Does Rural Labor Transfer Impact Chinese Agricultural Carbon Emission Efficiency? A Substitution Perspective of Agricultural Machinery

Abstract

In this paper, referring to Kaya’s method, the ratio of gross agricultural production (GAP) and agricultural carbon emission is defined as agricultural carbon emission efficiency (ACEE). Considering rural labor transfer (RLT) increases the agricultural machinery intensity (AMI), the two are substitutes for each other and may interact with agricultural carbon emission efficiency (ACEE). We constructed a Simultaneous Equations Model (SEM) of ACEE, RLT and AMI and analyzed the interaction mechanism of these three variables using the Three-Stage Least Squares (3SLS). The following conclusions are drawn. First, RLT and AMI significantly promote the improvement of ACEE, while the improvement of ACEE and AMI further promotes RLT. Secondly, the causal relationship and influence mechanism of ACEE, RLT and AMI are interactive and multi-directional. For example, an increase in AMI promotes ACEE, but an increase in ACEE inhibits an increase in AMI. Finally, China has significant regional heterogeneity, and different regions have different interaction mechanisms. Local governments should consider the local economic base and technological level when implementing policies. This paper extends the analytical framework of ACEE, RLT, and AMI and provides a reference for governments to make policies.

1. Introduction

In March 2021, China, the world’s largest developing country, pledged to achieve carbon peaking by 2030 and to be carbon neutral by 2060. With 17% of China’s greenhouse gases coming from agriculture, compared to 7% in the U.S. and 11% globally, China’s agriculture has the potential to reduce carbon emissions [1]. At the same time, the disposable income per capita of rural residents in China was only one-fourth of the disposable income per capita of urban residents [2]. The New Economics of Labor Migration (NELM) argues that farmers will decide their labor’s destination according to the utility maximization principle [3]. When the income gap between urban and rural areas widens, the phenomenon of farmers moving to non-agricultural areas and non-agricultural sectors occurs, which is defined by academics as rural labor force transfer [4,5]. At the time of reform and opening up in 1978, the urbanization rate of China’s resident population was 17.9%, and by 2020, the urbanization rate had grown to 63.9%, while the proportion of people employed in the primary sector had declined from 50.0% to 23.6% [2]. The trend of China’s rural labor force transferring to non-agricultural areas and non-agricultural production sectors is evident [6,7].

The dual economic structure theory proposed by Lewis divides developing countries into a lower-productivity agricultural sector and a higher-productivity industrial sector, which gradually moves closer to the industrial sector as rural labor is transferred [8]. In the early stage of rural labor force transfer, low rural productivity, scarcity and structural imbalance of labor force may occur in the phenomenon of idle farmland, which leads to a decline in agricultural production and the price of agricultural products as well as a decline in gross agricultural product, also inhibiting the increase in carbon emissions [9,10,11]. In the later stage of labor transfer, with the improvement of farmers’ income and production efficiency, the scale of operation may be realized through “farmland consolidation”, which will reduce the cost of agricultural production and increase gross agricultural product [12,13]. Scale operations supported by green agricultural production technologies can improve the efficiency of agricultural carbon emissions. For example, straw return technology increases soil fertility and avoids the increase in carbon emissions caused by straw burning, and no-till planting technology reduces the energy consumption associated with tilling farmland [14,15]. However, the traditional rough-scale model may further increase agricultural carbon emissions [16].

According to the theory of induced innovation, after the relative price of a factor increases, technological progress can enable abundant factors to substitute for scarce factors, offsetting the constraints on economic growth imposed by scarce factors [17]. The rural–urban income gap increases opportunity costs for farmers to engage in agricultural production. The potential price of agricultural labor rises, so producers choose labor substitutes to maintain output levels [18]. Labor substitutes include means of production such as pesticides, fertilizers, and machinery. Pesticides and fertilizers are indirect substitutes for labor, which increase output by consuming resources at the same rate as labor but do not contribute to reducing carbon emissions in agriculture [19]. Machinery is a direct substitute for labor, including tillage machinery, fertilizer application machinery, irrigation machinery, etc. [20]. Endogenous growth theory suggests that agricultural machinery is more advanced in productivity than traditional labor, reducing the misuse of fertilizers and pesticides by increasing resource use efficiency, thus reducing agricultural carbon emissions [21,22]. The “rebound effect” of energy economics, on the other hand, suggests that agricultural machinery improves energy efficiency but at the same time increases energy demand, resulting in energy consumption savings that fail to offset the increase in energy consumption, causing an increase in agricultural carbon emissions [23]. With the mechanization rate of agricultural cultivation in China rising from 29.1% in 2000 to 71.2% in 2020, the role of machinery in agricultural production and carbon emissions is becoming increasingly critical [2].

As shown in Figure 1, we find that the transfer of rural labor and the substitution of machinery after the transfer affect the gross agricultural product and carbon emissions. Referring to the methodology of Kaya et al., we define the ratio of gross agricultural output to agricultural carbon emissions as the efficiency of agricultural carbon emissions [24,25,26]. Referring to the method of Huang and Zheng and Gao, we defined the rural labor transfer ratio as (employment in rural areas—employment in agriculture)/employment in rural areas [27,28]. Meanwhile, the ratio of agricultural machinery power to cultivated area defines agricultural machinery intensity. Based on the above analysis, we propose Hypothesis 1:

Hypothesis H1a. 

