Proportion (%)
NRECDP villages 8 773.34 36.02 10 043.97 25.49
Non-NRECDP villages 9 161.74 63.98 13 708.83 74.51
1)
The standard of an absolute-poverty village is whether it is a registered poor village; the standard of a relative-poverty
village is 5.50 USD per capita per day (from the World Bank report, [Link]
le/10986/30418/[Link]); 1 USD≈6.75 CNY (2018 data).
2)
NRECDP, the National Rural E-commerce Comprehensive Demonstration Project.
3)
PCDI, per capita disposable income.
1006 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011

is believed that these villages could represent their counties’ for all villages. The result shows that the inverse mills ratio
ordinary characteristics. We therefore divide the effect of (IMR) is significantly positive, which means the selective bias
rural e-commerce on household income into two continuous is not allowed to be ignored. The endogeneity problems
stages. First, using all the observations to predict the should be controlled for by the Heckman two-step method.
probability of one village receiving the intervention (namely, It is thus more suitable to use the Heckman approach to
the NRECDP), and calculate the reverse mills ratio of each test the effect of rural e-commerce on income and poverty
observation. And second, using the samples involved in reduction. Columns (1) and (3) report the selection equation
the intervention to carry out the second stage regression regression results for the full sample and relative-poverty
analysis, while taking the reverse mills ratio as an additional villages. Columns (2) and (4) report the Heckman outcome
control variable to obtain consistent estimates (Chen 2012). equation regression results for the full sample and relative-
Consequently, there are two equations: the first stage poverty villages. There is a significant positive correlation
selection equation and the second stage main regression between the duration of the NRECDP involvement and the
equation. In the selection equation, the NRECDP village, a per capita disposable income. Along with the progress of
dummy variable, is the dependent variable. We use a Probit the NRECDP, the income effect of rural e-commerce is
model to estimate the selection equation. increasingly significant. We also examine the role of rural
prob(eprogram=1)=Φ(δ´X2) (2) e-commerce agents in the village. While it is not statistically
In eq. (2), prob (eprogram=1) is the probability of villages significant at the level of 10%, the coefficient is positive. In
being selected into the NRECDP, and X2 indicates the factors theory, rural e-commerce agents may play a positive role
influencing the village to be selected. However, in order in promoting ordinary farmers’ income. However, it could
to address the sample selection issue, there should be at be the case that rural e-commerce agents actually increase
least one explanatory variable that satisfies the exclusive their own income by rural e-commerce, rather than income
conditions. This variable should be correlated with the of the rural households.
NRECDP, but should not be directly correlated with per The regression results of other control variables
capita disposable income. In our paper, the number of generally are consistent with common sense and economic
mobile phones per capita (avrmobile) and whether a village intuition. It finds that skill training is positively correlated
belongs to a national poverty county (poverty village) are with PCDI, which is consistent with the findings of Wan
regarded as the exclusive explanatory variables in the and Zhou (2005) and Knight et al. (2010). The variable
selection equation. X2 includes avrmobile and poverty representing ratio of party members is positive and
village in addition to the variables in X1. According to eq. (2), statistically significant. This suggests that the identity of
we can change eq. (1) into eq. (3): party members is an important social capital. Villages
ln_avrincome=θ+ρ1NRECDP_time+ρ2eagent+λX1 close to urban suburbs appear to have higher per capita
+ηϕ(δ´X2)/Φ(δ´X2)+µ (3) disposable income, which is in line with Ma et al. (2020).
Among them, ln_avrincome represents the logarithm However, the variables “edu9ratio” and “eagent” are not
of the per capita disposable income of a rural household, statistically significant. “Nine-year compulsory education”,
NRECDP_time represents the implementation time of the to some extent, reflects the general quality of labor force.
NRECDP in villages, X1 is a vector of corresponding control The villages with more high-quality laborers are supposed
variables, f(d´X2)/F(d´X2) is the correction term, θ is the to be richer. E-commerce agents can be approximately
constant term, and μ is an error term. In this paper, we use regarded as an intermediary and therefore villages with
eqs. (1), (2), and (3) to test the income increasing effect and more e-commerce agents have a higher probability of
poverty reduction effect of rural e-commerce. developing e-commerce.
In the selection equation, the ratio of nine-year
4. Results and discussion compulsory education and the proportion of labor force
are negatively and significantly correlated with the choice
4.1. Empirical results of experiment villages. However, being a national poverty
village is positively and significantly correlated with the
We firstly present the determinants of rural e-commerce choice of experiment villages. This result suggests that
participation from a probit model. In step 2, we present the poor areas tend to be involved in the NRECDP. This is
basic model of the effects of rural e-commerce on per capita consistent with the aim of government policies.
disposable income based on eq. (3). Column (2) of Table 6 reports our preferred specification,
Table 6 shows the results of the estimation for eq. (3), including the regression results of the relative poverty
representing the effect of rural e-commerce on the PCDI sample. The point estimate from panel A indicates that the
 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011 1007

