Effect of Micro-Credit for Poverty Alleviation on Income Growth and Poverty Alleviation—Empirical Evidence from Rural Areas in Hebei, China

Abstract

Micro-credit for poverty alleviation is an important financial measure of targeted poverty reduction and rural revitalization in China. This paper employs the OLS model and Logit model to empirically test the effect of micro-credit for poverty alleviation on the income level and stability of income growth of farmers based on the field survey data of 458 registered poverty-stricken farmer households in Fuping County and Quyang County of Hebei Province. The results suggest that micro-credit for poverty alleviation can increase farmers’ income, stabilize the growth of their income, and exert significant short-term and long-term effects on income growth and poverty alleviation. The specialized farmer cooperatives, the scale of production and operation, the proportion of family labor force, and the education level of the head of the farmer’s household exert a significantly positive effect on the farmers’ income and the stable growth of their income. There is a significant interaction between micro-credit for poverty alleviation and specialized farmer cooperatives. The physical conditions of family members exert a negative effect on the stable growth of their income, and other financing channels have no significant effect.

1. Introduction

Financial poverty alleviation is used as an important antipoverty policy tool for developing countries [1,2]. With the successful completion of targeted poverty alleviation and the gradual promotion of rural revitalization strategy, the system innovation and system reform of rural finance has become increasingly important in China. Micro-credit for poverty alleviation plays a crucial role in the five-year transition period of connecting targeted poverty alleviation with a rural revitalization strategy. There are, however, some functional and systemic shortages in traditional financial poverty alleviation, such as difficulties in aiming at the targeted customers [3,4,5], complex transaction procedures [6,7], and high transaction cost [8,9], and thus the customization of micro-credit for poverty alleviation is not only an important financial measure for targeted poverty alleviation but also a vital channel for accurately connecting financial resources with the capital needs of small farmers. Sustainable financial poverty alleviation is the basic requirement for implementing targeted poverty alleviation and a rural revitalization strategy, and the healthy development of micro-credit for poverty alleviation is the foundation for sustainable financial poverty alleviation. The current mode of micro-credit for poverty alleviation in China is insufficient in terms of sustainability, so it is necessary to transform to an “inclusive” model and to gradually adjust the policy elements at procedure and operation levels so as to raise the serviceability and quality level of micro-credit for poverty alleviation.

The poverty reduction effect of micro-credit for poverty alleviation in developing countries has been an important part of academic research, and, currently, more urgent attention is being paid to the relationship between micro-credit for poverty alleviation and the stable income growth of farmers. The remainder of this paper is organized as follows: Section 2 details a review of the literature; Section 3 hosts a theoretical analysis; Section 4 is a description of the data and construction of the model; Section 5 outlines the analysis of the regression results and robustness tests; Section 6 hosts a discussion; and the conclusion and policy implications are provided lastly.

2. Literature Review

Research into micro-credit for poverty alleviation started with the practice of micro-credit for poverty alleviation in Bangladesh in the late 1970s, and its use began to increase day by day in the 1990s [10,11,12]. Poverty in poor areas can be alleviated through micro-credit, and this solution is supported by empirical evidence from seven developing countries [13]. The research of Amin et al. [14] showed that the provision of micro-credit by rural NGOs in Bangladesh could reduce poverty. Schrieder and Sharma [15] demonstrated that access and participation in micro-credit could enhance the welfare of the poor in developing countries. With the development of poverty alleviation in developing countries, especially from 2014 to date, a large number of bodies of research on micro-credit for poverty alleviation is emerging, with new progress being made in terms of researching breadth and depth.

There are different views regarding whether micro-credit for poverty alleviation can effectively affect income growth and poverty alleviation. Some research has advocated that micro-credit for poverty alleviation can effectively alleviate the poverty of farmers. Micro-credit for poverty alleviation is helpful for the poverty-stricken population to continue and expand their production and operation and further boost their income and improve their life quality [2,16,17,18]. Micro-credit for poverty alleviation is dynamic, which can continuously and accurately target the credit needs of the poverty-stricken population and constantly improve the comprehensive services and utility of information technology of micro-credit for poverty alleviation in order to have an effect of income growth and poverty alleviation. Such elements as line of credit, organizational characteristics [19], agricultural operation status [3], loan capitalization tendency [20], and socio-economic [5,21], political, and cultural factors [22,23] exert significant influences on obtaining micro-credit for poverty alleviation and alleviating the poverty of farmers. Micro-credit for poverty alleviation can effectively increase the income of the poverty-stricken population, especially women [24] and poverty-stricken farmer households with capital needs. However, as time passes, the effect of micro-credit for poverty alleviation on income growth and poverty alleviation is weakening, which can be confirmed by diminishing the marginal effect of income growth of poverty-stricken farmers with loans.

Some other researchers suggest that micro-credit for poverty alleviation fails to effectively increase farmers’ income, or that its effect is uncertain [25,26,27]. Poor households become poorer by the additional burden of debt. Micro-credit for poverty alleviation only provides a platform for poverty-stricken farmers to receive a loan, and the platform has limitations in terms of resolving the development of poverty-stricken farmers. The research of Karlan and Zinman [28], Angelucci et al. [22], Crepon et al. [29], and Banerjee et al. [30] has showed that increment in micro-credit cannot alleviate poverty or effectively increase family income. Augsburg et al. [31] and Tarozzi et al. [32] demonstrated the same phenomenon using a randomized controlled trial (RCT). According to the research of Thanh et al. [26], rural micro-credit has little impact on farmers’ income growth at the macro level, and this impact is even less than expected. Meanwhile, according to some empirical research [1,7], poverty-stricken farmers rely more on informal credit because there is rarely credit available for farmers, and accurate credit support to poverty-stricken farmers is not frequently supplied to help farmers out of poverty; rural financial size is favorable for poverty alleviation, but rural financial efficiency negatively impacts poverty alleviation.

