Rural infrastructure and poverty in China.

The study develops a theoretical framework of how irrigation and drainage infrastructure and rural transportation infrastructure influence poverty. Using panel data on 31 provinces in China from 2002 to 2017, this paper estimates basic and continuous difference-in-differences (DID) models to investigate the preliminary impact of irrigation and drainage infrastructure and rural transportation infrastructure on poverty and further explores the influence mechanisms of these rural infrastructures on poverty by using the mediating effect model. The results show that irrigation and drainage facilities infrastructure can directly reduce poverty. On the one hand, rural transportation infrastructure directly leads to rural hollowing out and aggravates rural poverty; on the other hand, it indirectly promotes poverty reduction by stimulating economic growth. Overall, the positive and negative effects of rural transportation infrastructure on poverty offset each other.

Introduction

Poverty reduction has always been a core component of the Sustainable Development Goals of the United Nations, which are deeply important for promoting the sustainable development of the world especially when facing the threat of normalizing extreme weather. On November 23rd 2020, China announced that it had eliminated absolute poverty nationwide by uplifting all its citizens beyond its set 2300 yuan per year, or around 1.52 dollars per day poverty line. Over the past 40 years, China has pursued many stimuluses to achieve this goal and more than 700 million people has been lifted out of poverty. One of the stimuluses is to promote a variety of infrastructure projects [1]. In fact, infrastructure plays a fundamental role in promoting growth and alleviating poverty in China, especially in rural areas greatly affected by extreme weather. Agriculture, rural areas and farmers’ issues have always been China’s top priorities since 2004 and there have been 17 No. 1 central documents, made by the government of China, prioritizing development of agriculture and rural areas with different key themes each year. In 2010, China’s No. 1 central document proposed to strengthen construction of rural infrastructure, especially for rural irrigation and drainage and transportation infrastructure, to reduce poverty. The central and local government funds as well as national debt funds needed to be invested to the construction and management of rural infrastructure. More specifically, the government arranged 818.3 billion yuan for agriculture, peasants, and rural areas in 2010. Among them, 86.2 billion yuan was allocated for small-sized irrigation projects and 132.3 yuan for rural roads and other public transport infrastructures, beyond the previous investment. After 2010, China’s No. 1 central document still attaches great importance to rural infrastructure of irrigation and transportation and keep investing to rural infrastructure to help rural poor people to deal with extreme weather.

Given that China has invested heavily to irrigation and transportation infrastructure especially in rural areas and the miracle of poverty reduction in the past 40 years, three important issues arise and needs to be figured out. First, did the investment to irrigation and transportation infrastructure in 2010 reduce poverty in practice? Second, what roles did irrigation and transportation infrastructure play in influencing poverty, respectively? Third, what is the mechanism behind the impact? Some studies maintain that rural transportation infrastructure, may not have social benefits and succeed in reducing poverty if rural laborers choose to transfer to urban areas due to the lack of agglomeration of local economies [2]. On the contrary, rural poverty may be deepened because of the loss of young and middle-aged rural labor force and the decline of abandoned villages, which is called siphon effect [3]. Moreover, irrigation and drainage infrastructure may also fail to count in that case. Dealing with these three problems can help China’s government reasonably adjust investment structure in rural infrastructure and consolidate the achievements on poverty reduction, as well as provide experience for those developing economies troubled by poverty. In addition, under the threat of extreme weather, it is of particular importance to clarify the role of irrigation and drainage infrastructure in affecting poverty. Irrigation and drainage infrastructure, due to the characteristics of relatively lower profits and slower return, usually attracts much less investment than rural transportation infrastructure.

Given the above discussions, this research aims to evaluate the impact and mechanisms of rural irrigation and transportation infrastructure on poverty and makes innovations in several ways. First, this paper theoretically clarifies how irrigation and drainage infrastructure and rural transportation can affect poverty. We supplement theoretical literature concerning how irrigation and drainage infrastructure influences poverty and the underlying mechanism, which are rare in previous research. In addition, we add to a growing body of literature on how rural transportation influences poverty, focusing not on the poverty-reducing effect but also on the poverty-aggravating effect. Second, this paper helps enrich empirical evidence on how rural irrigation and transportation infrastructure influences poverty by applying the basic and continuous DID models and the mediating effect model. We use updated data to make the results more convincible and two-stage least square method to handle endogeneity problem.

The rest of this paper is structured as follows. The next section reviews previous literature. The Framework and Hypotheses section analyzes the theoretical influencing mechanisms of the three kinds of rural infrastructure on poverty. The Materials and Methods section discusses materials and methods. The Results and Conclusion section present the empirical test and discuss the results, respectively. The last section concludes with policy implications, limitations and future research.

Literature review

Many empirical studies have suggested a positive role for rural infrastructure in allowing improvements to the quality of life of those living in poverty and alleviating poverty. Most of these studies highlight the impact of rural infrastructure on economic growth and thus indirectly on poverty, which is called a trickle-down effect [4-9]. The research by these authors illustrates that rural infrastructure such as rural transportation infrastructure can stimulate economic growth through gains in productivity, which in turn leads to increases in income and poverty alleviation. Sasmal, R., and Sasmal, J. (2016) and Chotia et al.,(2017) examine the connection between economic growth and poverty alleviation as well as how the two are connected to public infrastructure [10, 11]. The results reveal that economic growth may drive poverty reduction, and infrastructure features largely in both growing the economy and reducing poverty.

Some seminal works also pay attention to how rural infrastructure directly affects poverty. Most of the authors believe that rural infrastructure construction can directly reduce farmers’ production costs and improve productivity, thereby increasing their income and reducing poverty [12, 13]. Fan et al. (2005) reveal that investment in rural infrastructure can increase household income and pro-mote poverty reduction, with road infrastructure playing the most important role [6]. More recently, some scholars have pointed out that the improvement of rural infrastructure, especially rural transportation infrastructure, may promote urbanization. However, it may also lead to less productive capital and skilled labor in rural areas. In such “hollow villages”, the rural poor may fall into greater poverty [14]. This increase in poverty may offset the poverty reduction effect of rural infrastructure, which we can call the masking effect, often used in mediation analysis in psychology [15-17]. Other studies also explore the relationship between irrigation and drainage infrastructure and rural income or rural poverty [18].

Overall, previous studies have emphasized the importance of rural infrastructure for poverty reduction. Most studies focus on how rural transportation infrastructure affects poverty reduction and usually find the negative relationship between rural transportation infrastructure and poverty. However, few literatures manage to figure out the mechanisms of how the irrigation and drainage infrastructure influence poverty. The specific poverty reduction mechanism behind the irrigation and drainage infrastructure needs to be further determined. In addition, it is important to discuss the potential negative impact of rural transportation infrastructure on poverty reduction and make a comparison with the irrigation and drainage infrastructure. Therefore, this paper contributes to a growing but inconclusive body of literature by theoretically clarifying how irrigation and drainage infrastructure and rural transportation can affect poverty. In addition, this paper uses the basic and continuous DID models to study the impact of different rural infrastructures on poverty. This continuous DID model can solve certain endogeneity problems in the model. The two-stage least squares method and a change in the dependent variables are also used as tests of the robustness of the conclusion. Third, this paper uses the mediating effect model to further explore the mechanisms behind the effects of irrigation and drainage infrastructure and transportation infrastructure on poverty. Different from most existing studies, this study finds that rural transportation infrastructure both aggravates rural poverty and reduces poverty by promoting economic growth. Given the current policy situation and this realistic background, it is necessary to pay more attention to the irrigation and drainage infrastructure, which is related to agriculture production and peasants’ life. This consideration will further consolidate the gains in poverty alleviation and prevent future increases in poverty.

