Land Titling: A Catalyst for Enhancing China Rural Laborers’ Mobility Intentions?

Abstract

Land titling, a critical land institution reform aimed at enhancing tenure security, serves as a pivotal policy instrument to strengthen rural laborers’ mobility intentions. Leveraging a balanced panel dataset from the 2014 and 2016 China Labor-force Dynamic Survey (CLDS), this study employs a difference-in-differences (DID) model to evaluate the policy effects of the latest round of land titling on rural laborers’ mobility intentions. The results demonstrate that land titling significantly enhances rural laborers’ willingness to migrate. To ensure robustness, we incorporate individual and year fixed effects, cluster robust standard errors at the household level, and conduct multiple robustness tests, including placebo test, propensity score-matching difference-in-differences (PSM-DID), replacement of dependent variable, clustered adjustment, adding control variables and interaction fixed effects. Mechanism analysis reveals that land titling elevates laborers’ mobility intentions primarily by reducing land reallocation and stimulating investments in agricultural machinery. Heterogeneity analysis further identifies stronger effects in villages dominated by agricultural employment, and among middle-aged laborers. These findings highlight the nuanced role of tenure security in reshaping rural laborer dynamics and provide empirical support for optimizing land-related policies to facilitate structural transformation.

1. Introduction

Since the country’s reform and opening-up, the mobility of rural laborers has not only served as a critical driver of China’s rapid industrialization and urbanization in the past, but also remains a vital force in sustaining economic growth and enhancing total factor productivity. Therefore, rural laborer mobility persists as a significant practical issue in China’s socioeconomic development [1]. Reducing rural residents’ concerns about migrating to urban areas constitutes an essential pathway for advancing new-type urbanization and achieving the comprehensive establishment of a modern socialist society [2]. According to the Migrant Worker Monitoring Survey Report released by the National Bureau of Statistics, the total number of rural laborers engaged in off-farm employment reached 166.1 million in 2013, reflecting 1.7% annual growth. By 2023, this figure had risen to 176.5 million, yet marked a decline of 2.2% compared to the previous year [3]. This trend indicates a stabilization in the total outflow of rural laborer, with growth rates slowing or even contracting. A fundamental factor underlying this phenomenon lies in the diminished laborers’ mobility intentions caused by insufficient stability of rural land-use rights. Land serves as the foundation for rural laborers’ livelihoods and development, embodying both material security and emotional attachment [4]. Simultaneously, as the most fundamental institutional arrangement governing production relations, the stability of the land tenure system directly shapes rural laborers’ migration decisions and the level of urban-rural integrated development. At the same time, as the most basic production relations arrangement and system operation framework of the country, the stability of the land system is directly related to laborers’ mobility intentions and the level of urban–rural integration. As such, ensuring stable land property rights transcends individual interests; it constitutes an institutional prerequisite for optimizing laborer allocation between urban and rural areas and advancing Chinese-style modernization [4].

Enhancing rural laborers’ mobility intentions in China carries strategic importance for socioeconomic advancement. First, promoting urban–rural labor transitions through elevated mobility propensities effectively stimulates domestic economic circulation and underpins high-quality development [2]. As this developmental paradigm forms the cornerstone of socialist modernization, enabling laborers to access urban sectors and non-agricultural industries with superior productivity yields dual benefits, it elevates household earnings and living standards while simultaneously achieving dual goals of domestic demand expansion and sustainable economic upgrading. Second, mobility-driven laborer migration establishes fundamental prerequisites for agricultural transformation, specifically scaling, modernization, and specialization [5]. Progressive urbanization reduces agricultural workforce engagement, promoting land consolidation among new agricultural operators. This transition addresses land fragmentation challenges while accelerating mechanization processes, ultimately empowering modern agricultural entities to achieve industrialized and specialized production systems.

Existing studies across developing nations have demonstrated that land titling strengthens rural laborers’ migration propensity through enhanced tenure security [6,7]. China’s 2013 Central No. 1 Document formally initiated a nationwide campaign for comprehensive land titling, setting a five-year implementation timeline. By 2020, this initiative achieved remarkable coverage, with over 96% of rural households certified, extending benefits to roughly 200 million families. Within this policy framework, our research investigates two critical dimensions: first, whether land titling effectively promotes rural laborers’ mobility intentions; second, the operational mechanisms behind this effect, particularly through decreased frequency of land reallocations and increased investments in agricultural machinery. Additionally, the study analyzes variations in policy impacts across villages with differing levels of part-time employment and among workforce cohorts of varying age brackets. These dual investigations form the principal research focus of this paper.