Additionally, some studies have shown that the impact of micro-credit for poverty alleviation varied by income group, and that this program benefited more the moderately poor compared to the hardcore poor [33,34]. The use of micro-credit for poverty alleviation deviates from its target and thus weakens its effect on poverty alleviation. Credit sources fail to help poverty-stricken farmers out of poverty in terms of the original intention of micro-credit for poverty alleviation, meaning it cannot be fully obtained by poverty-stricken households, and these problems have not been solved effectively. The target of the micro-credit for poverty alleviation project has moved from low-income farmers and lower-middle-income farmers to middle-income farmers and even high-income farmers [35,36,37]. Such a situation has been caused by the sufficient demand of poverty-stricken farmers on micro-credit for poverty alleviation, the fact that credit institutions are facing sustainability pressure, and that the personnel of these institutions are more inclined to lend to high-income farmers engaged in non-agricultural business projects. Oliphant and Ma [38] believed that cognitive bias between borrowers and lenders in the rural micro-credit market in China results in the low efficiency of the micro-credit market. Target deviation is caused by a few reasons. With the development of inclusive financing in China, the social target of the institutions of micro-credit for poverty alleviation has deviated from poverty alleviation and low-income groups due to the commercialization of micro-credit for poverty alleviation, the expansion of the scale of institutions, and the self-exclusion of the poverty-stricken population, meaning micro-credit has deviated from its original intention; therefore, it is essential to optimize the governance structure of institutions, reduce trade costs, and strengthen social performance management. The bank of agriculture and commerce is the most important subject in the implementation of micro-credit for poverty alleviation. Micro-credit for poverty alleviation has deviated from its original intention because the bank does not completely understand the concept of micro-credit for poverty alleviation and is thus unwilling to actively grant micro-credit, and commercial financial institutions also find it hard to persist in poverty alleviation; therefore, the establishment of a policy-based micro-credit institution for poverty alleviation by the state is the best way to alleviate poverty with the use of micro-credit.

The actual effect on poverty alleviation and income growth is the implementation purpose and policy foothold of micro-credit for poverty alleviation, which should be considered in terms of long-term and short-term effects. The short-term effect can be measured by the influence of micro-credit for poverty alleviation on the current income level of farmers, and the long-term effect can be measured by confirming whether micro-credit for poverty alleviation promotes a stable increase in farmers’ income. The long-term effect is based on the short-term effect, which can better reflect the quality and essential requirements of the effect of micro-credit on poverty alleviation and income growth. The existing research has placed more emphasis on the short-term effect of micro-credit on poverty alleviation and income growth, instead of reviewing the impact on the stable improvement of farmers’ income [39]. The stable increase in farmers’ income is not only an important guarantee for high-quality poverty alleviation, but it is also an essential condition for promoting rural revitalization [40,41]. With the successful completion of poverty alleviation, the government should take proper measures to prevent people from returning to poverty after being lifted out of it, which is an important task [42,43,44]. Only when the characteristic poverty alleviation industries ensure long-term and healthy development and farmers’ income grows continuously and stably can the risk of returning to poverty be reduced effectively so as to ensure the quality of the poverty alleviation method. The relationship between micro-credit for poverty alleviation and the stable growth of farmers’ income is an important issue to be focused on.

3. Theoretical Analysis

Conroy [45] stated that micro-credit is the provision of a broad range of financial services such as deposits, loans, payment services, money transfers, insurance, and training to poor and low-income households and their micro enterprises. Micro-credit for poverty alleviation studied in this paper refers to the discount government micro-loan with the purpose of developing characteristic poverty alleviation industries of registered farmers, which does not include various types of non-discount government micro-loan or interest-free micro-loan for poverty alleviation in some areas and differs from commercial micro-credit for poverty alleviation and cooperatives’ micro-credit for poverty alleviation and farmers’ income. Farmers’ income includes production and operation income, transfer income, wage income, and asset income. Micro-credit for poverty alleviation has been widely recognized as a crucial tool for poverty alleviation and socioeconomic well-being. It can contribute to farmers’ income growth through a variety of pathways.

First, for those who benefit from micro-credit, the loan itself temporarily provides additional capital that can be used to increase the productive physical capital of the household [8]. For agricultural households, in particular, the demand for financial services arises from the funding needs of cyclical agricultural operations. Borrowing may also enable the household to take advantage of potentially profitable investment opportunities that are beyond the farmers’ own endowments and resources. More importantly, easing capital constraints through credit can reduce the opportunity costs of capital-intensive assets relative to family labor, thereby promoting the application of capital-intensive technologies and increasing labor productivity. This will further contribute to the long-term stable growth of farmers’ income.

Second, micro-credit can also contribute to income generation by increasing the risk-bearing capacity of a household. For example, if there is a riskier but higher-yielding contract in agricultural markets, the household is more willing to engage in this investment as long as they know that credit is available to cushion consumption and cope with income shortfalls [3]. This contributes to the potential income growth of the household. Furthermore, Phan et al. [46] found that small loans from micro-credit programs may help rural households cope with agricultural production shocks. Han et al. [4] argued that those farmers who receive micro-credit for poverty alleviation can use their bank savings to cope with the bad situation caused by risks shocks, such as illness, death, and bad weather, which reduces their vulnerability to poverty. These findings reinforce the important role of micro-credit in farmers’ livelihoods.