Framework and hypotheses

Irrigation and drainage infrastructure and poverty

On the one hand, irrigation and drainage infrastructure can directly reduce poverty. As one of the most basic public goods in rural areas, irrigation and drainage infrastructure can improve agricultural production conditions and grain yield by improving irrigation capacity. Farmers are able to adjust their crop structure, develop large-scale breeding programs and engage in processing and nonagricultural industries to eliminate poverty. At the same time, the improvement of irrigation and drainage infrastructure can enhance farmers’ ability to deal with disasters and reduce risks. Therefore, production efficiency will rise with lower agricultural production costs, thereby reducing poverty. In addition, irrigation and drainage infrastructure can release part of the rural labor force from farm work and optimize the work-time structure. This saved labor and time can be used for higher income activities. Thus, poverty can be reduced. On the other hand, irrigation and drainage infrastructure can indirectly reduce poverty by promoting the growth of agriculture, forestry, animal husbandry and fisheries [19]. According to the Cobb-Douglas production function, the elasticity of capital and labor may directly determine growth. Irrigation and drainage infrastructure can increase the output from a unit of capital and labor by reducing the impact of floods and other disasters on them. The growth in agriculture, forestry, animal husbandry and fisheries can continue, thus indirectly affecting poverty.

Based on the literature review, we have the following hypotheses:

Hypothesis 1a. Irrigation and drainage infrastructure can directly reduce poverty.

Hypothesis 1b. Irrigation and drainage infrastructure can indirectly reduce poverty through increasing the growth in local agriculture, forestry, animal husbandry and fisheries.

Rural transportation infrastructure and poverty

The relationship between rural transportation infrastructure and poverty remains complex. In terms of its direct impact, improvements in rural transportation infrastructure can reduce the transportation costs of farmers and expand the market opportunities for local agricultural products [20]. However, due to the remote location of and the lack of talent and capital concentration in rural areas, new roads transfer a large amount of labor to the city. This directly leads to the loss of local talent and the hollowing out of rural areas, which aggravates rural poverty [21, 22].

On the other hand, rural transportation infrastructure can reduce poverty indirectly by promoting tertiary industry and stimulating economic growth. The trickle-down effects from economic growth contribute to poverty alleviation. As one of the factors of production, investment in rural transportation infrastructure can promote the division of labor, improve production efficiency, and directly promote economic growth. The multiplier effect produced by investment can further stimulate the vitality of economic growth. In addition, as a public good, rural transportation infrastructure has externalities. This means that investment in rural transportation infrastructure can cause farmers to accumulate human capital and can promote manufacturing production. Additionally, transportation costs can consequently decrease, which can improve the overall investment structure to allow for additional investments and bring about economic growth [23-28]. Economic growth can reduce poverty by increasing employment opportunities and improving transfer payments, which are both trickle-down effects.

Therefore, we have hypotheses presented as below:

Hypothesis 2a. Rural transportation infrastructure may directly aggravate rural poverty.

Hypothesis 2b. Rural transportation infrastructure can promote tertiary industry growth and thus indirectly reduce poverty.

The mechanism can be pictured in Fig 1.

Fig 1

The mechanism of irrigation and drainage and rural transportation infrastructure on poverty.

The direct an indirect mechanism of irrigation and drainage and rural transportation infrastructure on poverty.

Materials and methods

Data source

Due to limited data availability, this paper uses panel data on 26 provinces in China from 2002 to 2017. The data are collected mainly from the 2003–2018 China Rural Statistical Yearbook, China Statistical Yearbook, and the National Bureau of Statistics. The ethnic regions in this paper are eight ethnic provinces in China, referring to the Inner Mongolia Autonomous Region, Ningxia Hui Autonomous Region, Xinjiang Uygur Autonomous Region, Tibet Autonomous Region and Guangxi Zhuang Autonomous Region, as well as Guizhou, Yunnan and Qinghai provinces. Poverty is more serious in these areas for environmental and historical reasons. Additionally, due to a lack of data, we do not consider Beijing, Shanghai, Hainan, or Tibet.

Methods

Definition of variables

Dependent variables. Rural poverty incidence rate (Poverty_rate) and the size of the rural poor population (Poverty_num). We define people in rural China whose income falls below the minimum income required by the government as the rural poor. The proportion of these people to the total rural population is the rural poverty incience rate. With the targeted poverty alleviation program in China, those who receive basic living allowances account for an increasing proportion of the impoverished population. They are good representatives for those who live in extreme poverty.

Core independent variables. Referring to the definition of [29], rural infrastructure is different from traditional public services. It caters to farmers’ production, life and development with a long service life and two of the most important are irrigation and drainage and rural transportation infrastructure. Irrigation and drainage infrastructure is closely related to peasants’ production and may help them increase family business income. While rural transportation infrastructure is closely related to non-agriculture employment and may help them increase wage income. Both irrigation and drainage and rural transportation infrastructure are crucial to poor peasants’ income and poverty reduction. We choose effective irrigation area to represent the construction of irrigation and drainage infrastructure as it can reflect the actual irrigation effect on cultivated land. And for rural transportation infrastructure, we choose rural road mileage to represent its construction.

Intermediary variable. According to the previous theoretical analysis, irrigation and drainage infrastructure can indirectly reduce poverty by promoting the growth of agriculture, forestry, animal husbandry and fisheries. Rural transportation infrastructure can indirectly reduce poverty by influencing the growth of the service industry [30]. Therefore, for irrigation and drainage infrastructure, we choose the added value of agriculture, forestry, animal husbandry and fisheries as the intermediary variable [31, 32]. While for rural transportation infrastructure, we choose and the added value of tertiary industries as intermediary variable [24].

Control variables. To improve the robustness of the model, this paper refers to previous studies [1, 12] and selects control variables from the three dimensions of economy, society and environment. The development of economy, society and environment may have influences on poverty through many ways. For example, economic development may reduce poverty through increasing consumption and expanding the channels of employment. And the development of society may affect the social welfare of the peasants and then pose impact on poverty. While the development of environment may directly influence the agricultural production and then affect welfare of the peasants and poverty. And referring to [9, 12, 27], the control variables at the economic level include per capita GDP, per capita industrial output, industrial structure, government expenditures, and rural residents’ consumption levels. From the social dimension, the control variables are population density, human capital, population urbanization rate, land urbanization rate, urban-rural income gap, rural electricity consumption, mechanization level, and level of financial support for agriculture. From the environmental dimension, the control variables are land area available for crop planting, reservoir capacity, soil erosion control, and grain yield. The proxy indicators for each control variable are shown in Table 1.

Table 1

Definition of the variables in the model.