Table 6 Empirical result of benchmark regression

Full sample Relative-poverty villages
Independent variables eprogram: ln_avrincome: eprogram: ln_avrincome:
selection equation (1) outcome equation (2) selection equation (3) outcome equation (4)
NRECDP_time=1 0.453* 0.395**
(0.235) (0.199)
NRECDP_time=2 0.456* 0.397**
(0.238) (0.199)
NRECDP_time=3 0.534** 0.446**
(0.229) (0.192)
NRECDP_time=4 0.607*** 0.417**
(0.229) (0.196)
NRECDP_time=5 0.747*** 0.405
(0.258) (0.255)
eagent 0.394* 0.023 0.469* 0.019
(0.205) (0.072) (0.259) (0.066)
edu9ratio –1.778** 0.004 –0.708 0.007
(0.856) (0.333) (0.927) (0.300)
proratio 0.562 0.276* 0.746 0.334**
(0.452) (0.167) (0.596) (0.150)
dependcy 0.011 0.019 –0.052 –0.007
(0.067) (0.026) (0.097) (0.025)
avrsubsidy 0.059 0.006 0.019 0.008
(0.110) (0.032) (0.133) (0.324)
partyratio 0.674 2.337*** 0.693 0.902
(3.144) (0.796) (3.763) (0.793)
ifsubsurbs –0.198 0.284** –0.219 0.169
(0.268) (0.118) (0.419) (0.135)
hardroadrate 0.304 0.082 0.027 0.032
(0.449) (0.154) (0.583) (0.137)
freezer 0.048 0.013 0.078 0.018
(0.052) (0.016) (0.099) (0.014)
avrmobile 0.041 0.087
(0.228) (0.262)
poverty-village 1.568*** 1.425***
(0.236) (0.277)
IMR 0.155** 0.144**
(0.068) (0.070)
Province variable Yes Yes Yes Yes
Constant –6.512*** 8.195*** –1.019 8.397***
0.809 (0.303) 0.765 (0.259)
Observations 365 211
Selected 92 78
Not selected 273 133
R2 0.524 0.481
***
Robust standard errors are presented in parentheses. , P<0.01; **, P<0.05; *, P<0.1.

first year of involvement in the NRECDP is positive and variables are similar to those examined in the full sample.
statistically significant. The following second to fourth year
show the same results. The coefficient of the fifth year is 4.2. Sensitivity analysis
positive but not statistically significant. As the NRECDP
progresses, the income effect of rural e-commerce shows In observational studies or quasi-natural experimental
a trend of increasing initially and then declining. studies, missing variable bias is usually the key factor
To summarize, the NRECDP’s effect on farmers’ income affecting the robustness of causal inference. We have
in the relative poverty villages is inverted U-shaped. In the controlled some variables in the benchmark regression.
first three years, the positive income effect increased year However, as long as the missing variables are correlated
by year. In the next two years, the effect declined, though with the explanatory variables or the explained variables
it was still positive. The regression results of other control appear in the unobservable confounding variables, the
1008 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011