Third, enhancing social capital in the rural area is another intermediary pathway for micro-credit for poverty alleviation to promote farmers’ income. Social capital can be seen as a consequence of micro-credit programs, and micro-credit can play a crucial role in the formation of, change in, and conversion of all kinds of social capital [12,47] because saving and credits in micro-credit programs are partly practiced in small, interrelated groups and contribute to the formation of a social network [48]. Agboola et al. [2] concluded that social capital affects productivity positively, and it is an important factor in improving the income of community members. The above analysis shows that micro-credit for poverty alleviation can contribute to the overall income growth of community groups by promoting the formation and development of social capital.

Therefore, based on the above discussion, the current study formulated the following hypotheses:

H1. 

Micro-credit for poverty alleviation can increase farmers’ income.

H2. 

Micro-credit for poverty alleviation can contribute to the stable growth of farmers’ income.

4. Materials and Methods

4.1. Data Sources and Sample Characteristics

All data used in this paper were obtained from the field investigation in Fuping County and Quyang County in Hebei, China, in June 2021. The investigation objects are the registered households, including poverty-stricken households and households lifted out of poverty. This investigation was conducted by means of door-to-door questionnaires, interviews with the farmers by village cadres, and formal discussion with the farmers by relevant authorities. In total, 1–2 poverty-stricken village(s) was/were sampled from each town, 13 villages (including Pingyang Town in Fuping County) and 18 towns (including Xiaomu Town in Quyang County) were fully covered, and 42 poverty-stricken villages were sampled. Furthermore, the registered households were selected randomly as the respondents, and two investigators worked in pairs to conduct the door-to-door questionnaires based on the system of the Poverty Relief Office of the State Council. In case of any doubt about any issue and data in the investigation process, the investigators should confirm information with the registered households and village cadres. The participants in formal discussion were from the Poverty Relief Office of the County, local financial supervision bureau (formerly the financial office), and other departments, and the interviewees were the village cadres and the first secretary of the village resident team. Information such as the size and purpose of micro-credit for poverty alleviation, operation procedures, and risk control were obtained in the investigation.

In total, 860 questionnaires were collected in this investigation, and because this paper researches the effect of micro-credit for poverty alleviation on the industry income of registered poverty-stricken farmers, another 483 questionnaires were chosen from the farmers involved in industries. This refers to the characteristic poverty alleviation industries developed by the poverty-stricken households with the help of micro-credit for poverty alleviation, as well as the farmers’ production and operation income. We reviewed these questionnaires and rejected some of them where information was missing or doubtful and could not be supplemented or confirmed by a telephone follow-up interview. Thus, 458 valid samples were obtained and used as the sample data for the empirical analysis. The basic characteristics of the sample are shown in

Table 1. Basic characteristics of the samples.

Variables	Options	Quantity	Proportion (%)
Fuping		215	46.9%
Quyang		243	53.1%
Households in poverty		166	36.2%
Households out of poverty		292	63.8%
Type of industry	Planting	192	41.9%
	Aquaculture	201	43.9%
	Other	65	14.2%
Net production and operation income	≤20,000	48	10.5%
	20,000–50,000 (inclusive)	167	36.5%
	50,000–80,000 (inclusive)	145	31.7%
	80,000	98	21.4%
Industry scale	≤30,000	50	10.9%
	30,000–80,000 (inclusive)	235	51.3%
	80,000–120,000 (inclusive)	125	27.3%
	120,000	48	10.5%
Micro-credit for poverty alleviation	Yes	366	79.9%
	No	92	20.1%
Farmer specialized cooperatives	Added	385	84.1%
	Not added	73	15.9%
Proportion of family labor force	≤0.25	110	24.0%
	0.33–0.50	203	44.3%
	≥0.60	145	31.7%
Schooling years of the head of the household	<6 years	38	8.3%
	6–9 years (inclusive)	278	60.7%
	>9 years	142	31.0%
Other financing channels	Yes	162	35.4%
	No	296	64.6%

Note: Schooling years include the years of formal academic education and the accumulated training period of skills and technologies received by the head of the household, because technology and skill training can directly affect the ability of farmers to engage in production and operation. Other financing channels include formal ones such as commercial banks, mutually funding cooperatives, and micro-credit companies, and informal channels such as private lending.

.

In terms of regional distribution, Fuping County and Quyang County are adjacent to each other in the central region of Hebei Province with mountains and hills as the main landform and a small number of plains, and thus their situation is representative. The sample of the two counties accounts for 46.9% and 53.1%, respectively. The majority of households are engaged in the planting of traditional crops, fungus, mushrooms, and Chinese herbal medicines and the breeding of livestock and poultry such as cattle, pigs, sheep, and chickens. With respect to the agricultural production and operation income, 68.1% of their income ranges from CNY 20,000 to 80,000, 21.4% of their income exceeds CNY 80,000, and 3.0% exceeds CNY 100,000. The overall income level is not high, but the per capita income exceeds the poverty standard.

Among the 458 registered households, 36.2% are in poverty, and 63.8% are out of poverty; 79.9% have applied for micro-credit for poverty alleviation, and 20.1% have not; and 84.1% have joined the farmer specialized cooperatives, and 15.9% have not. The farmer specialized cooperatives are highly recognized by most farmer households, and the farmers have a widespread demand on credit funds. In terms of the scale of production and operation, 51.3% investment expenditure lies between CNY 30,000 and 80,000, 27.3% is between CNY 80,000 and 120,000, 10.9% is less than CNY 30,000, 10.5% is higher than CNY 120,000, and 0.9% (just 4 households) exceed CNY 200,000. As a result, the industry investment scale of most households is small.