Variable type	Dimension	Variable	Symbol	Description
Dependent variable	Poverty	The size of the rural poor population	Poverty_num	The number of the rural poor population
		Rural poverty incidence rate	Poverty_rate	Rural minimum living guarantee population / Total rural population
		Irrigation and drainage infrastructure	Irrigation	Effective irrigation area (1000 HA)
Independent variable	Rural infrastructure	Rural transportation infrastructure	Road	Rural road mileage (km)
Intermediary variable	Industrial added value	The added value of agriculture, forestry, animal husbandry and fisheries	Primary industry	Added value of agriculture, forestry, animal husbandry and fishery (100 million yuan)
Control variable	Economy	The added value of tertiary industries	Tertiary industry	The added value of tertiary industries (100 million yuan)
		Per capita GDP	Pgdp	Per capita GDP (yuan/person)
		Per capita industrial output	Pindustrialization	Industrialization level (10000 yuan / person)
		Industrial structure	Industrial structure	The ratio of the sum of the primary industry and the secondary and tertiary industries
		Government expenditures	Government expenditures	Reflecting government expenditure (100 million yuan)
		Rural residents’ consumption levels	Consumption	Reflect the expenditure of rural residents (yuan / person)
	Society	Population density	Population	Number of people per unit land area (person / km2)
		Human capital	Human capital	Years of education per capita (years)
		Population urbanization rate	Popu_urban	Proportion of urban population
		Land urbanization rate	Land_urban	Built up area (10000 square kilometers)
		Urban-rural income gap	Urban-rural_gap	Income of urban residents / Rural residents
		Rural electricity consumption	Electricity	Rural electricity consumption (100 million kwh)
		Mechanization level	Mechanization	Total power of agricultural machinery (10000 kW)
		Level of financial support for agriculture	Afinance	Local expenditure on agriculture, forestry and water affairs (100 million yuan)
	Environment	Land area available for crop planting	Seed	Sown area of crops (1000 HA)
		Reservoir capacity	Reservoir	Total reservoir capacity (100 million cubic meters)
		Soil erosion control	Erosin	Soil erosion control area (1000 HA)
		Grain yield	Grain	Grain yield per unit area (kg / HA)

Difference-in-differences

In 2010, to improve peasants’ income and reduce poverty, the Chinese government issued No. 1 central document to strengthen investment to rural infrastructure especially irrigation and drainage infrastructure and rural transportation infrastructure. Therefore, this paper will use a difference-in-differences (DID) model and use the investment as a quasi-natural experiment. The impact of rural infrastructure on poverty will be evaluated in this way. This paper selects nonethnic areas as the experimental group and ethnic areas as the control group. The reasons are as follows. Due to natural resource endowments and historical developments, ethnic areas fall far behind nonethnic areas in economic and social terms. According to that situation in China and the theory of Development Poles, the investments in rural infrastructure in China is also developed-region-oriented. Referring to [32], DID applies to this case if the policy in 2010 has relatively larger effects on nonethnic areas than ethnic areas. Theoretically, the parallel trend assumption of DID is satisfied. We will test the assumption in the Results section. The model is as follows:

Basic difference-in-difference model. where Eq (1) represents the basic DID model. The dummy variable du = 1 if belonging to nonethnic areas with better natural resource and faster development, and 0 otherwise. dt2010 = 1 in the years after the policy stimulus in 2010. The coefficient, α1, therefore indicates the impact of rural infrastructure investment on rural poverty rate. The policy under study was implemented in 2010, which is the treatment period. du ⋅ dt2010 is the interaction between the group and time dummy variables; its coefficient denotes the net effect of policy implementation, which is of great importance.λ and v represent the province fixed effects and the time fixed effects, respectively.

Continuous difference-in-difference model.

In Eq (2), the group dummy variable du is replaced by the continuous variable X, which represents the different types of infrastructure construction in this article. That is called continuous DID because α represents the net effect of the change in each rural infrastructure on poverty [33].

Model of parallel trends. where the dummy variables of D equal zero, except for the following: D− equals 1 for experimental groups in the jth year before policy, while D+ equals 1 for experimental groups in the jth year after policy. We examine the trends of poverty rates before and after 3 years of the policy. Also, we add the year of the policy and estimate the dynamic effect of policy on poverty rates relative to the year of policy. More importantly, we can test the parallel trend of DID. If β (k = 1,2,3) before the policy is not significant, then the parallel trend assumption is satisfied. This means that the experimental group and the control group have similar trends before the policy. Similarly, if β (k = 4,5,6,7) after the policy is significant, it means that differences arise between the experimental group and the control group after the policy implementation.

Mediating effect model. Referring to Wen et al.(2004) and Wen et al.(2014) and to test the influence mechanisms of different types of rural infrastructure on poverty, this paper further uses the mediating effect model [34, 35]. Eqs (4)–(6) are the regression equations set by the intermediary effect test procedure. where the first step is to regress the dependent variable y on the independent variable X to confirm that X is a significant predictor of y in Eq (4). Then regress the mediator M on X to confirm that X is a significant predictor of M in Eq (5). Finally, regress Y on both X and M to confirm that the M is a significant predictor of Y in Eq (6). The test procedure of mediating effect test is shown in Fig 2 [36]. The masking effect indicates X may show no effect on y on the whole, as positive and negative offsets each other. Fig 2 plots the test procedure of mediating effect.

Fig 2

Test procedure of mediating effect.

The 3 test procedure of mediating effect based on Eqs (4)–(6).

Results

Descriptive statistical analysis

The descriptive statistics for the main variables are shown in Table 2. To further ensure the stability of the data and reduce problems such as collinearity and heteroscedasticity in the model, this paper conducts logarithmic transformation on the variables in the data except for the share of the population receiving the rural minimum living guarantee, industrial structure, human capital, population urbanization rate, and urban-rural gap. “ln” before a variable name indicates that logarithmic transformation has been carried out.

Table 2

Descriptive of the variables in the model.

Variable	Obs	Mean	Std. Dev.
ln_Poverty_num	399	1.409	0.426
Poverty_rate	407	0.062	0.051
ln_Irrigation	416	7.487	0.761
ln_Road	324	7.412	1.061
ln_Primary industry	416	6.986	0.944
ln_Tertiary industry	416	8.232	1.092
ln_Pgdp	416	10.06	0.72
ln_Pindustrialization	416	-0.074	0.781
ln_ Industrial structure	416	0.143	0.06
ln_ Gconsume	416	7.192	0.921
ln_ Rconsume	416	8.438	0.673
ln_ Population	416	5.191	1.149
Human_capital	416	8.414	0.782
Popu_urban	409	0.474	0.105
Land_urban	416	7.018	0.78
Urban-rural_gap	416	2.995	0.58
ln_ Electricity	416	4.604	1.325
ln_ Mechanization	416	7.789	0.786
ln_ Afinance	416	5.146	1.136
ln_ Seed	416	8.501	0.727
ln_ Reservoir	416	5.273	0.839
ln_ Erosin	416	8.034	0.794
ln_ Grain	416	8.487	0.216

Result of basic model

Using Eq (1), we first test the effect of rural infrastructure investment on poverty reduction and control for the fixed effects of provinces and years, as well as the control variables at the provincial level. First, following the benchmark from the DID model shown in Eq (1), the poverty incidence rate is taken as the explanatory variable. The effect is estimated by a multiway fixed effect model. Since the investment plan was promoted in 2010, this paper selects 2010 as the treatment period. The estimated results are shown in Table 3. In Table 3, Column (1) does not include control variables, while Column (2) does. The coefficient on the interaction between the time dummy variable and group dummy variable is the focus of our attention and reflects how rural infrastructure affects poverty. In Column (2), the interaction between the group dummy variable and the dummy variable for 2010 is significantly negative. This result suggests that rural infrastructure construction in 2010 effectively helped alleviate poverty in China.