Rural labor transfer will impact agricultural carbon emission efficiency.

Hypothesis H1b. 

Agricultural machinery intensity will impact agricultural carbon emission efficiency.

We extend the Kaya identity and derive the following equations:

$$C=\frac{C}{GAP}\times \frac{GAP}{M}\times \frac{M}{A}\times \frac{A}{P}\times P$$

(1)

$$ACEE=\frac{GAP}{C},TL=\frac{GAP}{M},AMI=\frac{M}{A},NS=\frac{A}{P}$$

In Equation (1), C represents the agricultural carbon emissions, M denotes the agricultural machinery quantity, A indicates the area of arable land, and GAP and P signify gross agricultural production and the rural labor force quantity, respectively. ACEE represents agricultural carbon emission efficiency, shown in the reciprocal form in Equation (1), and AMI indicates the machinery intensity. TL denotes the technical level of agricultural production, i.e., the GAP contribution per agricultural machinery unit, and NS represents the level of natural resources, i.e., the amount of arable land per capita. By applying a transformation to Equation (1), the following equation is obtained:

$$∆C=∆ACEE+∆AMI+∆P+∆TL+∆NS$$

(2)

In Equation (2), ∆ represents the rate of change, so ∆P reflects rural labor transfer (RLT). The level of agricultural technology (TL) and natural resources (NS) are exogenous factors in this paper. When the exogenous variables are controlled, ∆C is affected by ∆ACEE, ∆AMI and ∆P (RLT). After moving ∆C to the right side of Equation (2), any of ∆ACEE, ∆AMI and ∆P (RLT) moved to the left side of Equation (2) will be affected by the other two on the right. Based on the above analysis, we propose Hypothesis 2:

Hypothesis H2.

There are interactive influence mechanisms among agricultural carbon emission efficiency, rural labor transfer, and agricultural machinery intensity.

We construct a Simultaneous Equations Model of agricultural carbon emission efficiency, rural labor transfer, and agricultural machinery intensity to verify the two hypotheses. We use the 3SLS regression method to eliminate the endogeneity and analyze the three interaction mechanisms. Meanwhile, we divide China into four regions for heterogeneity analysis. The main contribution of this study lies in the following two aspects. First, from a theoretical research perspective, this paper constructs a three-way interactive influence framework of rural labor transfer, agricultural machinery intensity, and agricultural carbon emission efficiency instead of a single or unidirectional influence framework, which explains the influence mechanism of the three more scientifically. Secondly, from the perspective of practical application, this paper extends the study of rural labor transfer, agricultural machinery intensity, and agricultural carbon emission efficiency to the macro-provincial level, analyzes the heterogeneous differences in different regions, and helps to provide planning ideas for population migration, machinery input, and agricultural carbon reduction from a macro perspective.

2. Materials and Methods

2.1. Agricultural Carbon Emission Measurement

Volume 4 of the IPCC Guidelines for National Greenhouse Gas Inventories states that agricultural carbon emissions come from agricultural and forest land production activities, including tilling, irrigation, fertilizer application, pesticide application, use of agricultural films, agricultural machinery, etc. [29]. West and Marland systematically explored carbon emissions from small-scale agriculture and classified its carbon sources into four main categories: fertilizers, pesticides, agricultural irrigation, and seed cultivation [30]. Xu et al. measured agricultural carbon emissions from the perspective of energy consumption in agricultural production and selected six types of energy, such as gasoline and diesel, for estimation [31]. The former analyzes the carbon emissions from the agricultural production process, while the latter analyzes the carbon emissions from energy consumption in agricultural production. In this paper, we refer to the research of Zhuang to include the agricultural production process and energy consumption in the agricultural carbon emission measurement system and consider fertilizers, pesticides, agricultural films, land tilling, irrigation, and diesel fuel as sources of agricultural carbon emissions [32]. The calculation equation for the quantity of agricultural carbon emissions E is the sum of the product of each carbon source and its carbon emission factor, as follows:

$$E={\sum }_{i=1}^6{E}_i\times {\epsilon }_i$$

(3)

$E$ denotes the total agricultural carbon emissions, $E_i$ denotes the input of each carbon source, and ${\epsilon }_i$ denotes the carbon emission coefficient of each carbon source. The main carbon emission coefficients are shown in

Table 1. Agricultural carbon emission coefficients.

Source	Coefficient	Reference
Fertilizer	0.8956 kg/kg	Oak Ridge National Laboratory, USA
Pesticide	4.9341 kg/kg	Oak Ridge National Laboratory, USA
Diesel	0.5927 kg/kg	IPCC
Agricultural film	5.18 kg/kg	Institute of Agricultural Resources and Ecological Environment, Nanjing Agricultural University
Tilled farmland area	3.126 kg/hm2	School of Biology and Technology, China Agricultural University
Irrigation area	25 kg/hm2	Dubey and Lal

Note: Agricultural carbon emission coefficients refer to the work of Zhuang (2023) [32].

.