The household labor force in the investigated samples accounts for 46.0% on average. The 145 households with a labor force participation ratio higher than 46.0% are all households out of poverty, so labor force plays an important role in the process of poverty alleviation. The average schooling years of the farmers are 7.98, of which 60.7% lie between 6 and 9 years, 8.3% are below 6 years, 31.0% are more than 9 years, and only 11 persons have average schooling years of more than 12 years, accounting for 2.4%. The farmers generally receive short-term agriculture production technology training, but their overall education level is low. In the 458 samples, 111 farmer households have serious and chronic diseases, accounting for 24.2%. There are 162 farmer households with other financing channels, accounting for 35.4%.

4.2. Model Construction and Variable Setting

The empirical research in this paper is divided into two steps: the first is whether micro-credit for poverty alleviation can promote the income growth of farmers, and the second is whether micro-credit for poverty alleviation can improve the growth stability of farmers’ income. There is a successive relationship between these two questions. Only after providing a positive answer to the first question can the second question be researched.

4.2.1. Short-Term Effect of Micro-Credit on Poverty Alleviation

This effect is illustrated by testing the impact of micro-credit for poverty alleviation on the income level of the registered farmers. In view of targeted poverty alleviation, micro-credit for poverty alleviation can increase farmers’ income through supporting the development of the characteristic industries of the farmers so as to play its role in poverty alleviation. Therefore, the net production and operation income of the farmers is selected as the explanatory variable, and the empirical model is set as follows:

$$\begin{array}{l}income={\beta }_0+{\beta }_{1}microloan+{\beta }_{2}coprtv+{\beta }_{3}microloan\times croprtv+{\beta }_{4}scal\\ +{\beta }_{5}laborratio+{\beta }_{6}edu+{\beta }_{7}health+{\beta }_{8}othchan+\epsilon \end{array}$$

(1)

Income represents the logarithmic form of net production and the operation income of the farmers in 2020 as an explained variable. Microloan represents a dummy variable of participation of the farmers in micro-credit for poverty alleviation; when the farmers receive micro-credit for poverty alleviation, the value of the microloan is 1, and when the farmers do not receive it, the value of the microloan is 0. Coprtov is a dummy variable of the participation of the farmers in the farmer specialized cooperatives. When the farmers participate in the cooperatives, coprtov = 1, and when the farmers do not participate, coprtov = 0. In order to examine the income effect of micro-credit for poverty alleviation due to participation in the farmer specialized cooperatives, an interactive term of the farmer specialized cooperatives (coprtov) and micro-credit for poverty alleviation (microloan) is added in model (1). Scal is the industry scale, expressed by the logarithm of industrial investment; laborratio is the proportion of family labor force; edu is the schooling years of the head of the household; health refers to the physical condition of the members of the household; and othchan is a dummy variable of whether the farmers have other financing channels. If they do have one, othchan = 1; in the case that they do not have one, othchan = 0. Microloan, coprtove, and the interactive term microloan × coprtov are the most important variables in this paper.

4.2.2. Long-Term Effect of Micro-Credit on Poverty Alleviation

This effect is addressed by testing the influence of micro-credit for poverty alleviation on the stability of farmers’ income. In a target poverty alleviation task, the growth stability of the income of the registered farmers is a variable that requires much more special attention. After the farmers are lifted out of poverty, they all are at risk of returning to poverty, and the key point is to avoid and decrease such a risk until a long-term mechanism has been set up for the stable growth of farmers’ income, with industry being at the core of this. For the previously explained variable, the stable growth of farmers’ income has 2 values, which are binary response variables, that is, the growth is table if case y = 1, and the growth is not table in the case of y = 0. A logit binary choice model was used for regression estimation.

Assuming that $x_{k,}k=1,2,\cdot \cdot \cdot ,$ is the independent variable, representing the influence factors of the growth stability of farmers income, $y_i={\beta }_0+{\beta }_1{x}_{1i}+{\beta }_2{x}_{2i}+\cdot \cdot \cdot +{\beta }_k{x}_{ki}+u_i,i=1,2,\cdot \cdot \cdot ,n,$, and n is the number of samples. $k$ is the quantity of independent variables. $x_{ki}$ is the value of the individual characteristics of the i^th sample. $u_i$ is the random perturbation term which is independent of each other and has a mean value of 0.

The model above cannot be used as a model for practical research on binary choice due to the heteroscedasticity of the error terms, and thus it should be transformed into a utility model. According to Goldberger (1964) and Maddla (1983), a model for practical application can be obtained:

$y_i=1-F(-x_i\beta )+u_i,$ and $y_i$ is a regression of its conditional mean.

With the support of sample data, the parameter estimator can be calculated:

$$p=p(y=1/x)=F(x^{\prime }\beta )=\frac{\exp (x^{\prime }\beta )}{1+\exp (x^{\prime }\beta )}$$

$p(y=1){\beta }_k$ means that change in the “logarithmic probability ratio” will be caused by each additional x. If $x$ changes by one unit, the marginal effect on $p(y=1)$ is approximately expressed as $p(1-p){\beta }_k$.

The econometric model is constructed as follows:

$$\begin{array}{l}p(inration=1)={\beta }_0+{\beta }_{1}microloan+{\beta }_{2}coprtv+{\beta }_{3}microloan\times croprtv+{\beta }_{4}scal\\ +{\beta }_{5}laborratio+{\beta }_{6}edu+{\beta }_{7}health+{\beta }_{8}othchan+\mu \end{array}$$

(2)

Inratio is the explained variable, using the ratio of net production and operation income of the farmers in 2018 to that in 2017. If the ratio is larger than 1, it indicates that the net income in 2020 is more than that in 2019, and the income grows stably; if the ratio is less than or equal to 1, it indicates that the net income in 2020 is lower than that in 2019, and the income grows unstably, with a large risk of returning to poverty. Microloan is the core explained variable of the model.