Existing research on how land titling affects rural laborers’ mobility intentions remains limited, particularly regarding the analysis of causal mechanisms. Earlier studies mainly rely on OLS or panel models [8,9,10], which are vulnerable to endogeneity problems like sample selection bias and omitted variables. This study addresses these gaps by analyzing two-wave data (2014 and 2016) from the China Labor-force Dynamics Survey (CLDS). Using a difference-in-differences (DID) approach, it evaluates how land titling policies influence rural laborers’ willingness to migrate, offering more robust causal evidence than prior methods. We further investigate its intrinsic mechanisms and conduct heterogeneity analyses. The contributions of this research are threefold: first, in terms of research perspective, unlike prior studies that predominantly measure labor mobility through indicators such as “non-agricultural wages”, “non-agricultural household registration”, and “non-agricultural population”, this paper innovatively assesses “whether they intend to migrate to urban areas for employment” to capture rural laborers’ mobility intentions. This paper focuses on the understudied dimension of willingness, examining how enhanced tenure security influences the mobility intentions of rural laborers. Second, methodologically, we employ the DID framework for policy evaluation and reinforce the robustness of our findings using propensity score matching difference-in-differences (PSM-DID). Third, regarding research content, we conduct mechanism analysis to verify how land titling affects mobility intentions through land reallocation and agricultural machinery investments. Additionally, we perform heterogeneity analysis based on the part-time employment degree differences in villages and the age differences among laborers.

1.1. Policy Background

As a key institutional reform in China’s rural land management since the household responsibility system (HRS), land titling has developed through different stages marked by pilot trials and nationwide expansion [11]. Starting in 1978, the HRS initially allocated collective farmland equally based on family size. However, frequent land redistributions caused by population changes (births, deaths, aging) and urban migration gradually weakened farmers’ tenure security, leading to unclear property rights. To solve these issues, the 1980s saw the introduction of farmland contracts defining usage rights periods. Policy improvements continued with the 1984 Central Document No.1 setting 15-year contracts, followed by 1993 reforms introducing 30-year terms. Both were later written into the 2002 Rural Land Contracting Law requiring official certifications. Despite this progress, challenges like incomplete regulations, urban growth demands, and restricted movement of rural laborers continued affecting land distribution efficiency and rights enforcement.

The 2008 Central Document No.1 marked a paradigm shift by prioritizing nationwide land registration infrastructure development. Initial pilots launched in 2009 across Shenyang, Beijing, and Chengdu tested institutional designs for subsequent generalization. A critical juncture emerged in 2013 when Central Document No.1 formalized a five-year national titling completion roadmap, representing the systematic institutionalization of contemporary land titling reforms. Comparative analysis reveals three cardinal advancements distinguishing this reform cycle.

First, precision in rights delineation through technical institutionalization. The reform introduced geospatial technologies including professional cadastral surveying, digital mapping systems, and GPS boundary demarcation, enabling the precise categorization of land parcels, ownership structures, and spatial parameters. Concurrent establishment of a national registration database achieved comprehensive integration and dynamic monitoring of contractual management rights across administrative tiers, significantly enhancing data integrity for tenure governance.

Second, empowerment through expanded land rights. Post-titling institutional arrangements granted farmers enhanced autonomy in land utilization decisions beyond agricultural production, including mortgage collateralization and equity participation mechanisms. This rights expansion facilitated market-oriented transactions and financial instrument development, effectively unlocking previously dormant land asset value—a critical enabler for rural capital formation.

Third, legal reinforcement of tenure security. Comprehensive audits of contractual rights enabled standardized agreement refinements and nationwide certification of farmland management rights, creating legally binding documentation systems. These institutional safeguards not only strengthened perceived tenure security among landholders but also established juridical foundations for dispute resolution and rights enforcement during land transfers.

This study situates the 2013 reform as a critical institutional threshold for analyzing land titling’s impacts on rural laborer reallocation dynamics, recognizing its importance and national unity in property rights reform. This reform’s dual focus on tenure security enhancement and factor market development provides an optimal institutional context for examining contemporary rural transformation pathways.