We collected the quantities of six sources of agricultural carbon emissions from the “China Rural Statistical Yearbook”, which were combined with the coefficients in Table 1 to calculate specific data on China’s agricultural carbon emissions from 2000 to 2021. We used ArcGIS 10.8.1 to generate Figure 2 of China’s agricultural carbon emissions distribution in 2021. Figure 2 shows that agricultural carbon emissions have a specific clustering effect in 2021. Four neighboring provinces, Hebei, Henan, Shandong and Jiangsu, have the highest agricultural carbon emissions, exceeding 4.05 million tons. The five neighboring provinces of Anhui, Hubei, Hunan, Guangdong and Sichuan emit more than 2.59 million tons of carbon. Guangxi, Yunnan and Hainan in the Southwest; Jilin, Heilongjiang and Liaoning in the Northeast; Fujian, Jiangxi and Zhejiang in the Southeast; and Shaanxi and Xinjiang in the Northwest emit more than 1.71 million tons of carbon. The rest of the provinces emit less than 1.71 million tons of carbon. The provinces with relatively high carbon emissions are those in China with favorable geography and a better agricultural base.

2.2. Data and Statistical Description

This paper selects panel data from 31 provinces in China from 2000 to 2021 for analysis (excluding Hong Kong, Macau and Taiwan). The endogenous variables are agricultural carbon emission efficiency, rural labor force transfer and agricultural machinery intensity. Fifteen exogenous variables were selected based on three dimensions in

Table 2. Exogenous variables.

Type	Variables	Description	Abbreviation
Policy environment	Education level of rural labor force	Average years of education	Edu
	R&D capability of province	Ratio of R&D expenditure to GDP	RD
	Tax burden of province	Ratio of taxes to GDP	Tax
	Openness level of province	Ratio of exports and imports to GDP	Open
	Technical level of province	Ratio of technology market turnover to GDP	Tech
Economic development	Urbanization level of province	Percentage of urban population	Urban
	Agricultural investment of province	Investment in fixed assets in agriculture	AI
	Industrial structure of province	Ratio of agricultural output to total output	InS
	Population density of province	Population per unit area	PD
	Unemployment rate of province	Unemployment rate	Unemp
	Income of farmers	Farmers’ disposable income	Income
	Rural–urban income gap of province	Difference between urban and rural income	Gap
Natural environment	Quantity of rainfall of province	Quantity of rainfall	Rainfall
	Land slope of province	Average slope of land	Slope
	Disaster of province	Agricultural disaster area	Dia

: policy environment, economic development, and natural environment [33,34]. All raw data are derived from the “China Statistical Yearbook”, “China Rural Statistical Yearbook”, and “National Earth System Science Data Center”; some data are obtained by calculation.

Table 3. Descriptive statistics.

	Variable	Observations	Mean	Std. Dev.	Min	Max
Endogenous variables	ACEE	682	8.361	4.917	1.851	35.301
	RLT	682	0.88	1.177	−0.217	11.604
	AMI	682	0.6	0.35	0.132	2.698
Exogenous variables	Edu	682	7.338	0.986	2.236	10.25
	Rain	682	6.729	0.505	5.303	7.711
	Urban	682	50.9	16.178	13.89	89.6
	Open	682	0.292	0.36	0.008	1.721
	AI	682	4.737	1.334	0.097	7.367
	InS	682	0.141	0.089	0.002	0.573
	Dia	682	0.221	0.161	0.000	0.936
	Tech	682	0.012	0.023	0.00	0.175
	RD	682	0.014	0.011	0.001	0.065
	Tax	682	0.076	0.027	0.034	0.2
	Unemp	682	3.475	0.716	0.76	6.5
	Gap	682	2.778	0.553	1.842	5.646
	Slope	682	1.171	1.29	0.004	5.414
	Income	682	8.776	0.771	7.193	10.559
	PD	682	5.287	1.483	0.742	8.275

shows the descriptive statistics of all the data. The mean value of agricultural carbon efficiency is 8.361, the standard deviation is 4.917, the minimum value is 1.851, and the maximum value is 35.301. It shows a big difference in the agricultural carbon efficiency of different provinces over different periods. The mean value of rural labor transfer is 0.88, the standard deviation is 1.177, the minimum value is –0.217, and the maximum value is 11.604. Rural labor generally moves to non-agricultural areas, but there is also a return of labor to the countryside. The mean value of agricultural machinery intensity is 0.6, the standard deviation is 0.35, the minimum value is 0.132, and the maximum value is 2.698, with minor fluctuations.

Figure 3 shows the trend of China’s agricultural carbon efficiency, rural labor transfer, and agricultural machinery from 2000 to 2021. ACEE’s unit is CNY 10,000 per ton, AMI’s unit is 10,000 watts per hectare, and RLT is the ratio of rural labor transfer. From 2000 to 2015, all three maintained growth; however, after 2015, agricultural carbon efficiency and machinery intensity began to diverge, and machinery intensity decreased. The main reason for this is that China implemented a new “Environmental Protection Law” in 2015, which imposed stricter requirements on local governments and enterprises to reduce pollution, and some old agricultural machines were banned. Meanwhile, after the Paris Agreement was enacted in 2016, the Chinese government began to fulfil the agreement’s content and placed carbon emissions under stricter control. For example, increasing fines for polluters, incorporating environmental governance into government performance appraisals, and encouraging clean energy use led to sustained growth in agricultural carbon efficiency [35]. Since 2019, due to the impact of COVID-19, farmers’ incomes have significantly reduced, and many farmers have moved to cities in search of jobs, leading to a rapid rise in RLT [36].