validity of the benchmark regression results would be statistically significant, even at the 10% level of significance.
doubtful. Therefore, referring to the research ideas of Frank The findings suggest that the covariates’ balancing property
(2000), we analyzed the sensitivity of the estimated results was satisfied (Anang et al. 2017; Ma et al. 2020).
of the previous regression by using the bias percentage. By Table 8 presents the results based on the kernel-based
so doing, we further ensure the robustness of the empirical matching technique. The effect of rural e-commerce on the
results. Bias percentage is defined as the proportion of household income of all villages is shown in column (2).
observable samples replaced. Bias percentage makes The results show that the coefficient of involvement in the
the inference invalid, i.e., the minimum bias degree that NRECDP is positive and statistically significant. It means
can change the conclusion of causal inference. So we that e-commerce has a significantly positive effect on the
can effectively address the problem of missing variables rural households’ income. Column (4) shows the result for
caused by unobservable factors. the relative poverty villages. After rematching samples, the
Table 7 represents the results of sensitivity analysis for impact of rural e-commerce on personal income also shows
variables that passed the significance test in the previous a U-shaped pattern. Combined with the regression results of
regression. The bias percentage of the core explanatory different samples, it can be concluded that rural e-commerce
variable, NRECDP_time, is 64.95%. It means that the has a significant and positive impact on income and reducing
regression result of NRECDP_time on per capita disposable poverty. However, the effect appeared to be heterogeneous
income in the village will be changed when the bias of the among villages. For the relative poverty villages, the impact
unobservable missing variable reaches 64.95%. Among of rural e-commerce on personal income shows an inverted
the three significant control variables, partyratio has the U-shaped pattern. With the continuous progress of the
lowest missing variable bias (40.74%), ifsubsurbs has the NRECDP, the marginal effect of rural e-commerce on income
highest missing variable bias (84.74%), and the average tends to become stable gradually.
deviation percentage of the control variables is 60.64%. As shown in Tables 6 and 8, household income could
The result shows significant robustness. be improved by rural e-commerce significantly. Rural
e-commerce could play a more statistically significant role
4.3. Robustness test in enhancing income. The ideal model of poverty alleviation
is to enhance the ability of self-development of low-income
To further examine the impact of rural e-commerce on
groups. Generally, villagers who would like to engage in
farmer’s income, we conducted a propensity score matching
e-commerce need to go through vocational training.
(PSM) model as a robustness check (Ogutu et al. 2014;
Rural households’ lack of skills is the main reason
Qin and Chong 2018). PSM estimation results are shown
affecting their participation in e-commerce. According to
in Table  8. Based on the observed characteristics, PSM
a specific survey of 9 553 rural households nationwide by
makes it possible to match villages with e-commerce income
the NFRS in 2016, the proportion of farmers who had a
and those without. In the PSM estimation framework, the
strong willingness (very or relatively strong) to engage in
selection equation can be estimated by adopting a binary
e-commerce was only 8.2%. It is not hard to imagine that
choice model (a logit model in this study). By doing so,
the proportion of those who actually engaged in e-commerce
propensity scores for villages could be generated in the
would be even less. The third national agricultural census
sample. According to the distance defined by propensity
offers a similar finding. In 2016, the proportion of large-
scores, the PSM approach matches the villages participating
scale agricultural business households and agricultural
in the NRECDP and those not. The potential selectivity
business units engaging in agricultural e-commerce was
effects arising from observable factors could be alleviated.
1.6%, of which 1.1% were large-scale agricultural business
The kernel-based matching technique is adopted to show
households NFRS in 2016 showed that the proportion of
that none of the mean differences of the selected variables
households that were unable to engage in e-commerce was
between the NRECDP-villages and non-NRECDP villages is
as high as 49.1%. As mentioned above, e-commerce may
be a digital divide for rural people.
Table 7 Empirical result of sensitivity analysis Overall, we tried to examine the poverty alleviation effect
Variables Bias in full Bias in relative-poverty of e-commerce in this paper. As a beginning study on
detected samples (%) villages (%)
e-commerce demonstration policies, this paper revealed that
NRECDP_time 64.95 64.77
proratio 52.13 41.79 e-commerce could help poor people to get rid of poverty.
partyratio 40.74 – The NRECDP was considered as a quasi-experiment.
ifsubsurbs 84.74 – Because the mechanism variables under the NRECDP were
Mean 60.64 53.28 unavailable, the mechanism through which e-commerce
 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011 1009