1.2. Literature Review

Since the implementation of land titling policies, academic research on their effects has predominantly focused on land as the object of study, emphasizing specific aspects such as land transfer, credit collateralization, and agricultural investment. However, relatively limited attention has been devoted to farmers as active laborer participants, with particularly scarce literature directly examining how land titling influences rural laborers’ mobility intentions. Rural laborer mobility refers to changes in workers’ employment types and geographic locations, while mobility intention reflects their underlying motivations and subjective inclinations to alter these dimensions based on comprehensive considerations of economic, familial, and personal factors [12]. This study appropriately operationalizes laborer mobility intention through the indicator of “farmers’ willingness to migrate for urban employment”.

According to Todaro’s rural–urban migration theory, laborers’ decisions to migrate primarily hinge on the expected income differential between urban and rural areas [13]. The Todaro model emphasizes that migration choices are based not merely on actual wage gaps but on a rational calculation comparing urban expected income (urban wages multiplied by employment probability) with rural earnings, adjusted for migration costs. Within this framework, the impact of land titling on labor mobility can be analyzed through two dimensions. On the one hand, as farmland traditionally functions as a social safety net for rural laborers, land titling legally secures land rights, substantially reducing the risk of land loss during migration. This mitigates the opportunity cost in migration decisions: while Todaro’s model assumes zero income upon migration failure, certified land acts as a “risk buffer”, enabling laborers to return to agricultural production if urban employment fails, thereby increasing their tolerance for urban job market uncertainties. On the other hand, by stabilizing land tenure, the policy indirectly improves laborers’ expectations of urban non-agricultural income—when land loss risk approaches zero due to institutional safeguards, farmers weigh urban potential earnings more heavily in their expected income calculations, rather than overestimating threats to rural livelihoods. Thus, land titling enhances the positive incentive effect of “expected income differentials” in Todaro’s model through dual mechanisms—risk reduction and income expectation optimization—ultimately promoting more proactive rural labor mobility toward urban sectors. Land titling enhances tenure security, substantially reducing land dispossession risks during off-farm employment, thereby potentially strengthening laborers’ mobility intentions.

The existing literature explores this relationship through multiple dimensions: regarding non-agricultural employment, the land titling policy significantly reduces the probability of land being bound to inefficient laborers, thereby promoting laborers’ engagement in non-agricultural sectors [14,15]. In terms of rural laborer migration rates, regions with lower land loss risks and higher land transfer levels exhibit elevated out-migration ratios among farmers [16,17]. In the context of rural-to-urban transition, studies demonstrate that land titling facilitates laborers’ urban settlement as “new citizens” by lowering relocation costs and enhancing tenure security [18,19]. Concerning migration duration, empirical evidence indicates that the policy prolongs farmers’ migration periods, with more pronounced effects on those possessing higher human capital [20]. Collectively, these impacts on non-agricultural employment, rural laborer migration rates, urban integration, and migration duration underscore the policy’s role in optimizing human capital allocation between rural and urban areas. Building on this foundation, the first hypothesis is proposed:

Hypothesis 1 (H1).

Land titling positively enhances rural laborers’ mobility intentions.

Land reallocation involves adjusting farmers’ land management rights based on population changes to maintain fair access to land-use rights within rural communities. It mainly appears as “major reallocation” (full redistribution across the village) or “minor reallocation” (partial adjustments during contracts) [21]. In theory, this practice raises tenure insecurity by increasing land loss risks, reducing possible rental income from land transfers. In contrast, land titling strengthens tenure security through official registration, offering stable property rights for rural laborers working in cities. Studies agree that land reallocation worsens tenure instability, raises transaction risks in land transfers, and limits rural laborers’ mobility. Meanwhile, land titling cuts reallocation frequency by clarifying ownership, boosting awareness of rights protection, and weakening informal rules. Specifically, first, land titling legally confirms land usage rights, improving stability and lowering reallocation chances. Second, it strengthens the “endowment effect” of land as vital property, encouraging rural laborers to resist reallocation and keep contracts stable [21]. Third, as an informal practice, land reallocation loses influence under the formal rules created by titling [22]. Together, the policy reshapes farmers’ understanding, secures their rights identity, reduces reallocations, and weakens its negative impact on labor mobility, ultimately boosting rural laborers’ willingness to move [23]. Based on this, the hypothesis states:

Hypothesis 2 (H2).