Table 3

Result of benchmark model of DID.

	(1) Poverty_rate	(2) Poverty_rate
duxdt2010	-0.0485***	-0.0436***
	(-4.71)	(-5.46)
Control for Year	Yes	Yes
Control for Province	Yes	Yes
Control Variables	No	Yes
Constant	0.077***	1.62*
	(23.01)	(2.05)
Observations	407	400
R-squared	0.8203	0.8791

Note:t statistics in parentheses,

* p <0.10,

** p <0.05,

*** p <0.01

However, this conclusion does not reveal irrigation and drainage infrastructure and rural transportation infrastructure can influence rural poverty. To address these questions, this article will further use the continuous DID model to investigate the effects of irrigation and drainage infrastructure and rural transportation infrastructure on poverty reduction. The test results based on Eq (2) are shown in Table 4. As shown in this table, Columns (1) and (3) show the estimation result without control for variables. Added with control variables to increase the robustness of the results, Columns (2) and (4) reveal how irrigation and drainage infrastructure and rural transportation infrastructure influence poverty, respectively. Column (2) shows that the coefficient on interaction term between irrigation and drainage infrastructure and the year 2010 is negative and has a significance level of 5%. The results prove that irrigation and drainage infrastructure can effectively reduce poverty. In Column (4), the interaction coefficient between rural transportation infrastructure and the year 2010 is not statistically significant. This result indicates that unlike the other two types of rural infrastructure, rural transportation has no overall significant impact on poverty.

Table 4

Result of continuous model of DID.

	(1) Poverty_rate	(2) Poverty_rate	(3) Poverty_rate	(4) Poverty_rate
ln_Irrigation×t2010	-0.00999	-0.0177**
	(-1.66)	(-2.75)
ln_Rroad×t2010			0.0075**	-0.0031
			(2.21)	(-0.75)
Control for Year	Yes	Yes	Yes	Yes
Control for Province	Yes	Yes	Yes	Yes
Control Variables		Yes		Yes
Constant	0.0954***	0.968	0.0456***	0.574
	(4.69)	(0.96)	(3.05)	(0.54)
Observations	407	400	311	311
R-squared	0.7818	0.8690	0.8367	0.8588

To check the robustness of above results, we display the result of parallel trend test, instrumental variable estimation and changing the dependent variable. Table 5 displays the parallel trend test results. Referring to [37], we choose the period three years before and after the implementation of the policy (2010) to test for common trends. As shown in Table 5, control variables are included in Columns (2) while not in Columns (1) to ensure the robustness of the regression results. Column (2) shows that the regression coefficients on the interaction terms between the time dummy variables and group dummy variables are not significant in 2007, 2008 or 2009. This finding shows that before the implementation of rural infrastructure construction in 2010, the incidence of poverty in ethnic areas and nonethnic areas experienced the same trend. The fluctuations in the two groups are not significantly different. The experimental group and the control group conform to the DID common trend assumption. Moreover, after 2010, the regression coefficients on the interaction terms between the time dummy variables and group dummy variables are significantly negative in 2011, 2012 and 2013. This result indicates that the trends in poverty in the experimental group and the control group were different after the implementation of the policy in 2010 and we cannot reject the test of parallel trends in pretreatment periods.

Table 5

Parallel trend test.

	(1)	(2)
Pre3	-0.00679	-0.0119
	(-0.97)	(-1.00)
Pre2	-0.00649	-0.00771
	(-0.83)	(-0.80)
Pre1	-0.0108*	-0.00995
	(-1.75)	(-1.23)
Current	-0.0291	-0.0240*
	(-1.69)	(-1.72)
Aft1	-0.0344**	-0.0242**
	(-2.12)	(-2.28)
Aft2	-0.0371**	-0.0269**
	(-2.33)	(-2.35)
Aft3	-0.0384**	-0.0239***
	(-2.67)	(-2.88)
Control for Year	YES	YES
Control for Province	YES	YES
Control Variables		YES
Constant	0.0693***	0.441
	(20.01)	(0.45)
Observations	407	400
R-squared	0.7656	0.8395

Note:t statistics in parentheses,

* p <0.10,

** p <0.05,

*** p <0.01

Fig 3 plots the impact of implementation of the policy on rural poverty rates. We consider a 6-year window, spanning from 3 years before the until 3 years after deregulation. The dashed lines represent 95% confidence intervals. Specifically, we report estimated coefficients in Eq (3). Fig 3 illustrates the same key points as Table 5: there is no trend in poverty rates in two groups prior to the policy. Next, note that poverty rates fall immediately after policy, such that Aft1 is negative and significant at the 5% level. Thus, the mechanisms and channels connecting rural infrastructure with the infrastructure must be fast acting.

Fig 3

The dynamic impact of rural infrastructure on poverty.

The impact of implementation of the policy on rural poverty rates 3 years before the until 3 years after deregulation.

Table 6 shows the result of instrumental variable estimation. The instrumental variables are lagged irrigation and drainage infrastructure and lagged rural transportation infrastructure. Columns (1) and (3) in Table 6 do not include the control variables and reflect the impact of irrigation and drainage infrastructure and rural transportation infrastructure on poverty, respectively. Columns (2) and (4) in Table 6 include the control variables to the models from Columns (1) and (3), respectively. In Column (2), the coefficient on the interaction terms between ln_Irrigation and dummy variable t2010 is significantly negative, which is consistent with the above results. In Column (4), the coefficient on the interaction term between ln_Road and dummy variable t2010 is not significant, which is also consistent with the previous results. Once again, irrigation and drainage infrastructure can promote poverty reduction. These results show that the findings from the above analysis are robust. However, the effect of rural transportation infrastructure on poverty reduction is not clear.

Table 6

Result of two-stage least square method.

	(1) Poverty_rate	(2) Poverty_rate	(3) Poverty_rate	(4) Poverty_rate
ln_Irrigation×t2010	-0.00984***	-0.0166***
	(-2.94)	(-4.89)
ln_Road×t2010			0.00714***	0.0027
			(3.61)	(1.26)
Control for Year	YES	YES	YES	YES
Control for Province	YES	YES	YES	YES
Control Variables		YES		YES
Constant	0.00385	0.771	0.0577***	0.262
	(0.77)	(1.46)	(19.01)	(0.46)
Observations	382	378	286	286
R-squared	0.6715	0.8032	0.3844	0.5273

Note:t statistics in parentheses,

* p <0.10,

** p <0.05,

*** p <0.01

Table 7 presents the regression results of replacing rural poverty rate with the size of the rural poor population in Eqs (1) and (2). To enhance robustness, the logarithm of the rural poor population is taken. Column (1) shows the estimation results based on Eq (1). Column (2) includes control variables in the model from Column (1). Columns (3) and (5) are the estimation results from Eq (2), showing the impact of irrigation and drainage and rural transportation infrastructure on the size of the rural poor population, respectively. Columns (4) and (6) add control variables based on (3) and (5). As shown in Columns (2), (4) and (6) in Table 6, the dependent variable is replaced by the size of the rural poor population. The coefficients on interaction term between du and t2010 is -0.168 and has a significance level of 5%. The coefficients on interaction term between ln_Irrigation and t2010 is -0.0579 sand has a significance level of 10%. While the coefficients on interaction term between ln_Road and t2010 is not significant. These results are consistent with the previous regression results.