Table 8 Empirical result based on propensity score matching model

Full sample Relative-poverty villages
Independent variables eprogram: ln_avrincome: eprogram: ln_avrincome:
selection equation (1) outcome equation (2) selection equation (3) outcome equation (4)
NRECDP_time=1 0.470** 0.276
(0.237) (0.222)
NRECDP_time=2 0.463** 0.351*
(0.235) (0.216)
NRECDP_time=3 0.598*** 0.453**
(0.227) (0.208)
NRECDP_time=4 0.599*** 0.365*
(0.225) (0.210)
NRECDP_time=5 0.789*** 0.421
(0.259) (0.343)
eagent 0.396* 0.108 0.301* 0.025
(0.236) (0.087) (0.316) (0.089)
edu9ratio –0.643 0.250 2.425* 0.366
(1.018) (0.398) (1.425) (0.383)
proratio 0.338 0.133 0.475 0.437**
(0.511) (0.210) (0.723) (0.204)
dependcy –0.004 0.008 –0.108 –0.030
(0.071) (0.031) (0.115) (0.034)
avrsubsidy –0.109 0.001 –0.205 0.012
(0.178) (0.046) (0.211) (0.419)
partyratio –0.099 0.286 1.722 0.578
(3.452) (0.992) (4.056) (0.887)
ifsubsurbs –0.092 0.265** 0.117 0.169
(0.281) (0.124) (0.460) (0.152)
hardroadrate 0.253 0.348* 0.376 0.108
(0.499) (0.183) (0.711) (0.172)
freezer –0.026 0.011 –0.298 0.022
(0.068) (0.023) (0.215) (0.022)
avrmobile 0.019 0.131
(0.242) (0.286)
poverty-village 1.724*** 1.599***
(0.301) (0.358)
IMR 0.251*** 0.131*
(0.074) (0.081)
Province variable Yes Yes Yes Yes
Constant –7.107*** 8.005*** –1.217 8.461***
(1.044) (0.322) (0.872) (0.303)
Observations 308 178
Selected 71 59
Not selected 237 119
R2 0.401 0.357
***
Robust standard errors are presented in parentheses. , P<0.01; **, P<0.05; *, P<0.1.

contributed to poverty alleviation had not been examined in China. The Heckit model was employed to address
in depth. the potential selectivity bias from both observed and
unobserved factors. The PSM approach was adopted to
5. Conclusion and policy recommendations test the robustness of empirical findings. To the best of our
knowledge, our study is innovative and comprehensive and
Studies that investigate the wider poverty-reduction effects the first one that identifies a statistical association between
through e-commerce are relatively small in quantity. This rural e-commerce and household income based on village-
paper fills this literature gap by examining the effect of rural level data. Our findings indicate that rural e-commerce
e-commerce on farmers’ income. We drew upon cross- has a significantly positive effect on personal income.
sectional analysis by using village-level data collected Moreover, this effect in the relative poverty villages is
from 31 provinces, municipalities and autonomous regions inverted U-shaped. The robustness of the findings was
1010 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011