Land titling reduces land reallocation frequency, enhances tenure security, and thereby increases rural laborers’ mobility intentions.

Agricultural machinery investment encompasses mobile productive assets in agriculture that are distinct from land-based investments [24]. Mechanization theoretically reduces labor constraints in farming while elevating production costs, prompting liberated workers to seek urban employment opportunities with higher remuneration. Academic consensus identifies three primary pathways through which land titling enhances mechanization: rights stabilization, land market facilitation, and credit accessibility improvement [25]. Firstly, formalized land ownership reinforces tenure security and exclusivity, thereby extending farmers’ investment horizons and encouraging capital-intensive machinery adoption. Secondly, by mitigating transaction uncertainties in land markets, titling promotes plot consolidation and scale operations, essential prerequisites for mechanized farming. Thirdly, the enhanced collateral value of titled land expands agricultural financing options, effectively alleviating credit constraints that traditionally hinder equipment purchases [26]. This institutional innovation consequently accelerates machinery adoption, which in turn amplifies labor mobility by reducing agricultural workforce demands. The resultant labor reallocation effect channels human resources toward urban labor markets [27]. Based on this mechanistic analysis, we posit that:

Hypothesis 3 (H3).

Land titling increases agricultural machinery investment, promotes labor substitution through mechanization, and consequently enhances rural laborers’ mobility intentions.

2. Materials and Methods

2.1. Data Sources

This study utilizes the China Labor-force Dynamics Survey (CLDS) database, a nationally representative micro-database jointly established by Sun Yat-sen University (Guangzhou, China) and 27 domestic partner institutions. As one of China’s most authoritative micro-level datasets, the CLDS project involved collaborative efforts from over 800 researchers, supervisors, and interviewers. In 2012, the survey successfully covered 303 villages, 10,612 households, and 16,253 individuals across China. Subsequent waves of tracking surveys were conducted in 2014, 2016, and 2018 to maintain temporal continuity.

The CLDS database comprises three interconnected questionnaires—community, household, and individual—focusing on the status and dynamics of China’s labor force. Its research scope encompasses education, employment, migration, health, social participation, economic activities, grassroots governance, and other interdisciplinary themes. To ensure national representativeness, the survey covers 29 provinces/municipalities (excluding Hong Kong, Macau, Taiwan, Tibet, and Hainan), targeting all working-age family members (15–64 years old) in sampled households. Methodologically, CLDS employs a multistage, stratified, and probability proportional to size (PPS) sampling framework to address China’s rapidly evolving socioeconomic context. Notably, it pioneered a rotating panel tracking approach in China, where samples are randomly divided into four cohorts. Each cohort undergoes four consecutive tracking waves (six years) before replacement, effectively balancing longitudinal continuity with cross-sectional survey advantages. This dual design accommodates both dynamic social changes and the need for comparative temporal analysis. To investigate the impact of land titling on rural laborers’ mobility intentions, we extracted 5016 valid observations from the 2014 and 2016 survey waves. Key variables include land titling, rural-to-urban migration decisions, land reallocation, agricultural machinery investments, urban settlement intentions, and covariates at individual, household, and village levels. Given the absence of the key variable “land titling” in the 2012 CLDS data and substantial missing values in this critical variable within the 2018 CLDS data, this study utilizes two waves of CLDS panel data from 2014 and 2016. To enhance data quality, we implemented rigorous data processing and sample selection procedures. Specifically, the process involved three steps. First, we retained only two categories of samples: those with current agricultural hukou status and those previously registered as agricultural hukou holders but now converted to resident hukou status. Non-agricultural hukou samples and those with prior non-agricultural hukou status were systematically excluded to focus on rural laborer force analysis. Second, we applied a 1% two-sided winsorization to continuous variables to mitigate outlier effects. Finally, samples with severe missing values in primary variables were removed to ensure data integrity and analytical accuracy.

2.2. Definition of Varibles

1. Dependent variable: rural laborer mobility intention. The study measures rural labor mobility intention using variables that reflect mobility propensity and motivation. Drawing on the CLDS database, migration intention is identified through the question: “Do you plan to seek employment outside your hometown?” Respondents who answered affirmatively were coded as 1, while those who declined were coded as 0. The analysis specifically targets rural residents without prior migration experience or urban settlement status.