2.3. Model and Method

In this paper, we first conduct the Granger causality test on the three endogenous variables in

Table 4. Granger causality test.

Causal Flow	Coefficient	Standard Error	Z Statistic	p Value
RLT→ACEE	1.129	0.299	3.77	0.000
AMI→ACEE	2.679	0.664	4.04	0.000
ACEE→RLT	0.008	0.003	2.98	0.003
AMI→RLT	0.189	0.029	6.52	0.000
ACEE→AMI	-0.006	0.002	−2.44	0.015
RLT→AMI	0.039	0.021	1.88	0.006

[37]. We find that rural labor force transfer and agricultural machinery intensity granger cause carbon emission efficiency, consistent with Hypothesis 1. With the exception of agricultural carbon emission efficiency, which does not have a Granger causal link with agricultural machinery intensity, the three core explanatory variables all have Granger causality, which is almost consistent with Hypothesis 2.

The Granger causality test results show a causal relationship among agricultural carbon emission efficiency, rural labor force transfer, and agricultural machinery intensity. Since it is difficult to effectively present the interactions among the variables in the system in a single-equation model, this paper constructs the Simultaneous Equation Model (SEM) of agricultural carbon emission efficiency, rural labor force transfer, and agricultural machinery intensity. The advantage of SEM is that it considers the causal relationship between variables and effectively solves the endogeneity problem [38]. The system of simultaneous equations in this paper is

$${\mathrm{ACEE}}_{\mathrm{it}}={\mathsf{\alpha }}_0+{\mathsf{\alpha }}_1\times {\mathrm{RLT}}_{\mathrm{it}}+{\mathsf{\alpha }}_2\times {\mathrm{AMI}}_{\mathrm{it}}+{\sum }_{n=3}^8{\alpha }_n\times U_{nit}+{\epsilon }_{\mathrm{it}}$$

(4)

$${{\mathrm{RLT}}_{\mathrm{it}}}_{ }={{\mathsf{\beta }}_0}_{ }+{{\mathsf{\beta }}_1}_{ }\times {{\mathrm{ACEE}}_{\mathrm{it}}}_{ }+{{\mathsf{\beta }}_2}_{ }\times {{\mathrm{AMI}}_{\mathrm{it}}}_{ }+{\sum }_{k=3}^8{\beta }_k\times X_{kit}+{\phi }_{\mathrm{it}}$$

(5)

$${{\mathrm{AMI}}_{\mathrm{it}}}_{ }={{\mathsf{\gamma }}_0}_{ }+{{\mathsf{\gamma }}_1}_{ }\times {{\mathrm{ACEE}}_{\mathrm{it}}}_{ }+{{\mathsf{\gamma }}_2}_{ }\times {\mathrm{RLT}}_{\mathrm{it}}+{\sum }_{m=3}^8{\gamma }_m\times Y_{mit}+{\psi }_{\mathrm{it}}$$

(6)

In the SEM system of this paper, the three endogenous variables ACEE_it, RLT_it, and AMI_it are used as both explained variables and explanatory variables in different equations, so there are three equations in total. $U_{nit}$, $X_{kit}$ and $Y_{mit}$ represent exogenous variables in different equations, where i represents different provinces and t represents different times. α_n, β_k and γ_m represent the coefficients of different equations. ε_it, φ_it and ψ_it represent the error terms of different equations. The six most explanatory exogenous variables were selected for each of the three equations. $U_{nit}$ refers to exogenous variables for Equation (4), including education level, agricultural investment, disaster, quantity of rainfall, openness level, and industrial structure. $X_{kit}$ refers to exogenous variables for Equation (5), including urbanization level, population density, tax burden, rural–urban income gap, education level, and openness level. $Y_{mit}$ refers to exogenous variables for Equation (6), including income, unemployment rate, technical level, land slope, rural–urban income gap, and R&D capability.

Because SEM is a complex system composed of multiple equations, there is the possibility of mutual causality among the variables, and it is necessary to identify the model of the simultaneous equations to determine whether it can be estimated by regression. The Simultaneous Equation Model (SEM) system constructed in this paper contains 3 endogenous variables (K = 3) and 15 predetermined variables (G = 15). Equation (4) contains three endogenous variables (Ki = 3) and six predetermined variables (Gi = 6) and the rank of its matrix = K − 1 = 2, which meets the rank condition of identification. Meanwhile, according to the order condition, G − Gi= 15 − 6 = 9 is greater than Ki − 1 = 3 − 1 = 2 in Equation (4), so it is over-identified. The recognition results of the rank and order conditions of Equations (5) and (6) are the same as those of Equation (4), which are also over-identified. Therefore, after judging the rank and order conditions, it can be concluded that SEM meets the identification conditions, and the estimated parameters in all equations are estimable and analyzable in this paper [39].