Table 7

Result of changing the dependent variable.

	(1) ln_Poverty_num	(2) ln_Poverty_num	(3) ln_Poverty_num	(4) ln_Poverty_num	(5) ln_Poverty_num	(6) ln_Poverty_num
du×t2010	-0.354***	-0.168***
	(-4.86)	(-2.92)
ln_Irrigation×t2010			-0.0309	-0.0579*
			(-0.63)	(-1.89)
ln_Road×t2010					0.0189	0.0119
					(1.03)	(0.94)
Control for Year			YES	YES	YES	YES
Control for Province			YES	YES	YES	YES
Control Variables				YES		YES
constant	1.527***	-3.353	1.515***	-6.417	1.483***	-15.27**
	(62.85)	(-0.29)	(9.00)	(-0.60)	(18.41)	(-2.57)
Observations	399	392	399	392	311	311
R-squared	0.6887	0.7559	0.6569	0.7532	0.7601	0.8589

Note:t statistics in parentheses,

* p <0.10,

** p <0.05,

*** p <0.01

Result of mechanisms test

From the above regressions, we can see that irrigation and drainage infrastructure can promote poverty reduction, but their poverty reduction mechanisms need to be verified. Although rural transportation infrastructure has no significant impact on poverty, it remains to be seen whether it has a “masking effect” on poverty. Using the mediating effect test model proposed by [35, 36], this section explores the internal influence mechanisms of the two types of rural infrastructure on poverty. The results are shown in Table 8.

Table 8

Result of mediating effect.

	Irrigation and drainage infrastructure			Rural transportation infrastructure
	(1) Poverty_rate	(2) ln_Primary industry	(3) Poverty_rate	(4) Poverty_rate	(5) ln_Tertiary industry	(6) Poverty_rate
ln_Irrigation×t2010	-0.0177***	-0.0061	-0.0171***
	(-5.45)	(-0.87)	(-5.37)
ln_Primary industry			0.0927***
			(3.84)
ln_Road×t2010				0.0023	0.0142***	0.0056**
				(0.96)	(3.77)	(2.40)
ln_Tertiary industry						-0.2309***
						(-6.13)
Control for Year	YES	YES	YES	YES	YES	YES
Control for Province	YES	YES	YES	YES	YES	YES
Control Variables	YES	YES	YES	YES	YES	YES
constant	1.128**	1.89***	0.9534*	-0.495	-9.265***	-2.635***
	(2.12)	(1.61)	(1.82)	(-0.69)	(-8.32)	(-3.48)
Observations	400	400	400	311	311	311
R-squared	0.869	0.998	0.8724	0.884	0.9994	0.8989

Note:t statistics in parentheses,

* p <0.10,

** p <0.05,

*** p <0.01

In Table 8, Columns (1)-(3) examine the poverty reduction mechanisms behind irrigation and drainage infrastructure. The growth of agriculture, forestry, animal husbandry and fisheries is selected as the intermediate variable for irrigation and drainage infrastructure. The mediating effect test procedure is shown in Fig 1. As shown in Column (1) in Table 8, the construction of irrigation and drainage infrastructure has a significant negative impact on poverty, so we continue to carry out the intermediary effect test. In Column (2), the coefficient on irrigation and drainage infrastructure, which estimates the infrastructure’s effect on the intermediary variable of rural economic growth, is not significant. In Column (3), the coefficient on irrigation and drainage infrastructure is significantly negative, while that on the growth of agriculture, forestry, animal husbandry and fisheries is significantly positive. Therefore, further bootstrap tests are needed, and the results are shown in Table 9. The coefficients on both the direct and indirect effects of irrigation and drainage infrastructure are negative. However, the confidence interval for the indirect effect after correction is [-.0016786, .001194]. The value 0 is included in the interval, indicating that the indirect effect is not significant. The confidence interval for the direct effect after correction is [-.0300072, -.0099262], which excludes 0.

Table 9

Result for bootstrapping.

		Observed Coef.	Bias	Bootstrap Std. Err.	Normal-based [95% Conf. Interval]
ln_Irrigation×t2010	Indirect effect	-0.00057362	-0.0001354	0.00082755	[-.0021794, .0010257] (P)
					[-.0016786, .001194] (BC)
	Direct effect	-0.0171468	0.0004506	0.00400739	[-.0248126, -.009654] (P)
					[-.0300072, .-.0099262] (BC)
ln_Road×t2010	Indirect effect	-0.0032976	0.0000615	0.00103574	[.0053486, .0013547] (P)
					[.0056422, .0015639] (BC)
	Direct effect	-6.71096	0.0003991	0.00227134	[.0015815, 0.0104943] (P)
					[.0003989, .0098849] (BC)

(P) percentile confidence interval; (BC) bias-corrected confidence interval

In Table 8, Columns (4)-(6) are used to test the impact of rural transportation infrastructure on poverty. According to the regression results in Table 4, the overall impact of rural transportation infrastructure on poverty is not significant. However, how rural transportation infrastructure influences poverty has not been confirmed. Therefore, this section tests whether the transportation infrastructure has both indirect and direct impacts on poverty that create a masking effect. Due to the convenience brought by rural transportation infrastructure, on the one hand, the loss of the rural population leads to the hollowing out of rural areas and aggravates poverty. At the same time, the rural population mostly flows to the labor-intensive tertiary industry in cities and towns. This situation will lead to increases in wages and remittances to rural areas, reducing poverty. Therefore, this section selects the growth of the tertiary industry as the intermediary variable explaining how rural transportation infrastructure affects poverty. The results are shown in Table 8. The coefficient on rural transportation infrastructure in Column (4) is not significant. According to Fig 1, the impact of rural infrastructure on poverty may be masked. In Column (5), the coefficient on rural transportation infrastructure, which estimates its effect on the tertiary industry, is significantly positive. In Column (6), the coefficient of the growth of the tertiary industry on rural poverty is significantly negative, and the impact of rural transportation infrastructure on rural poverty is significantly positive. Table 9 displays the bootstrapping test results for rural transportation infrastructure. The results show that the corrected direct effect is significantly positive at the 95% level. The indirect effect is significantly negative at the 95% level.