2. Core explanatory variable: land titling. Following Sun et al. [25], land titling is operationalized through the household survey question: “Has your household obtained the Rural Land Contract Management Right Certificate?” Households with land titling are coded as 1, and those without as 0. This measurement captures the completion of land tenure formalization, a pivotal step in China’s rural land institutional reforms.

3. Control variables: aligned with methodologies from Sun et al. [25] and Weng and Hu [28], controls span three dimensions. These variables include individual characteristics (age, marital status), political affiliation (Communist Party membership), and health status, accounting for demographic influences on migration decisions. Factors such as age, marriage, health, and so on were considered to have an impact on the results. Household attributes included car, single-child status, and agricultural occupation, controlling for economic capacity and livelihood patterns. Village features included agricultural land area, addressing localized economic structures that shape income opportunities.

Table 1. Meanings of variables and descriptive statistics.

Variables	Definition	Mean	S.D.
Rural laborer mobility intention	1 = yes, 0 = no	0.050	0.218
Land titling	1 = yes, 0 = no	0.520	0.500
Age	Years	45.540	14.834
Married	1 = yes, 0 = no	0.867	0.340
Communist Party membership	1 = yes, 0 = no	0.054	0.226
Health	Health status: 5 = very healthy, 4 = healthy, 3 = average, 2 = relatively unhealthy, 1 = very unhealthy	3.581	1.029
Car	1 = yes, 0 = no	0.170	0.376
Single-child	1 = yes, 0 = no	0.075	0.263
Agricultural occupation	1 = yes, 0 = no	0.706	0.456
Agricultural land area	Total agricultural land area (acres)	5683.154	7313.424

Notes: The total area of agricultural land is the original value shown in the table, and the corresponding value is used in the regression analysis.

summarizes variable definitions and descriptive statistics, maintaining consistency with prior empirical frameworks.

2.3. Modeling

This study employs the difference-in-differences (DID) model to identify the causal effect of land titling policies on rural laborers’ mobility intentions. Following the inherent logic of DID methodology, we treat specific policy implementations as quasi-natural experiments by constructing treatment groups (affected areas post-policy implementation) and control groups (consistently unaffected regions). This empirical strategy effectively isolates policy impacts from confounding factors through systematic comparisons of pre–post intervention differences between these groups, thereby mitigating potential endogeneity issues in policy effect estimation. The baseline regression model is specified as follows:

$$Y_{it}={\beta }_0+{\beta }_1{DID}_{it}+Z_{it}+{\delta }_i+{\alpha }_t+{\epsilon }_{it}$$

(1)

In Equation (1), $Y_{it}$ represents the dependent variable (rural labor mobility intention), $i$ represents laborers, and $t$ represents the survey year. ${DID}_{it}$ represents the core explanatory variable (land titling). ${DID}_{it}={treatment}_i\times {post}_t,$ If the household has received the land titling, then ${treatment}_i=1$; otherwise, it is $0$. When $t\ge 2013$, ${post}_t=1$; otherwise, it is $0$. ${\beta }_0$ represents the intercept term, ${\beta }_1$ represents the policy effect coefficient of land rights confirmation, $Z$ represents a series of control variables including personal characteristics, household characteristics, and village characteristics, ${\delta }_i$ represents individual fixed effects to control for individual factors that affect laborers’ plans to work in cities but do not change over time; ${\alpha }_t$ represents year fixed effects to control for macroeconomic factors and other policy factors that change over time but do not vary with individual laborers; ${\epsilon }_{it}$ represents the random disturbance term.

3. Results

3.1. Benchmark Regression Results

To examine the impact of land titling on rural laborer mobility, we employ a stepwise regression approach for Equation (1). Initially, we regress the dependent variable solely on the core explanatory variable (land titling). Subsequently, we incorporate control variables encompassing individual, household, and village characteristics. Finally, we enhance model robustness by introducing individual fixed effects, year fixed effects, and cluster-robust standard errors at the household level. As shown in Column (1) of

Table 2. Impact of land titling on rural laborer mobility: baseline regression results.