The Three-Stage Least Squares (3SLS) method is the optimal GMM estimator when the disturbance terms satisfy the conditional homoskedasticity and is better than the Two-Stage Least Squares (2SLS) method [40]. Considering the potential correlation of the endogenous variables and the possible correlation among the stochastic disturbance terms of the equations, the Three-Stage Least Squares (3SLS) method was used to estimate Equations (4)–(6) as a whole [41]. In order to eliminate the heteroskedasticity problem to a certain extent, some variables were logarithmized. The correlation coefficients of the endogenous variables of the equations showed that the correlation coefficients were less than 0.4, indicating no significant multicollinearity problem.

China has a vast territory, and each region’s economic development and geomorphological characteristics are different, with significant regional heterogeneity. Resource endowment, agricultural production conditions, and the level of economic development all affect the interaction mechanism among the three variables. Therefore, this article needs to analyze China as a whole and China’s regional heterogeneity. According to the National Bureau of Statistics of China, China is divided into four regions: East, Central, West and Northeast in

Table 5. Distribution of the Eastern, Central, Western and Northeastern regions.

Eastern	Central	Western	Northeastern
Beijing, Tianjin, Hebei, Fujian, Shanghai, Zhejiang,	Shanxi, Anhui, Jiangxi, Hebei, Hubei, Henan	Inner Mongolia, Guangxi, Chongqing, Sichuan,	Heilongjiang, Liaoning, Jinlin
Shandong, Guangdong, Hainan		Guizhou, Yunnan, Tibet, Shaanxi,
		Ningxia, Xinjiang, Qinghai, Gansu

[42]:

3. Results and Discussion

3.1. ACEE, RLT, and AMI Have an Interactive Effect Mechanism

First, we analyze the interaction mechanism for China as a whole.

Table 6. Results for all of China.

Equation (4)	ACEE	Coefficient	Standard Error	t-Value	p-Value	Significance Level
	RLT	2.777	0.324	8.57	0.000	***
	AMI	8.263	1.068	7.74	0.000	***
	Constant	−25.318	3.733	−6.78	0.000	***
Equation (5)	RLT	Coefficient	Standard Error	t-value	p-value	Significance level
	ACEE	0.046	0.01	4.54	0.000	***
	AMI	1.407	0.163	8.66	0.000	***
	Constant	−4.696	0.545	−8.62	0.000	***
Equation (6)	AMI	Coefficient	Standard Error	t-value	p-value	Significance level
	RLT	0.109	0.033	3.29	0.001	***
	ACEE	−0.03	0.006	−4.65	0.000	***
	Constant	−0.693	0.355	−1.95	0.051	*

Note: *** and * show significance at 1% and 10%.

shows the results of regression of the SEM consisting of Equations (4)–(6) using 3SLS. In the result of Equation (4), the estimated coefficients of rural labor transfer and agricultural machinery intensity are significantly positive at the 1% level; accelerating the transfer of rural labor force and increasing the intensity of agricultural machinery will increase the efficiency of agricultural carbon emissions. Firstly, the transfer of rural labor force leads to the need for farmers to increase the input of other material factors to replace labor. Fertilizers, pesticides, and other means of production increase agricultural carbon emissions while significantly increasing total agricultural output, leading to a rise in agricultural carbon emission efficiency [43]. Secondly, although the rising intensity of agricultural machinery increases the amount of agricultural carbon emissions, the use of mechanical irrigation technology, mechanical fertilizer technology, mechanical farming technology and large-scale operation improves the efficiency of the utilization of factors of production, which has a more positive impact on the improvement of total agricultural output [44]. Therefore, in general, the positive contribution of agricultural machinery inputs to carbon emission efficiency is more significant than the negative inhibiting effect.

As shown in the result of Equation (5), the estimated coefficients of agricultural carbon emission efficiency and agricultural machinery intensity are significantly positive, and the increase in carbon emission efficiency and agricultural machinery intensity positively promotes rural labor transfer. The increase in agricultural carbon emission efficiency means that the same carbon emission can produce more output, the input and output of agricultural factors are more harmonious with the ecological environment, the demand for necessary rural labor decreases, and the phenomenon of farmers going out to work increases [45]. The increase in agricultural machinery input reflects the technological progress of agricultural production, and the popularization of agricultural mechanization makes it possible for agricultural production to produce more output with less labor input, improves labor productivity, and promotes the transfer of rural labor to non-agricultural industries [46].

The result of Equation (6) shows that the estimated coefficients of rural labor transfer are significantly positive at the 1% level, and the estimated coefficients of agricultural carbon emission efficiency are significantly negative at the 1% level. The transfer of rural labor has a positive effect on increasing the intensity of agricultural machinery, while the increase in agricultural carbon emission efficiency reduces the input of agricultural machinery. The positive effect of rural labor force transfer on agricultural machinery input is that with the reduction of the surplus rural labor force, farmers must use agricultural machinery to replace labor force production to maintain the level of total agricultural output, leading to an increase in agricultural machinery input per unit of arable land [47]. The increased efficiency of agricultural carbon emissions will increase the income of local governments and farmers. In the context of carbon reduction in China, the government will actively provide new carbon-reducing technologies and cleaner production methods and establish environmental protection laws that will enable farmers to change their production habits, decreasing the intensity of agricultural machinery [48].