Discussion

This study develops a theoretical framework of how irrigation and drainage infrastructure and rural transportation infrastructure influence poverty. Along with provincial panel data on China from 2002 to 2017, this paper uses a basic DID model to examine how investment policy on rural infrastructure influences poverty in rural China. Then a continuous DID model is used to investigate how two of the most important rural infrastructures, irrigation and drainage infrastructure and rural transportation infrastructure, influence rural poverty in China. Instrumental variable estimation and a change in the independent variable are used to enhance the robustness of the results. In addition, the mediating effect model is used to investigate the influence mechanisms of irrigation and drainage infrastructure and rural transportation infrastructure on rural poverty in China.

This study found that rural infrastructure construction in 2010 can help reduce poverty in China. This study also revealed that irrigation and drainage infrastructure can directly reduce poverty, which is consistent with previous research [38-40]. Therefore, H1a is proved, and there is no evidence to support H1b. And there is no enough evidence for the indirect effect on poverty reduction in this study.

Besides, existing research on impact of rural infrastructure on poverty usually claims that rural infrastructure especially rural road may economic growth and therefore reduce poverty [41, 42]. This study found that rural transportation infrastructure may aggravates rural poverty although it can reduce poverty by promoting the growth of the tertiary industry, which creates an “illusion” of no impact. Rural transportation infrastructure directly leads to rural hollowing out and aggravates rural poverty, indicating there is the siphon effect. In addition, rural transportation infrastructure promotes the growth of the tertiary industry, therefore indirectly promoting rural poverty reduction. Accordingly, the positive and negative effects offset each other, resulting in these effects being masked overall. Therefore, H2a and H2b are proved. These findings corroborate the ideas of Asher and Novosad(2020), who maintained that transportation infrastructure is costly investment and may not necessarily cause positive economic impacts [22].

Given above discussions, this article makes the following theoretical and empirical contributions to the literature as mentioned earlier: First, theoretically, we enrich literature on the poverty-reducing effect of irrigation and drainage infrastructure on poverty. Besides, this paper contributes to a wide literature by clarifying the poverty-aggravating and poverty-reducing and effect of rural transportation infrastructure and the underlying mechanisms. Second, empirically, the basic and continuous DID models and the mediating effect model are used to estimate the effects of rural infrastructure on poverty. Besides, two-stage least square method has been adopted to deal with the endogeneity problem, supplementing literature on estimating the causal relationship between rural infrastructure and poverty. Additionally, with updated and more extensive data, we empirically confirm the direct poverty-reducing effect of irrigation and drainage infrastructure and the direct poverty-aggravating and indirect poverty-reducing and effect of rural transportation infrastructure. These findings expand empirical research on the relationship between irrigation and drainage infrastructure and rural transportation infrastructure and poverty.

Conclusions, implications and future research

Conclusions

The final conclusions are as follows: First, irrigation and drainage infrastructure can effectively reduce rural poverty in China, and rural transportation infrastructure has no clear impact on poverty on the whole. Second, irrigation and drainage infrastructure can reduce poverty directly, and no evidence of indirect influence mechanisms has been found in this study. Third, for rural transportation infrastructure, it may bring about the siphon effect and directly aggravate rural poverty in China on the one hand. On the other hand, it can promote economic growth by driving the development of the tertiary industry and indirectly promotes poverty reduction in rural China through the trickle-down effects of economic growth. However, the two opposite effects offset each other, and rural transportation infrastructure has no obvious impact on poverty overall.

Implications

Under the background of achieving the Sustainable Development Goals and realizing the aim of no poverty and facing the challenging of frequent extreme weather, this paper has clear policy significance for the global government on how to further strengthen the current poverty reduction effects of various rural infrastructure, consolidate the existing achievements in poverty alleviation and prevent poverty levels from increasing. In the context of urbanization, developing countries, represented by China, mostly focus on rural transportation to achieve poverty reduction. However, this paper finds that although rural transportation infrastructure can promote economic growth and achieve poverty reduction, it can also promote the decline of rural areas, and the poorest people are abandoned in rural areas. Moreover, the lack of irrigation and drainage infrastructure has a significant direct effect on poverty. In addition, under the background that COVID-19 may exist for a long time around the world, some migrant workers have chosen to return to their hometown to engage in agricultural production and entrepreneurship, and even participate in the industrialization of agriculture. In this situation, the government is required to pay more attention to the construction of irrigation and drainage infrastructures which is of great importance to people’s livelihood in rural areas. In addition, the government needs to increase financial input to poor rural communities to consolidate the achievements in poverty alleviation. Second, the reason why rural transportation infrastructure may aggravate poverty is that it can accelerate the outflow of the rural population. This means that although such infrastructure can promote urbanization, it may lead to rural hollowing out. However, urbanization is an inevitable trend in economic development worldwide. The government should, in combination with this trend, reposition rural functions, support local development projects with comparative advantages, and provide equal public services for the poorest rural populations.

Limitations and future research

Although we have figured out the effect and mechanisms of irrigation and drainage and rural transportation infrastructure on poverty with empirical models, there are some limitations in this study. First, this article mainly focused on the irrigation and drainage and rural transportation infrastructure and did not consider other rural infrastructure such as electricity and telecommunications infrastructure, which are also important to poverty reduction. Future research can explore the impact and mechanism of electricity infrastructure on poverty. Second, the construction of rural infrastructure, especially transportation infrastructure, often has spatial spillover effects on nearby regions. The spillover effects were not considered in this article owing to limitations of paper length. Future study can consider both time and space effects of rural infrastructure on poverty by applying spatial Difference-in-Difference models [43, 44]. Third, this study only adopted panel data in different provinces during 2002–2017 due to data availability. Data covering a longer period and smaller units such as macro-county and even micro-household level can help provide a more accurate and convinced analysis on the impact of rural infrastructure on poverty, which can be studied in future research.

(DTA)

Click here for additional data file.

24 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Jun Yang

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

2. Thank you for stating the following in your Competing Interests section:

[NO authors have competing interests].

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

4. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ.

Additional Editor Comments:

The authors estimated basic and continuous difference-in-differences (DID) models to investigate the preliminary impact

of irrigation and drainage infrastructure and rural transportation infrastructure on poverty and further explores the influence mechanisms of these rural infrastructures on poverty by using the mediating effect model. The topic is interesting. Authors are suggested to make the following changes to the paper.

The introduction should further highlight the scientific problems, motivations, and possible innovations of the paper.

It is suggested to update the data of the paper, so that the results may be more meaningful.

Discussion: The contribution of the paper should include both theoretical and empirical aspects, and the author is advised to rewrite

from these two aspects.

It should contain a summary of current research gaps and an outlook for future research.

The study examined the effects of rural transportation infrastructure on poverty using DID models and mediating effect model. The methods are sound, the analyses are reasonable, and the structure is complete. However, there are still some small problems that need to be fixed.

1. The citation format in the text needs to be checked and in accordance with the standard of the journal.

2. “This increase in poverty may offset the poverty reduction effect of rural infrastructure, which is called the masking effect of rural infrastructure.” (Line80-81) This sentence requires literature support.

3. The clarity of Figure 1-3 needs to be improved. There may be problems with Figure 1-2 displays that need to be checked.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The study examined the effects of rural transportation infrastructure on poverty using DID models and mediating effect model. The methods are sound, the analyses are reasonable, and the structure is complete. However, there are still some small problems that need to be fixed.