5. Results

Table 2. Variable setting and descriptive statistics.

	Variable	Connotation or Assignment	Expected Direction of Action		Mean	Standard Deviation
Explained variable	Production and operation income	Net production and operation income in 2020, in logarithm	Net production and operation income		5.654	2.714
	Stability of income growth	Production and operation income in 2020/production and operation income in 2019. If the ratio is >1, y = 1; if the ratio is ≤1, y = 0.		Income stability	0.865	0.342
Core explanatory variable	Is there any micro-credit for poverty alleviation?	Yes = 1, no = 0	+	+	0.799	0.401
Control variable	Do the farmers participate in the farmer specialized cooperatives?	Yes = 1, no = 0	+	+	0.841	0.366
	Interactive item of “Is there any micro-credit for poverty alleviation?” and “Do the farmers participate in the farmer specialized cooperatives?”	“Is there any micro-credit for poverty alleviation?” * “Do the farmers participate in the specialized farmer cooperatives?”	+	+	0.679	0.467
	Industry scale	Assign value by capital investment 5-1	+	+	7.887	3.761
	Proportion of household labor force	Number of household labor force/number of household population	+	+	0.460	0.198
	Level of education	Years of schooling of the head of household, in logarithm	+	+	7.983	1.798
	Physical conditions of family members	With chronic disease or serious disease = 1, without = 0	_	_	0.245	0.430
	Other financing channels	Yes = 1, no = 0	_	_	0.356	0.479

Note: * denotes the interaction effect between variables. + means that the explanatory variable has a positive effect on the explained variable and vice versa.

shows the statistical description of each variable. Furthermore, the Eviews10.0 econometric analysis software is used in this paper to fit the test and estimation of the sample data and to analyze the effect of micro-credit for poverty alleviation on income growth and poverty alleviation.

5.1. Short-Term Effect of Micro-Credit on Poverty Alleviation

The necessity of adding an interactive term is determined by comparing the goodness of fit of the equations with the interactive term and those without the interactive term. According to

Table 3. Model (1) regression results.

Variable	Interactive Term Not Added		Interactive Term Added
	Coefficient (Standard Error)	t Value	Coefficient (Standard Error)	t Value
c	−1.398 *** (0.291)	−4.801	−1.298 *** (0.348)	−3.731
Is there any micro-credit for poverty alleviation?	0.235 *** (0.133)	3.754	0.568 ** (0.311)	2.791
Do the farmers participate in the farmer specialized cooperatives?	0.391 ** (0.147)	2.665	0.178 *** (0.303)	5.894
“Is there any micro-credit for poverty alleviation?” and “Do the farmers participate in the farmer specialized cooperatives?”	-	-	0.181 ***(0.343)	5.264
Production and operation scale	0.406 *** (0.018)	21.767	0.405 *** (0.019)	21.713
Proportion of household labor force	1.640 *** (0.302)	5.427	1.646 *** (0.303)	5.440
Education level of the head of household	0.316 *** (0.035)	8.948	0.318 *** (0.035)	8.954
Physical condition of family members	−0.207 ** (0.124)	−2.173	−0.209 ** (0.124)	−2.485
Is there any other financing channel?	1.319 (0.136)	0.279	1.323 (0.137)	0.285
$R^2$	0.829		0.829
Log likelihood	−702.309		−702.168
χ2	14.067		15.507
F value	311.573		272.223
Sample quantity	458

Note: **, and *** represent the significant value at the statistic level of 5%, and 1%, respectively, as follows.

, when comparing the chi-square difference in and the distribution critical value of the two models, the p value is less than 0.05, indicating that there is a significant interactive effect of the two variables.

According to Table 3, the core variable (micro-credit for poverty alleviation) has a significant positive influence on the income level of the farmers, and the short-term effect on poverty alleviation is obvious, which is in line with expectations, and the income has grown by an average of CNY 5680. Policy support should focus on the cultivation and development of characteristic poverty-stricken industries. After exploration and development for several years, many farmers’ industries have progressed to the growth stage. Although the overall scale of the industry is still small, farmers generally have a strong desire to expand the scale of production and operation, so there is a strong emphasis on funds. Micro-credit for poverty alleviation meets the needs of farmers in terms of funds to develop their industries, especially since the end of 2017, when the Poverty Relief Office of the State Council prescribed a strict limit to the intended use of micro-credit for poverty alleviation after eliminating such situations as “individual loan for enterprise use” or “individual loan for social use” in order to achieve accurate docking with the farmers’ industries and obviously promote the growth of farmers’ income.

Participation in farmer specialized cooperatives can also effectively increase farmers’ income, with an average growth of CNY 1780. The interactive term of micro-credit for poverty alleviation and the farmer specialized cooperatives is significant at the statistical level of 1%, the regression coefficient is 0.181, and the interactive effect is significant. Compared with the farmers with micro-credit for poverty alleviation who do not participate in farmer specialized cooperatives, the farmers who participate in the cooperatives have grown by an average of CNY 1810, indicating that participation in the cooperatives can effectively improve the effect of micro-credit on income growth and poverty alleviation. Although the farmer specialized cooperatives are not real cooperatives at their current stage, the cooperatives play an important role in the actual production and operation of farmers. The farmers provide services to the member farmers of the cooperatives from industry selection, investment, production and operation, financing, process management, technology guidance, and market sale. These functions have been recognized by the farmers. Additionally, compared with non-member farmers, membership can indicate that the farmers have a better credit level and ability to develop their industries, can obtain micro-credit more easily, can apply for micro-credit in the development of their industries to form a good circulation and interaction relationship between credit capitals and industry development, and can promote the growth of farmers’ income.