3.2. Significant Heterogeneity across the Four Regions of China

Table 7. Results of the Eastern region.

Equation (4)	ACEE	Coefficient	Standard Error	t-Value	p-Value	Significance Level
	RLT	1.976	0.214	9.22	0.000	***
	AMI	8.263	1.068	7.74	0.000	***
	Constant	−67.625	6.455	−10.48	0.000	***
Equation (5)	RLT	Coefficient	Standard Error	t-value	p-value	Significance level
	ACEE	0.256	0.041	6.31	0.000	***
	AMI	1.398	0.454	3.08	0.002	***
	Constant	−5.873	1.756	−3.34	0.001	***
Equation (6)	AMI	Coefficient	Standard Error	t-value	p-value	Significance level
	RLT	−0.38	0.058	−6.51	0.000	***
	ACEE	−0.024	0.015	−1.58	0.114
	Constant	−0.693	0.355	−1.95	0.051	*

Note: *** and * show significance at 1% and 10%.

shows the results of the Eastern region; rural labor transfer and agricultural machinery intensity in the East significantly contribute to agricultural carbon emission efficiency at 1% level. Agricultural carbon emission efficiency and agricultural machinery intensity significantly contribute to rural labor transfer at 1% level. This is consistent with national results. However, rural labor transfer in the East significantly reduces agricultural machinery intensity, and agricultural carbon emission efficiency does not affect machinery intensity. This is because industries in Eastern China are mainly based on manufacturing and services, with an excellent economic base and a high level of education among farmers. When the transfer of agricultural labor occurs, farmers do not only seek substitution of the factors of production. However, they are more likely to combine agriculture with manufacturing and service industries, reducing the number of agricultural machinery used, and the intensity of agricultural machinery is detached from the effect of carbon emission efficiency [49].

According to

Table 8. Results of the Central region.

Equation (4)	ACEE	Coefficient	Standard Error	t-Value	p-Value	Significance Level
	RLT	−2.614	0.744	−3.51	0.000	***
	AMI	−5.182	1.69	−3.07	0.002	***
	Constant	−56.352	5.43	−10.38	0.000	***
Equation (5)	RLT	Coefficient	Standard Error	t-value	p-value	Significance level
	ACEE	0.072	0.016	4.40	0.000	***
	AMI	−0.031	0.232	−0.13	0.893
	Constant	−1.421	1.093	−1.30	0.194
Equation (6)	AMI	Coefficient	Standard Error	t-value	p-value	Significance level
	RLT	−0.167	0.041	−4.04	0.000	***
	ACEE	−0.02	0.008	−2.51	0.012	**
	Constant	−2.705	0.417	−6.49	0.000	***

Note: *** and ** show significance at 1% and 5%.

, the interaction mechanism of the three variables in the Central region is almost diametrically opposite to that of the whole of China. Rural labor transfer and agricultural machinery intensity significantly reduce agricultural carbon emission efficiency at 1% level. Agricultural machinery intensity does not affect rural labor transfer, while carbon emission efficiency significantly increases rural labor transfer. Both rural labor transfer and agricultural carbon emission efficiency significantly reduce agricultural machinery input. The main reason for this is that Central China is the least economically developed and most populous region in China, and many farmers will flock to work in cities or other economically developed regions every year, leading to the most challenging reversal of the labor transfer trend in Central China [50]. The low level of economic development means that farmers cannot implement cleaner production methods to cope with the labor migration gap, and the misuse of pesticides and fertilizers emits large amounts of carbon dioxide. The low level of technology leads to low carbon productivity of machinery and low carbon efficiency in agriculture. The huge urban–rural income gap reduces farmers’ willingness to engage in agricultural production, and agricultural machinery and labor gradually lose their substitution properties, even to the extent that arable land is abandoned and occupied; then, the number of agricultural machinery is forced to decrease. Therefore, even if the efficiency of agricultural carbon emissions increases, it will not be able to change rising trend in the transfer of agricultural labor and the decrease in the intensity of agricultural machinery.

As

Table 9. Results of the Western region.

Equation (4)	ACEE	Coefficient	Standard Error	t-Value	p-Value	Significance Level
	RLT	9.457	2.072	4.56	0.000	***
	AMI	2.198	1.112	1.98	0.048	**
	Constant	−17.454	5.662	−3.08	0.002	***
Equation (5)	RLT	Coefficient	Standard Error	t-value	p-value	Significance level
	ACEE	0.014	0.004	3.56	0.000	***
	AMI	0.219	0.061	3.57	0.000	***
	Constant	1.009	0.248	4.06	0.000	***
Equation (6)	AMI	Coefficient	Standard Error	t-value	p-value	Significance level
	RLT	3.139	0.283	11.11	0.000	***
	ACEE	−0.119	0.014	−8.36	0.000	***
	Constant	3.139	0.283	11.11	0.000	***

Note: *** and ** show significance at 1% and 5%.

shows, the Western region is in line with China’s results, mainly because of its vast territory, large population, and medium level of economic development, which reflects the situation in China as a whole [51].