1. The citation format in the text needs to be checked and in accordance with the standard of the journal.

2. “This increase in poverty may offset the poverty reduction effect of rural infrastructure, which is called the masking effect of rural infrastructure.” (Line80-81) This sentence requires literature support.

3. The clarity of Figure 1-3 needs to be improved. There may be problems with Figure 1-2 displays that need to be checked.

Reviewer #2: The authors estimated basic and continuous difference-in-differences (DID) models to investigate the preliminary impact

of irrigation and drainage infrastructure and rural transportation infrastructure on poverty and further explores the influence mechanisms of these rural infrastructures on poverty by using the mediating effect model. The topic is interesting. Authors are suggested to make the following changes to the paper.

The introduction should further highlight the scientific problems, motivations, and possible innovations of the paper.

It is suggested to update the data of the paper, so that the results may be more meaningful.

Discussion: The contribution of the paper should include both theoretical and empirical aspects, and the author is advised to rewrite

from these two aspects.

It should contain a summary of current research gaps and an outlook for future research.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

7 Mar 2022

Dear Editors and Reviewers,

We appreciate the opportunity to modify our manuscript according to the critical comments concerning our manuscript entitled “Rural Infrastructure and Poverty in China” (Manuscript Number: PONE-D-22-00776). We would like to thank the editors and the review team for your constructive comments and suggestions on our paper. They are all valuable and very helpful for revising and improving our paper, and simultaneously they are of great significance to give instructions to our research. We have studied the comments carefully and made revisions which we hope to meet with approval.

With a concerted effort to address all of the valuable advice, we have revised the paper to a more accurate and concise style including citation format and figures, highlight the scientific problems, motivations, and possible innovations, rewrite the theoretical and empirical contribution of the paper, provided additional information for the dataset, and have updated the referencing literature. We believe that the recommended changes have dramatically enhanced the manuscript’s quality and its contribution, and we hope you will agree.

Also, it should be noted that we highlight the important changes in yellow in our revised manuscript in pdf file. Below, we provide our detailed responses in tabular form to explain how your points have been included in the revision. As you will see, we have made every possible attempt to address your concerns, and we hope you will find our revision acceptable. Once again, thank you for helping us to improve the manuscript greatly.

Responses to Journal Requirements

Comment 1

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Responses 1

Thanks for the guideline. The revised manuscript, as well as the title page, and supplementary materials are all arranged according to the PLOS ONE style.

Comment 2

Thank you for stating the following in your Competing Interests section:

[NO authors have competing interests].

Please complete your Competing Interests on the online submission form to state any Competing Interests. If you have no competing interests, please state "The authors have declared that no competing interests exist.", as detailed online in our guide for authors at http://journals.plos.org/plosone/s/submit-now

This information should be included in your cover letter; we will change the online submission form on your behalf.

Responses 2

Thanks for the reminder. We have stated “The authors have declared that no competing interests exist” in the revised cover letter.

Comment 3

In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

Responses 3

Thanks for the reminder. Apology for our negligence. The minimal data set underlying the results described in your manuscript can be found at the website of National Bureau of Statistics of China (http://www.stats.gov.cn/), China Statistical Yearbook (https://data.cnki.net/yearbook/Single/N2021110004) and China Rural Statistical Yearbook (https://data.cnki.net/yearbook/Single/N2021120010). In addition, we have updated our Data Availability statement in the cover letter to reflect the information we provide. Besides, In the revised revision, we upload the study’s minimal underlying data set as Supporting Information files to improve the research repeatability.

Comment 4

PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ.

Responses 4

Thanks for the reminder. I have updated my information and fetched the ORCID iD.

Responses to Reviewer #1

Comment 1

The citation format in the text needs to be checked and in accordance with the standard of the journal.

Responses 1

Thanks a lot for the reminder. Apology for our negligence. In the manuscript we have revised the citation format adopted by PLOS. Besides, we also renewed the reference literature to prove the findings in this study and revised the reference style according to PLOS requirements, which has been highlighted in the revised manuscript.

Please refer to Reference part in the revised manuscript for more details.

Comment 2

“This increase in poverty may offset the poverty reduction effect of rural infrastructure, which is called the masking effect of rural infrastructure.” (Line80-81) This sentence requires literature support.

Responses 2

Thanks for this suggestion of supplementation. We feel sorry that we had neglected to provide enough literature support. We have modified this problem in the revised manuscript as follows:

“This increase in poverty may offset the poverty reduction effect of rural infrastructure, which we can call the masking effect, often used in mediation analysis in psychology [15-17].”

Please refer to Line 83-85, Page 3 in the revised manuscript for more details.

Comment 3

The clarity of Figure 1-3 needs to be improved. There may be problems with Figure 1-2 displays that need to be checked.

Responses 3

Thanks for the suggestion for improving the clarity and displaying problems of Figure 1-3 of this manuscript. By applying the PACE tool, we have checked and revised Figure 1-3 in TIFF files according to figure preparation guidelines adopted by PLOS.

Please see in the revised Figures attached with manuscript.

References for Reviewer #1

1. MacKinnon, D. P., Krull, J. L., Lockwood, C. M. (2000). Equivalence of the mediation, confounding and suppression effect. Prevention science, 1(4), 173-181.

2. Lin, B. Q. (2003). Economic growth, income inequality, and poverty reduction in People's Republic of China. Asian development review, 20(2), 105-124.

3. Lu, H., Zhao, P., Hu, H., Zeng, L., Wu, K. S., Lv, D. (2022). Transport infrastructure and urban-rural income disparity: A municipal-level analysis in China. Journal of Transport Geography, 99, 103292.

Responses to Reviewer #2

Comment 1

The introduction should further highlight the scientific problems, motivations, and possible innovations of the paper.

Responses 1

Thank you very much for this constructive suggestion. During the revision stage, we tried to further highlight the scientific problems, motivations, and possible innovations of the paper in introduction. Motivated by China’s great investment to irrigation and transportation infrastructure especially in rural areas and the miracle of poverty reduction, we proposed the scientific problems cover whether the investment to irrigation and transportation infrastructure in 2010 reduced poverty in practice, what roles irrigation and transportation infrastructure played in influencing poverty and what is the underlying mechanisms. The possible innovations of the paper are highlighted from the theoretical and empirical aspects. We also add literature to support the viewpoint. The scientific problems, motivations, and possible innovations are enriched as following:

“Given that China has invested heavily to irrigation and transportation infrastructure especially in rural areas and the miracle of poverty reduction in the past 40 years, three important issues arise and needs to be figured out. First, did the investment to irrigation and transportation infrastructure in 2010 reduce poverty in practice? Second, what roles did irrigation and transportation infrastructure play in influencing poverty, respectively? Third, what is the mechanism behind the impact? Some studies maintain that rural transportation infrastructure, may not have social benefits and succeed in reducing poverty if rural laborers choose to transfer to urban areas due to the lack of agglomeration of local economies [2]. On the contrary, rural poverty may be deepened because of the loss of young and middle-aged rural labor force and the decline of abandoned villages, which is called siphon effect [3]. Moreover, irrigation and drainage infrastructure may also fail to count in that case. Dealing with these three problems can help China’s government reasonably adjust investment structure in rural infrastructure and consolidate the achievements on poverty reduction, as well as provide experience for those developing economies troubled by poverty. In addition, under the threat of extreme weather, it is of particular importance to clarify the role of irrigation and drainage infrastructure in affecting poverty. Irrigation and drainage infrastructure, due to the characteristics of relatively lower profits and slower return, usually attracts much less investment than rural transportation infrastructure.