	Rural Laborer Mobility Intention
	(1)	(2)	(3)
Land titling	0.014 *** (0.005)	0.016 *** (0.005)	0.022 *** (0.007)
Age		−0.000 (0.000)	−0.000 (0.000)
Married		0.005 (0.009)	0.016 (0.015)
Communist Party membership		0.009 (0.012)	−0.001 (0.018)
Health		−0.006 ** (0.002)	−0.003 (0.004)
Car		−0.009 (0.007)	−0.021 ** (0.010)
Single-child		−0.023 ** (0.009)	−0.015 (0.015)
Agricultural occupation		0.006 (0.006)	0.001 (0.011)
Agricultural land area		0.001 (0.002)	−0.010 ** (0.005)
Year fixed effect	No	No	Yes
Individual fixed effect	No	No	Yes
Observations	8322	7168	5016
Adjusted R-square	0.001	0.004	0.555

Notes: **, and *** denote significance levels at 5% and 1%, respectively; robust standard errors clustered at the household level are reported in parentheses; values 0.000 and −0.000 result from rounding to three decimal places.

, the baseline regression without control variables reveals a statistically significant positive association at the 1% level. This relationship persists in Column (2) after controlling for demographic and socioeconomic covariates. Column (3) further demonstrates robustness through fixed effects and clustered error adjustments, with coefficients remaining positive and statistically significant across specifications. These consistent results substantiate that land titling significantly enhances rural laborer mobility, thereby confirming Hypothesis 1.

3.2. Robustness Tests

1. Placebo test: to validate the robustness of our regression findings and address potential confounding from unobserved heterogeneity, we implemented a comprehensive placebo test framework following contemporary methodological advancements in quasi-experimental research. Drawing on the counterfactual validation approaches advocated by Ferraro et al. [29] and Chetty et al. [30], we randomized the land titling policy exposure across 5016 non-treated laborers in our sample. This randomization procedure, rigorously replicated through 500 independent iterations, simulates a null distribution of treatment effects under the assumption of no causal relationship. The placebo test results, as illustrated in Figure 1, demonstrate that the randomly assigned land titling coefficients are predominantly clustered around zero, forming a normal distribution that is distinctly different from the true estimate of 0.022. This divergence confirms that the baseline regression results remain robust and are not biased by unobserved confounding factors. The application of placebo test as a robustness check for difference-in-differences models has gained increasing recognition in the recent literature, consistent with methodologies employed by Shen et al. [31] and Zheng et al. [32].

2. PSM-DID model: to mitigate sample selection bias between treatment and control groups, this study matches laborers using propensity score matching (PSM) based on relevant covariates to enhance group homogeneity. The matched sample is then subjected to difference-in-differences (DID) analysis, incorporating individual fixed effects, year fixed effects, and cluster-robust standard errors at the household level to strengthen result credibility. Radius matching is employed for the PSM procedure [33,34,35]. Figure 2 illustrates that the treatment and control groups exhibit significant differences prior to matching, which are substantially reduced post-matching, passing balance tests, and alleviating selection bias. As shown in

Table 3. Robustness test Two: PSM-DID.

	Rural Laborer Mobility Intention
	(1)	(2)	(3)
Land titling	0.014 *** (0.005)	0.017 *** (0.005)	0.022 *** (0.007)
Control variable	No	Yes	Yes
Year fixed effect	No	No	Yes
Individual fixed effect	No	No	Yes
Observations	8279	7125	4988
Adjusted R-square	0.001	0.004	0.556

Notes: *** represent the significance level of 1%, respectively; the robust standard error of clustering at the household level is in parentheses.

, the estimated coefficients, signs, and significance levels align closely with the baseline results in Table 2. This consistency confirms the robustness of the finding that land titling significantly enhances rural laborers’ mobility intentions.

3. Replacement of dependent variable: intention to settle in urban areas. The substantial mobility of China’s rural laborer force has contributed to urbanization. However, due to institutional barriers such as household registration system, rural migrants working in cities for extended periods often fail to obtain urban household registration, resulting in circular migration patterns at the micro-level. Whether laborers intend to settle in urban areas reflects their motivation for permanent relocation. To verify the robustness of baseline regression results, we replace the dependent variable with “whether individuals plan to settle in urban areas”, derived from the CLDS individual questionnaire. The sample size for this analysis is 6668. As shown in

Table 4. Robustness Test Three: replacing the explained variable.