confirmed by the estimation results of the PSM models. from the  National Natural Science Foundation of China
We concluded that rural e-commerce could actually (71673274); the Outstanding Innovative Talents Cultivation
promote rural income and enhance rural development. Funded Programs 2019 of Renmin University of China.
The ideal model for poverty alleviation is to enhance the
capacity of low-income groups so that they can continue the Declaration of competing interest
development path independently. E-commerce empowers
some poor people, especially through skill training programs The authors declare that they have no conflict of interest.
usually provided by the government’s e-commerce projects.
Our field survey found that professional skills training for References
e-commerce contributes to increasing income. E-commerce
could be combined with other poverty alleviation approaches Acker J C, Mbiti I M. 2010. Mobile phones and economic
and intellectual support. E-commerce poverty alleviation development in Africa. Journal of Economic Perspectives,
would allow poor people to obtain information via internet, 24, 207–232.
which enables poor households to “open their vision” Akca H, Sayili M, Esengun K. 2007. Challenge of rural people
and “accept more insight”. This kind of “vision” and to reduce digital divide in the globalized world: Theory and
practice. Government Information Quarterly, 24, 404–413.
“insight” directly determines poor people’s initiative to get
Anang B T, Bäckman S, Rezitis A. 2017. Production technology
rid of poverty. The introduction of e-commerce to rural
and technical efficiency: Irrigated and rain-fed rice farms in
areas allows some poor people to actively integrate into
northern Ghana. Eurasian Economic Review, 7, 95–113.
modern society and share the benefits of modernization.
Baorakis G M, Kourgiantakis M, Migdalas A. 2002. The impact
E-commerce also can efficiently spread anti-poverty stories of e-commerce on agro-food marketing: The case of
through the internet, which could create a social atmosphere agricultural cooperatives, firms and consumers in Crete.
of caring for poor areas. British Food Journal, 104, 580–590.
The policy implication of our paper is significant. First, Chan M. 2015. Mobile phones and the good life: Examining
investment in e-commerce in recent years is critical in the relationships among mobile use, social capital and
increasing income and alleviating poverty. The policy subjective well-being. New Media & Society, 17, 96–113.
designs should seek to improve poor areas’ access to hard Chen Y. 2012. Village-based networks and wage of rural-to-
(internet) and soft (education) infrastructure. In the future, urban migrants: Estimating the causal effects of networks
the main direction of poverty alleviation through e-commerce using combined identification strategies. Chinese Journal
is to introduce digital technology and business forms into of Sociology, 4, 68–92.
Clark C, Gorski P. 2002. Multicultural education and the
poor areas. Combined with the introduction of e-commerce
digital divide: Focus on socioeconomic class background.
into rural production and management, the digitizing of all
Multicultural Perspective, 4, 25–36.
kinds of business entities should be accelerated. On the
Fafchamps M, Minten B. 2012. Impact of SMS based agricultural
platform of e-commerce, human resources can be improved
information on Indian farmers. The World Bank Economic
to achieve the revitalization of rural talents. In the poverty Review, 26, 383–414.
areas, construction of information infrastructure should Frank K A. 2000. Impact of a confounding variable on a
be implemented. Network speed-up and fee reduction regression coefficient. Sociological Methods and Research,
projects should be adjusted to better serve the poverty 29, 147–194.
areas. 5G network deployment should be ahead of Gao Y, Zang L, Sun J. 2018. Does computer penetration
schedule in rural areas. More e-commerce outlets should be increase farmers’ income? An empirical study from China.
constructed in poverty-stricken areas. Those outlets should Telecommunications Policy, 42, 345–360.
integrate various types of social capital resources such as Goldfarb A, Tucker C. 2019. Digital economics. Journal of
e-commerce, express, logistics, commerce, and finance, etc. Economic Literature, 57, 3–43.
Hartje R, Hübler M. 2017. Smartphones support smart labour.
Successful and inspirational stories of poverty alleviation
Applied Economics Letters, 7, 467–471.
in rural areas could be promoted on the internet. Regional
Heckman J. 1979. Sample selection bias as a specification
brands’ development strategy could specifically focus on the
error. Econometrica, 47, 154–161.
utilization of e-commerce. To summarize, only if the digital
Heckman J, Robb R. 1985. Alternative methods for evaluating
divide were bridged, rural income would increase through the impact of interventions: An overview. Journal of
the “digital dividend”. Ecnometrics, 30, 239–267.
Jalali A A, Okhovvat M R, Okhovvat M. 2011. A new applicable
Acknowledgements model of Iran rural e-commerce development. Procedia
Computer Science, 3, 1157–1163.
The authors gratefully acknowledge the financial support Jha S K, Pinsonneault A, Dubé L. 2016. The evolution of an
 PENG Chao et al. Journal of Integrative Agriculture 2021, 20(4): 998–1011 1011