The control variable has a positive influence on the farmers’ income level at a significance level of 1%, in which the family income of farmers increases by an average of CNY 4050 when an additional CNY 10,000 is added to the scale of production and operation, and family income increases by an average of CNY 160 when the proportion of the family labor force increases by 1%, and family income increases by an average of CNY 3180 with each additional year increase in the years of schooling of the head of the household. The physical condition of family members has a negative influence on farmers’ income at a significance level of 1%, which is consistent with the previous expectations. If family members suffer from serious diseases or chronic diseases, their income will be affected negatively and significantly. Additionally, the income of farmer households with members suffering from serious or chronic diseases is CNY 2090 lower than healthy households on average.

“Whether there is any other financing channel does not significantly affect the farmers’ income” fails to pass the t test, which is inconsistent with expectations. In combination with previous statistical analysis, failure to pass the test is possibly because the scale of the characteristic poverty alleviation industries of the registered poverty-stricken farmers is small at their current stage, the demand gap of credit funds of a single farmer is small, and micro-credit for poverty alleviation can only just meet the credit fund demands of the farmers at present.

5.2. Long-Term Effect of Micro-Credit for Poverty Alleviation on Poverty Alleviation

Based on the analysis of the regression results of model (1), micro-credit for poverty alleviation can significantly promote the growth of farmers’ income, and the regression equation of model (2) is constructed on this basis:

$$\begin{array}{l}{x}^{\prime }\beta =\underset{\left(2.825\right)}{1.793}+\underset{\left(2.888\right)}{0.947\times microloan}+\underset{\left(2.557\right)}{0.473\times coprtov}+\underset{\left(1.989\right)}{1.118\times mc}+\underset{\left(3.743\right)}{0.257\times scal}\\ +\underset{\left(2.704\right)}{2.434\times laborration}+\underset{\left(2.539\right)}{0.268\times edu}-\underset{\left(1.850\right)}{0.055\times health}\end{array}$$

(3)

Refer to

Table 4. Model (2) regression results.

Variable	Coefficient	Standard Error	Value z
c	1.793	0.982	2.8253
Is there any micro-credit for poverty alleviation?	0.947 ***	1.0666	2.8880
Do the farmers participate in the farmer specialized cooperatives?	0.473 **	0.8498	2.5569
“Is there any micro-credit for poverty alleviation?” and “Do the farmers participate in the specialized farmer cooperatives?”	1.118 **	1.1298	1.9894
Production and operation scale	0.257 ***	0.0687	3.7433
Proportion of household labor force	2.434 ***	0.8999	2.7041
Education level of the head of household	0.268 **	0.1055	2.5391
Physical condition of family members	−0.055 *	0.3432	1.850
Is there any other financing channel?	-	-	-
LR value	45.6436
Log likelihood	−158.7619
Sample quantity	458

Note: *, **, and *** represent the significant value at the statistic level of 10%, 5%, and 1%, respectively.

for the regression results:

The core variable concerned in this paper, micro-credit for poverty alleviation, has a significant positive influence on the stable growth of farmers’ income, which is in line with the expectation. The participation of farmers in the farmer specialized cooperatives can effectively promote the stable growth of farmers’ income. According to regression data, the interactive term of micro-credit for poverty alleviation and the farmer specialized cooperatives can also significantly promote the stable growth of farmers’ income, with a significant interactive effect. Participation in the farmer specialized cooperatives can significantly promote micro-credit for poverty alleviation to boost the stable growth of farmers’ income.

In terms of the marginal effect of the variable, the possibility of stability is increased by an average of 3.79% when the core variable, micro-credit for poverty alleviation, increases by 1%; the possibility of stability is increased by an average of 1.89% when the control variable, the farmer specialized cooperatives, increases by 1%; and the possibility of stability is increased by an average of 4.47% when the interactive term of the two variables increases by 1%. According to the comparison with the regression results of model (1), micro-credit for poverty alleviation has a significant positive effect on farmers’ income and income growth stability, but the effect of the variable of the interactive term on income growth stability in model (2) is far larger than that on the income level in model (1). Compared with the short-term effect, the interactive effect of two variables can be seen fully in the long term, so that the stable growth of farmers’ income can be effectively promoted. Compared with non-member farmers, member farmers have a stronger funds demand on industry development, so they are more eager to obtain micro-credit for poverty alleviation. Because information asymmetry between the member farmers and the bank is at a lower level, it is easy for the bank to supervise the use of micro-credit for poverty alleviation, and the bank also tends to provide micro-credit to the member farmers. According to the investigation results of these two counties, micro-credit is essentially used in industry development by member farmers, and the income of 92.6% of the farmers continuously increased in the most recent two years.

Control variables such as the scale of production and operation, the physical condition of family members, the proportion of farmer family labor force, and the schooling years of the head of the farmer household exert a positive influence on the stable growth of farmers’ income at significance levels of 1%, 1%, and 5%. The effect of the variable of the proportion of the farmer family labor force is especially significant. The possibility of stable income growth increased by an average of 9.74% when the variable increases by 1%. In combination with model 2, the labor resources of the farmer’s family is an important factor which can influence farmers’ income and their growth stability regardless of the term (long or short). The characteristic poverty alleviation industries at their current stage are mainly centered around traditional small-scale planting and breeding industries and a small number of service industries related to agricultural production and operation, for example, mushroom cultivation, herb cultivation, and beef cattle breeding, which takes the family as the main body. These industries do not require too many funds and do not have high requirements for technology level and labor quality; the industry chain is short, the number of those involved in labor plays an important role in industry development, and the development scale is always determined by the number of family members involved in the labor.