	Rural Laborer Mobility Intention
	(1)	(2)	(3)
Land titling	0.003 (0.005)	0.005 (0.005)	0.018 * (0.009)
Control variable	No	Yes	Yes
Year fixed effect	No	No	Yes
Individual fixed effect	No	No	Yes
Observations	9942	8598	6668
Adjusted R-square	0.000	0.014	0.618

Notes: * represent the significance level of 10%, respectively; the robust standard error of clustering at the household level is in parentheses.

, after absorbing individual fixed effects, year fixed effects, and incorporating clustered robust standard errors at the household level, agricultural land titling policy demonstrates a positive and statistically significant influence on farmers’ intentions to settle in urban areas.

4. Adding control variables: gender and educational level. Considering that the migration intentions of rural laborer are also influenced by gender and educational level [36,37], this study incorporates gender and individual educational level into the baseline model specified in Equation (1) to examine their impact on rural labor migration intentions after controlling for these factors. The results in column (2) of

Table 5. Robustness Test Four: adding control variables.

	Rural Laborer Mobility Intention
	(1)	(2)
Land titling	0.017 *** (0.005)	0.022 *** (0.008)
Gender	−0.009 * (0.005)	−0.008 (0.009)
Educational level	0.001 (0.001)	0.002 (0.001)
Control variable	Yes	Yes
Year fixed effect	No	Yes
Individual fixed effect	No	Yes
Observations	7160	5008
Adjusted R-square	0.004	0.556

Notes: * and *** represent the significance level of 10% and 1%, respectively; the robust standard error of clustering at the household level is in parentheses.

demonstrate that after including these control variables and accounting for individual and year fixed effects, the estimated coefficient of land titling remains consistent with that in column (3) of Table 2, showing a statistically significant positive effect on rural labor migration intentions at the 1% level.

5. Clustered adjustment: drawing on relevant research [38,39], we employed county-level clustered robust standard errors. Given that Chinese society is relationship-based and familiar, laborers’ migration decisions are often influenced by their social networks. However, external influencing factors vary across geographical regions. We thus assumed that migration decisions among laborers within the same county are interdependent, whereas those across different counties remain independent. The sample size obtained through county-level clustering remains smaller than that derived from household-level clustering due to missing administrative codes for certain counties where sampled laborers reside. Based on Model (1) as shown in

Table 6. Robustness Test Five: clustered adjustment.

	Rural Laborer Mobility Intention
	(1)
Land titling	0.022 *** (0.008)
Control variable	Yes
Year fixed effect	Yes
Individual fixed effect	Yes
Observations	4904
Number of districts and counties clustering	92
Adjusted R-square	0.555

Notes: *** represent the significance level of 1%, respectively; the robust standard error of clustering at the household level is in parentheses.

, adjusting the standard errors to the county level revealed that the coefficient of land titling exhibited minimal divergence from Table 2, while still demonstrating a statistically significant positive impact on rural labor mobility at the 1% level.

6. Interaction fixed effects: we incorporate city × year fixed effects and county × year fixed effects into the regression model based on relevant research [40]. To account for spatiotemporal heterogeneity in economic–demographic characteristics and policy shocks across cities and counties over time, we augment the baseline specification with city × year and county × year fixed effects. As shown in Columns (1) and (2) of

Table 7. Robustness Test Six: interaction fixed effects.

	Rural Laborer Mobility Intention
	(1)	(2)
Land titling	0.022 *** (0.007)	0.022 *** (0.008)
Control variable	Yes	Yes
Year fixed effect	Yes	Yes
Individual fixed effect	Yes	Yes
City × year fixed effect	Yes	No
County × year fixed effect	No	Yes
Observations	4898	4896
Adjusted R-square	0.565	0.566

Notes: *** represent the significance level of 1%, respectively; the robust standard error of clustering at the household level is in parentheses.

, the estimated coefficient of land titling remains consistent with Column (3) in Table 2, demonstrating a significantly positive impact on rural laborers’ mobility intentions at the 1% level. This robustness check confirms that the core mechanism linking tenure security to laborer reallocation persists even after absorbing finer-grained spatial-temporal variations.