ICT platform-enabled ecosystem for poverty alleviation: The NBSC. 2019. China’s rural poor population decreased by 13.86
case of EKutir. MIS Quarterly, 40, 431–445. million in 2018. [2020-5-20]. [Link]
Knight J, Li S, Deng Q. 2010. Education and the poverty trap in zxfb/201902/t20190215_1649231.html (in Chinese)
rural China: Closing the trap. Oxford Development Studies, Ogutu S O, Okello J J, Otieno D J. 2014. Impact of information
38, 1–24. and communication technology-based market information
Leng C, Ma W, Tang J, Zhu Z. 2020. ICT adoption and income services on smallholder farm input use and productivity: The
diversification among rural households in China. Applied case of Kenya. World Development, 64, 311–321.
Economics, 52, 3614–3628. Parker C, Ramdas K, Savva N. 2016. Is it enough? Evidence
Leong C, Pan S L, Newell S, Cui L. 2016. The emergence of self- from a natural experiment in India’s agriculture markets.
organizing e-commerce ecosystems in remote villages of Management Science, 62, 2481–2503.
China: A tale of digital empowerment for rural development. Qin C, Chong T T L. 2018. Can poverty be alleviated in China?
MIS Quarterly, 40, 475–484. Review of Income and Wealth, 64, 192–212.
Li L, Du K, Zhang W, Mao J. 2016. Poverty alleviation through Semykina A, Wooldridge J M. 2010. Estimating panel data
government-led e-commerce development in rural China: models in the presence of endogeneity and selection.
An activity theory perspective. Information Systems Journal, Journal of Econometrics, 157, 375–380.
29, 914–952. Shimamoto D, Yamada H, Gummert M. 2015. Mobile phones
Liu M, Feng X, Wang S, Qiu H. 2019. China’s poverty alleviation and market information: Evidence from rural Cambodia.
over the last 40 years: successes and challenges. Australian Food Policy, 57, 135–141.
Journal of Agricultural and Resource Economics, 59, 1–20. Song M X, Yang M X. 2019. Leveraging core capabilities and
Ma W, Grafton R Q, Renwick A. 2020. Smartphone use and environmental dynamism for food traceability and firm
income growth in rural China: Empirical results and policy performance in a food supply chain: A moderated mediation
implications. Electronic Commerce Research, 20, 713–736. model. Journal of Integrative Agriculture, 18, 1820–1837.
Ma W, Renwick A, Yuan P, Ratna N. 2018. Agricultural Tao T Y, Zhao M. 2012. An ontology-based information retrieval
cooperative membership and technical efficiency of apple model for vegetables e-commerce. Journal of Integrative
farmers in China: An analysis accounting for selectivity bias. Agriculture, 11, 800–807.
Food Policy, 81, 122–132. Wan G, Zhou Z. 2005. Income inequality in rural China:
Ministry of Commerce of China. 2019. E-commerce in China. Regression-based decomposition using household data.
China Commerce and Trade Press, Beijing. (in Chinese) Review of Development Economics, 9, 107–120.
Mintert J, Andresen D, Schroeder T. 2003. Improving efficiency Zeng Y, Guo H, Jin S. 2018. Does e-commerce increase
in business-to-business information transfers: A web- farmers’ income? Evidence from Shuyang County, Jiangsu
based solution in the beef sector. International Journal of Province, China. Chinese Rural Economy, 2, 49–64. (in
Information Management, 23, 415–424. Chinese)

Executive Editor-in-Chief HUANG Ji-kun

Managing Editor WENG Ling-yun