In

Table 10. Results of the Northeastern region.

Equation (4)	ACEE	Coefficient	Standard Error	t-Value	p-Value	Significance Level
	RLT	5.614	1.975	2.84	0.004	***
	AMI	5.911	3.616	1.63	0.102
	Constant	−33.643	11.605	−2.90	0.004	***
Equation (5)	RLT	Coefficient	Standard Error	t-value	p-value	Significance level
	ACEE	0.025	0.014	1.78	0.075	*
	AMI	1.032	0.256	4.03	0.000	***
	Constant	0.035	0.651	0.05	0.958
Equation (6)	AMI	Coefficient	Standard Error	t-value	p-value	Significance level
	RLT	0.547	0.06	9.17	0.000	***
	ACEE	−0.01	0.008	−1.35	0.178
	Constant	−0.188	0.3	−0.63	0.531

Note: *** and * show significance at 1% and 10%.

, there is no mechanism for interaction between agricultural machinery intensity and carbon emission efficiency in the Northeast region, but the rest is consistent with the whole of China. The main reason is that the Northeast has a perfect agricultural and industrial base as a significant food-producing province and a traditional industrial base [52]. The three Northeastern provinces now have 448 million mu of arable land, accounting for 23.4 per cent of the country’s total (with per capita arable land being 3.4 times the national average) and possessing the country’s leading agricultural machinery technology and a large-scale operation level [2]. This situation has led to a low marginal carbon contribution of agricultural machinery in the Northeast, and it is challenging to increase machinery inputs to promote carbon efficiency and change the existing pattern of agricultural machinery with increased carbon efficiency. When rural labor is transferred, farmers prefer to seek substitution of other factors of production to improve yields rather than agricultural machinery.

3.3. Results Are Robust

This paper adopts the method of replacing the two endogenous variables of rural labor force transfer and agricultural machinery intensity with the negative growth rate of the agricultural labor force and comprehensive mechanization rate to conduct the robustness test. When the rural labor force undergoes positive transfer, the agricultural labor force will have negative growth, so we choose the negative growth rate of the agricultural labor force to replace the rural labor force transfer. Although it cannot reflect the structural change resulting from rural labor force transfer, it can reflect the direction change. The comprehensive mechanization rate of crop cultivation, planting, and harvesting is chosen as a replacement indicator and is measured by the Ministry of Agriculture of China [53]. It is calculated by the weighted average of 0.4, 0.3 and 0.3 for ploughing, seeding and harvesting, respectively. The level of mechanized ploughing refers to the percentage of mechanized ploughing area in the sown area that should be ploughed (the sown area that should be ploughed is equal to the total sown area minus the no-till sown area). The level of mechanized sowing and the level of mechanized harvesting refers to the percentage of mechanized sowing area and mechanized harvesting area in the sown area and the harvested area, respectively. The comprehensive mechanization rate of crop ploughing, seeding and harvesting directly reflects the level of mechanized crop operations in the region [54].

The 2SLS method is also used to regress the original data to conduct robustness testing. As shown in

Table 11. Robustness test by replacing variables.

Equation (4)	ACEE	Coefficient	Standard Error	t-value	p-Value	Significance Level
	RLT	26.188	2.22	11.80	0.000	***
	AMI	1.553	0.312	4.97	0.000	***
	Constant	2.034	0.502	4.05	0.000	***
Equation (5)	RLT	Coefficient	Standard Error	t-value	p-value	Significance level
	ACEE	0.039	0.004	10.86	0.000	***
	AMI	0.059	0.012	4.75	0.000	***
	Constant	−0.075	0.017	−4.48	0.000	***
Equation (6)	AMI	Coefficient	Standard Error	t-value	p-value	Significance level
	RLT	4.332	0.698	6.21	0.000	***
	ACEE	−0.34	0.046	−7.33	0.000	***
	Constant	−2.246	0.274	−8.20	0.000	***

Note: *** show significance at 1%.

and

Table 12. Robustness test using 2SLS.

Equation (4)	ACEE	Coefficient	Standard Error	t-Value	p-Value	Significance Level
	RLT	1.83	0.338	5.42	0.000	***
	AMI	8.036	1.105	7.27	0.000	***
	Constant	−27.374	3.868	−7.08	0.000	***
Equation (5)	RLT	Coefficient	Standard Error	t-value	p-value	Significance level
	ACEE	0.037	0.01	3.57	0.000	***
	AMI	1.229	0.172	7.13	0.000	***
	Constant	−4.572	0.575	−7.95	0.000	***
Equation (6)	AMI	Coefficient	Standard Error	t-value	p-value	Significance level
	RLT	0.105	0.035	3.04	0.002	***
	ACEE	−0.042	0.007	−6.29	0.000	***
	Constant	−1.047	0.387	−2.71	0.007	***

Note: *** show significance at 1%.

, the robustness test of the replacement variable is consistent with the original results in the direction and significance of the regression coefficients, and the results using 2SLS are very close to the original results, indicating that the regression results of this article are robust [40].

4. Conclusions and Prospects

In this paper, we use the provincial panel data of China from 2000 to 2021 to construct a Simultaneous Equation Model of agricultural carbon emission efficiency, rural labor transfer, and agricultural machinery intensity and use the 3SLS method for regression analysis to draw the following conclusions.