The stable growth of farmers’ income will be affected negatively at a significance level of 10% if family members have serious or chronic diseases, but the effect is slight. In view of marginal effects, the possibility of stable income growth decreases by only an average of 0.22% when the variable of the physical condition of family members increases by 1%.

With the promotion of health poverty alleviation, medical poverty alleviation systems have been basically established, including basic medical insurance for urban and rural residents. In 2017, the government integrated basic medical insurance for urban and rural residents and new rural cooperative medical insurance to establish a unified basic medical insurance system for urban and rural residents, contract services for chronic diseases, and serious diseases insurance and serious diseases assistance. The degree of the medical insurance has been effectively improved by a series of proper measures, such as a tiered diagnosis and the treatment of common diseases, “one-stop” payment for hospitalization, and the establishment of a chronic disease registration system. On this basis, a special assistance fund is generally set up by the local government for the farmer households still in poverty after benefitting from the above policies, or insurance companies are introduced to set up a commercial insurance system to ensure basic needs are met for them. Thus, the cost of the medical assistance of those with chronic and serious diseases in the registered poverty-stricken farmers’ households is not the burden of the family anymore, meaning the limitation of the development of characteristic poverty alleviation industries is greatly weakened, and the variable of the physical condition of family members does not significantly influence the industry income of the farmers anymore.

5.3. Robustness Test

In order to check whether the regression results are robust, we performed a robustness test by replacing the core variable and changing the samples.

5.3.1. Replace the Core Explained Variable and Regress Again

In this paper, the core explained variable in models (1) and (2), “Is there any micro-credit for poverty alleviation?”, is replaced by “Is there any production and operation loan?”. The results are shown in

Table 5. Robustness regression results of the effect of micro-credit for poverty alleviation on increase growth and poverty alleviation (replacing the core explained variable).

Variable	Model 1 Production and Operation Income	Model 2 Stability of Income Growth
c	−0.112 * (0.021)	1.322 ** (0.234)
Is there any production and operation loan?	0.323 *** (0.012)	0.333 ** (0.121)
Do the farmers participate in the specialized farmer cooperatives?	0.221 * (0.023)	0.263 *** (0.011)
“Is there any micro-credit for poverty alleviation?” and “Do the farmers participate in the specialized farmer cooperatives?”	0.157 ** (0.029)	0.201 *** (0.210)
Production and operation scale	0.214 ** (0.190)	0.312 * (0.091)
Proportion of household labor force	0.345 ** (0.056)	0.131 * (0.098)
Education level of the head of household	−0.012 (0.135)	−0.028 * (0.310)
Physical condition of family members	−0.098 ** (0.143)	−0.103 ** (0.078)
$R^2$	0.687	0.693
Sample quantity	458

Note: *, **, and *** represent the significant value at the statistic level of 10%, 5%, and 1%, respectively.

. Additionally, the regression results suggest that a non-consumption loan can effectively raise the farmers’ income level and promote stable income growth, and the estimation results of other control variable (excluding the variable of “Is there any other financing channel?”) is in line with the content described earlier. Obviously, the research conclusion described earlier is still valid after changing the explained variable to “Is there any production and operation loan?”.

5.3.2. Change the Samples and Regress Again

The data are derived from an investigation in Wangqing County and Longjing City, Jilin Province, China, in December 2020. Both counties are border regions which are inhabited by various ethnic groups, governed by Yanbian Korean Autonomous Prefecture, and are made up of mainly mountainous terrain. Their characteristic industries mainly concentrate on the cultivation and processing of auricularia and cattle breeding. The situation of these two counties is representative and obviously different from that of the two counties in Hebei, so a comparative study can be made between them. The investigation was performed in the form of a door-to-door questionnaire, interview, and formal discussion. In total, 33 administrative villages were randomly selected from the two counties, and 787 valid questionnaires were collected, from which 482 samples were selected as the changing samples, and then a regression analysis was performed. Micro-credit for poverty alleviation has a significantly positive influence on the farmers’ income level and income growth stability, with the regression coefficient being 0.211 and 0.328 (the regression results are not presented in the paper due to space limitations), respectively. The farmer specialized cooperatives also exercised a positive influence, with regression coefficients of 0.296 and 0.233, respectively. Additionally, the positive interactive effect of the two variables is significant, with regression coefficients of 0.267 and 0.187, respectively. After changing the samples, the core variable, micro-credit for poverty alleviation, still plays a significant role in income growth and poverty alleviation. The positive influence of the farmer specialized cooperatives has a significant interactive effect with the core variable, and the estimation results of other control variables are in line with the results described earlier. As a result, the research conclusion discussed earlier is robust.

6. Discussion

From the previous empirical analysis, it is clear that micro-credit for poverty alleviation can effectively contribute to the growth of farmer household income and the stability of farmer household income growth. This result is generally consistent with the findings of Khandker and Koolwal [3] and Félix and Belo [5]. Meanwhile, the flexible mode of operation and increasingly sophisticated risk management and early warning mechanisms have effectively reduced the operational risks of micro-credit for poverty alleviation, providing an effective guarantee for its safe operation. In addition, the role of specialized farmer cooperatives cannot be ignored. It can facilitate the role of micro-credit for poverty alleviation. In fact, specialized farmer cooperatives are very common in rural areas of China, and almost all aspects of the operation of micro-credit for poverty alleviation are closely related to them. Specifically, farmers with financial needs are basically members of specialized farmer cooperatives, and specialized farmer cooperatives will influence micro-credit for poverty alleviation from loan eligibility, fund disbursement, purpose supervision, rollover repayment, etc. Sandra et al. [49] also pointed out the importance of the effect of farmer specialized cooperatives on poverty alleviation with the use of micro-credit. Therefore, it is necessary to strengthen the financial service function of specialized farmer cooperatives and other organizations in developing countries. Under a series of policy guarantees, funds are more precisely matched to those in need of funds, which makes farmers who originally relied on informal channels to obtain funds turn to this formal platform. From a macro perspective, rural financial markets will become more efficient.