Given the above discussions, this research aims to evaluate the impact and mechanisms of rural irrigation and transportation infrastructure on poverty and makes innovations in several ways. First, this paper theoretically clarifies how irrigation and drainage infrastructure and rural transportation can affect poverty. We supplement theoretical literature concerning how irrigation and drainage infrastructure influences poverty and the underlying mechanism, which are rare in previous research. In addition, we add to a growing body of literature on how rural transportation influences poverty, focusing not on the poverty-reducing effect but also on the poverty-aggravating effect. Second, this paper helps enrich empirical evidence on how rural irrigation and transportation infrastructure influences poverty by applying the basic and continuous DID models and the mediating effect model. We use updated data to make the results more convincible and two-stage least square method to handle endogeneity problem.”

Please refer to Line 26-61, Page 2 in the revised manuscript for more details.

Comment 2

It is suggested to update the data of the paper, so that the results may be more meaningful.

Responses 2

Thanks a lot for your constructive suggestion of updating the data. We had tried to collect the latest data from the website of National Bureau of Statistics of China and China Statistical Yearbook, where authoritative data in China is provided. Unfortunately, one of the key control variables, Government expenditures, is only available by 2017 due to data availability. Therefore, data covering 2002-2017 is latest data we have access to. Besides, we think current panel data covering 16 years and 26 provinces is enough to support the findings of this study on how irrigation and drainage infrastructure and rural transportation affect poverty and the underlying mechanisms. But we are really grateful for your inspiring advice and we hope the data can updated in the future study so that the results may be more meaningful, which we highlight in the summary of current research gaps and an outlook for future research.

Please refer to Line 502-506, Page 16 in the revised manuscript for more details.

Comment 3

Discussion: The contribution of the paper should include both theoretical and empirical aspects, and the author is advised to rewrite from these two aspects.

Responses 3

Thanks for this suggestion. We have attempted to refine the contribution in discussion part, from the theoretical and empirical aspects, to present the importance of this research. The revised contribution is as following:

“Given above discussions, this article makes the following theoretical and empirical contributions to the literature as mentioned earlier: First, theoretically, we enrich literature on the poverty-reducing effect of irrigation and drainage infrastructure on poverty. Besides, this paper contributes to a wide literature by clarifying the poverty-aggravating and poverty-reducing and effect of rural transportation infrastructure and the underlying mechanisms. Second, empirically, the basic and continuous DID models and the mediating effect model are used to estimate the effects of rural infrastructure on poverty. Besides, two-stage least square method has been adopted to deal with the endogeneity problem, supplementing literature on estimating the causal relationship between rural infrastructure and poverty. Additionally, with updated and more extensive data, we empirically confirm the direct poverty-reducing effect of irrigation and drainage infrastructure and the direct poverty-aggravating and indirect poverty-reducing and effect of rural transportation infrastructure. These findings expand empirical research on the relationship between irrigation and drainage infrastructure and rural transportation infrastructure and poverty.”

Please refer to Line 438-452, Page 14 in the revised manuscript for more details.

Comment 4

It should contain a summary of current research gaps and an outlook for future research.

Responses 4

Thanks for this suggestion of supplementation. We feel sorry that we had neglected this part. We have modified this problem in the revised manuscript as follows:

“Although we have figured out the effect and mechanisms of irrigation and drainage and rural transportation infrastructure on poverty with empirical models, there are some limitations in this study. First, this article mainly focused on the irrigation and drainage and rural transportation infrastructure and did not consider other rural infrastructure such as electricity and telecommunications infrastructure, which are also important to poverty reduction. Future research can explore the impact and mechanism of electricity infrastructure on poverty. Second, the construction of rural infrastructure, especially transportation infrastructure, often has spatial spillover effects on nearby regions. The spillover effects were not considered in this article owing to limitations of paper length. Future study can consider both time and space effects of rural infrastructure on poverty by applying spatial Difference-in-Difference models [43, 44]. Third, this study only adopted panel data in different provinces during 2002-2017 due to data availability. Data covering a longer period and smaller units such as macro-county and even micro-household level can help provide a more accurate and convinced analysis on the impact of rural infrastructure on poverty, which can be studied in future research.”

Please refer to Line 491-506, Page 15-16 in the revised manuscript for more details.

References for Reviewer #2

1. Xu, Z., Sun, T. (2021). The Siphon effects of transportation infrastructure on internal migration: evidence from China’s HSR network. Applied Economics Letters, 28(13), 1066-1070.

2. Donaldson, D. (2018). Railroads of the Raj: Estimating the impact of transportation infrastructure. American Economic Review, 108(4-5), 899-934.

3. Lipton, M., Litchfield, J., Faures, J. M. (2003). The effects of irrigation on poverty: a framework for analysis. Water policy, 5(5-6), 413-427.

4. Unver, O., Wahaj, R., Lorenzon, E., Mohammadi, K., Osias, J. R., Reinders, F., ... Sangjun, I. M. (2018). Key and smart actions to alleviate hunger and poverty through irrigation and drainage. Irrigation and Drainage, 67(1), 60-71.

5. Choi, K. S., Labhsetwar, V. K. (2021). Sustainable agricultural growth for rural development in Asia: a review. Irrigation and Drainage, 70(3), 470-478.

6. Aschauer, D. A. (1989). Is public expenditure productive? Journal of monetary economics, 23(2), 177-200.

7. Ke, X., Chen, H., Hong, Y., Hsiao, C. (2017). Do China’s high-speed-rail projects promote local economy? —New evidence from a panel data approach. China Economic Review, 44, 203-226.

8. Dube, J., Legros, D., Th´eriault, M., Des Rosiers, F. (2014). A spatial difference-in-differences estimator to evaluate the effect of change in public mass transit systems on house prices. Transportation Research Part B: Methodological, 64, 24-40.

9. Delgado, M. S., Florax, R. J. (2015). Difference-in-differences techniques for spatial data: Local autocorrelation and spatial interaction. Economics Letters, 137, 123-126.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

14 Mar 2022

18 Mar 2022

Thank you very much for your appreciating the manuscript. As you suggested, we have used the LaTeX template provided by PLOS and compiled the LaTeX files as PDF to ensure the format of the main text should be coherent with the requirement of Plos One.

Besides, we have revised the reference style as outlined in the ICMJE sample references, which is required by Plos One.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

23 Mar 2022

Rural Infrastructure and Poverty in China

PONE-D-22-00776R2

Dear Dr. Wu,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Jun Yang

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Accept

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: It is recommended for publication.

Reviewer #2: The authors have adequately addressed comments raised in a previous round of review and I feel that this manuscript is now acceptable for publication.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

4 Apr 2022

PONE-D-22-00776R2

Rural Infrastructure and Poverty in China

Dear Dr. Wu:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jun Yang

Academic Editor

PLOS ONE