Firstly, the results of model 1 in Table 6 are consistent with Hypothesis 1. Rural labor transfer and agricultural machinery intensity will impact agricultural carbon emission efficiency, and the effect will be significantly positive at the national level. This is because the overall modernization level of Chinese agriculture is still low, and the growth of marginal output value from machinery substitution due to rural labor transfer is higher than the growth of marginal carbon emissions.

Secondly, the results of model 1 to model 3 in Table 6 are consistent with Hypothesis 2. There are interactive influence mechanisms among agricultural carbon emission efficiency, rural labor transfer, and agricultural machinery intensity. The three causal relationships and transmission mechanisms are not unidirectional but interactive and complex. The interaction of the two variables may be in different directions. For example, the increase in the intensity of agricultural machinery promotes the efficiency of agricultural carbon emissions. However, the increase in the efficiency of agricultural carbon emissions inhibits the increase in the intensity of agricultural machinery, which suggests that the marginal contribution of machinery to the efficiency of carbon emissions from agricultural production is decreasing and that mechanization and large-scale production cannot be relied upon to improve the efficiency of agricultural carbon emissions.

Finally, Table 7, Table 8, Table 9 and Table 10 results show significant regional heterogeneity in China. Especially in the Central region, rural labor transfer and agricultural machinery intensity reduce the efficiency of agricultural carbon emissions, and the increase in agricultural machinery will alleviate rural labor transfer. It is the exact opposite of the other three regions. In the East, due to a better economic base and deeper integration of agriculture with secondary and tertiary industries, when rural labor force shifts occur, farmers prefer to move to secondary and tertiary industries rather than increase the intensity of machinery to maintain output. In the Northeastern region, because the rate of agricultural machinery is already high, the marginal contribution of machinery to carbon efficiency is meager, and increasing machinery intensity does not improve carbon efficiency, nor does an increase in carbon efficiency affect machinery intensity.

Based on the above conclusions, this paper puts forward the following policy recommendations.

First, from the perspective of rural labor force transfer, labor force transfer will increase mechanical strength, and increased mechanical strength will further promote labor force transfer. Moreover, the current trend of rural labor force transfer is difficult to reverse. The government should promote the transfer of rural land from farmers to agribusiness to achieve large-scale operation. Land transfer can effectively use the idle or abandoned farmland after the large-scale transfer of rural labor, and more professional and enthusiastic operators can operate the transferred land on a large scale, providing a good situation for other factors of production to replace the labor force and improving the efficiency of agricultural production.

Secondly, given the complex interaction mechanism of agricultural carbon emission efficiency, agricultural village labor transfer, and agricultural machinery intensity, the government should do more than just increase the intensity of machinery to improve the efficiency of agricultural carbon emissions when dealing with the rural labor shortage. The government should strengthen research on agricultural machinery technology and the renewal of clean agricultural machinery to cope with the gap in the amount of rural labor and the increase in cost. At the same time, it should increase subsidies for farmers to purchase cleaner production tools and materials and provide training to farmers to promote green agriculture development in China.

Finally, due to the significant heterogeneity of China’s regions, local governments should tailor their policies to the local conditions based on their economic base, technological level and farmers’ perceptions. For example, in the economically developed Eastern region, the government should encourage the integration of agriculture with production and service industries and encourage farmers to move to non-agricultural industries instead of non-agricultural regions. In the less developed Central areas, the government should correctly handle the relationship between rural labor transfer and agricultural mechanization, give full play to the comparative advantages of labor and machinery, and improve the total efficiency of agricultural production. In the Northeastern region, which has the best industrial base, the government should speed up the updating of green technology of agricultural machinery and give full play to its industrial advantages.

A summary of the theoretical contributions as well as the practical implications of this paper can therefore be presented:

Firstly, this paper incorporates the substitution of agricultural machinery into a system for analyzing the effect of rural labor transfer on the efficiency of agricultural carbon emissions. Moreover, through theoretical analysis and extension of Kaya identity, it is argued that labor transfer, machinery substitution and carbon emission efficiency are not unidirectional conduction mechanisms but multidirectional and complex interaction influence mechanisms. The theoretical framework proposed in this paper is proved correct by constructing the linkage model and using the 3SLS regression method to test Hypotheses 1 and 2. This paper extends the understanding of the relationship between rural labor transfer, agricultural machinery intensity and agricultural carbon emission efficiency. It enriches the theory of the influence mechanism of the three.

Secondly, this paper puts forward relevant policy recommendations based on the interaction mechanism of rural labor transfer, agricultural machinery intensity and agricultural carbon emission efficiency. It provides a new perspective for the government to deal with rural labor transfer, machinery input, and agricultural carbon reduction. At the same time, this paper analyzes the heterogeneity of China into four regions and finds the differences in the interaction mechanisms of the four regions, which provides a reference for local governments to implement policies.

However, this paper uses provincial data, which is a weaker explanation of problems in cities and counties. Additionally, this paper controls the level of technology and natural resources as exogenous variables and ignores their effects. Therefore, it is still an object of future research.