More importantly, at the micro level, farmers’ profitability and household economic wellbeing will be effectively increased. With the help of micro-credit for poverty alleviation, farmers’ incomes will continue to rise, and then micro-credit will be increasingly recognized by farmers in rural areas and widely implemented. In the long run, this is beneficial to the improvement of farmers’ profitability and household economic wellbeing. First, as small-scale and subsistence farming is usually less profitable in rural areas, rural households tend to look for non-farm self-employment opportunities. When provided with the new sources of funding, micro-credit borrowers may use the loan for off-farm income diversification [26]. For example, they can start up their own small businesses, such as opening trading shops or groceries, providing repairing services (motorbike, electric appliances), selling small items, etc. With the help of credit, these rural households may have better opportunities to obtain more decent work through non-farm activities [50]. Second, facilitating access to credit is likely to result in the acquisition of new skills and the upgrading of existing skills, thereby increasing the capacity of rural households to generate income and improve their livelihoods [51]. Baye [9] also pointed out that the potential for enhancing household economic wellbeing through micro-credit provision and access is immense.

7. Conclusions and Policy Implications

This paper conducts an empirical analysis of the influence of micro-credit for poverty alleviation on farmers’ income level and stable income growth based on the field investigation data of 458 registered poverty-stricken farmer households in Hebei, China. In terms of theoretical significance, this paper enriches the research findings in terms of the poverty reduction effects of micro-credit for poverty alleviation in developing countries and provides empirical support for the “Chinese version”. In terms of practical significance, this paper provides a reference for the implementation of micro-credit for poverty alleviation policies in other developing countries. Specifically, the conclusions of this paper are as follows:

• The short-term effect on income growth and poverty alleviation is significant. With control variables such as farmer specialized cooperatives, the scale of production and operation, the proportion of the family labor force, the education level of the head of the household, the physical condition of family members, and other financing channels, micro-credit for poverty alleviation has a significantly positive influence on farmers’ income level. The interactive effect of micro-credit and the farmer specialized cooperatives is significant, and participation in the cooperatives can have an improvement effect on poverty alleviation.
• There is a significant long-term effect on poverty alleviation. Micro-credit for poverty alleviation can significantly improve the stable growth of farmers’ income. In control variables, farmer specialized cooperatives, the scale of production and operation, the proportion of family labor force, and the education level of the head of the household exert a significantly positive influence on stability, whereas the physical condition of family members has a negative influence on stability, while “other financing channels” do not significantly affect stability. The interactive effect of micro-credit for poverty alleviation and the farmer specialized cooperatives can significantly promote the stable growth of farmers’ income.
• In control variables, farmer specialized cooperatives play a prominent and positive role in promoting the growth and stability of farmers’ income. Additionally, participation in the cooperatives is helpful to enhance the durability of the effect of micro-credit on income growth and poverty alleviation and effectively reduce the risk of returning to poverty.

Based on the research conclusions above, the following suggestions are proposed for further developing the effect of micro-credit for poverty alleviation on income growth and poverty alleviation in developing countries.

The first step is to further improve the accuracy of micro-credit for poverty alleviation in developing countries. The purpose of micro-credit for poverty alleviation essentially meets the requirements for developing characteristic poverty alleviation industries, and the capital demand of most farmers is also increasing, so the limit of micro-credit restricts the further development of farmers’ industries. With the promotion of the rural revitalization strategy, it is necessary to launch “loan limit + x” micro-credit products for poverty alleviation in good time, which means CNY “X0000” will be added on the basis of the existing micro-credit limit. On the premise of ensuring that farmers do not return to poverty, the loan interest is shared by the financial authority, and farmers to gradually weaken the quasi-public nature of micro-credit for poverty alleviation and transit to market-oriented operation.

The second step is to prevent the scale of the industry of some farmers in some developing countries from expanding too fast. The healthy development of farmers’ industries is the basis and ultimate purpose for the effect of micro-credit for poverty alleviation on income growth and poverty alleviation. On the whole, the scale of farmers’ industries in developing countries is not large, but unbalanced development is a widespread problem. Some farmers’ industries are large and expand too fast, meaning corresponding management difficulties and market risks also increase, so their micro-credit faces a high risk. It is necessary to plan and guide the development of farmers’ industries to ensure that farmers’ production and operation is on a moderate scale, reduce market risks, and guarantee the safety of micro-credit.

The third step is to strengthen the financial service function of the farmer specialized cooperatives and other organizations in developing countries. In view of the importance of the farmer specialized cooperatives in poverty alleviation due to micro-credit, it is necessary to further improve the organizational structure and the operating mechanism of the cooperatives, especially to strengthen their financial service functions. The cooperatives should be involved in all important links in micro-credit for poverty alleviation, such as the examination of application conditions, loan use supervision, risk management and control, and the evaluation of the poverty alleviation performance.

This study provides an effective demonstration of the poverty-reducing effects of micro-credit. Nevertheless, there are still some limitations of this paper that require further research in the future. First, from the perspective of micro-credit for poverty reduction, profitability improvement is more important than income increase, but it is difficult to quantify. Second, the data in the sample are limited to the central region of China. Therefore, future studies should use China’s more impoverished Western samples and conduct more in-depth research to quantify profitability improvements.