3.3. Mechanism Analysis

1. Land Titling, Land Reallocation, and Laborers’ Mobility Intentions.

Since the implementation of the Household Responsibility System, frequent land reallocation driven by contractual term limitations and demographic changes have emerged as a primary factor undermining land tenure stability. Whether the new round of land titling achieves policy effectiveness in reducing land reallocation and enhancing tenure stability requires empirical validation. Using the village-level survey data from the China Labor-force Dynamics Survey (CLDS), this study measures land reallocation through the binary indicator “Has village land been readjusted? (Yes = 1, No = 0)”. As shown in

Table 8. Mechanism analysis 1.

	Land Reallocation
	(1)	(2)	(3)
Land titling	−0.054 *** (0.009)	−0.060 *** (0.009)	−0.137 ** (0.054)
Control variable	No	Yes	Yes
Year fixed effect	No	No	Yes
Individual fixed effect	No	No	Yes
Observations	10,552	9321	7386
Adjusted R-square	0.004	0.013	0.613

Notes: **, *** represent the significance level of 5% and 1%, respectively; the robust standard error of clustering at the household level is in parentheses.

, the implementation of the land titling policy reduces the occurrence of land reallocation at the 5% level. The decreased frequency of land reallocation contributes to improve the stability of land tenure system and enhance the security of property rights, thereby mitigating laborers’ concerns about losing land upon migration to urban areas and ultimately enhancing rural laborers’ mobility intentions.

2. Land Titling, Agricultural Machinery Investment, and Laborers’ Mobility Intentions.

This study draws on the methodological framework established by Sun et al. (2020) [25], utilizing data from the China Labor-force Dynamics Survey (CLDS) household questionnaire. Specifically, agricultural machinery investment is measured through the survey item: “Does your household own a tractor? (Yes = 1, No = 0)”. As shown in

Table 9. Mechanism analysis 2.

	Agricultural Machinery Investment
	(1)	(2)	(3)
Land titling	0.107 *** (0.007)	0.091 *** (0.008)	0.027 * (0.015)
Control variable	No	Yes	Yes
Year fixed effect	No	No	Yes
Individual fixed effect	No	No	Yes
Observations	10,819	9370	7450
Adjusted R-square	0.021	0.061	0.822

Notes: *, *** represent the significance level of 10% and 1%, respectively; the robust standard error of clustering at the household level is in parentheses.

, the implementation of land titling significantly enhances farmers’ willingness to invest in agricultural machinery. Increased machinery investment facilitates laborer substitution in agricultural production, thereby releasing more rural laborers into non-agricultural sectors and providing greater opportunities for rural laborers to migrate to urban areas. This substitution effect ultimately strengthens rural laborers’ intentions to pursue laborer mobility through the improvement of agricultural mechanization.

3.4. Heterogeneity Analysis

1. Based on the perspective of village part-time employment degree.

Drawing on the classification framework of part-time employment degree proposed by Zhang and Qian [41] and Yang et al. [42], this study categorizes villages based on the proportion of laborers engaged in non-agricultural industries. Villages with less than 10% non-agricultural labor are classified as pure agricultural villages, those with 10–50% as agriculture-dominant villages, and those exceeding 50% as non-agriculture-dominant villages.

Table 10. Heterogeneity analysis 1.

	Degree of Part-Time Employment in the Village
	Pure Agricultural Villages	Agriculture-Dominant Villages	Non-Agriculture-Dominant Villages
Land titling	0.033 (0.022)	0.021 ** (0.009)	0.018 (0.039)
Control variable	Yes	Yes	Yes
Year fixed effect	Yes	Yes	Yes
Individual fixed effect	Yes	Yes	Yes
Observations	694	4252	398
Adjusted R-square	0.599	0.554	0.494

Notes: ** represent the significance level of 5%, respectively; the robust standard error of clustering at the household level is in parentheses.

demonstrates the differential impacts of land titling on laborers’ mobility intentions across villages with varying part-time employment degree. The land titling policy exhibits heterogeneous impacts across village typologies, shaped by distinct economic structures and labor allocation patterns. In agriculture-dominant villages (characterized by majority engagement in farming alongside partial non-agricultural labor participation), the policy generates significantly positive outcomes. Enhanced tenure security through land titling reduces perceived land rights uncertainties for potential migrant laborers, thereby amplifying labor mobility intentions while preserving agricultural productivity. Non-agriculture-dominant villages demonstrate minimal responsiveness, as over 50% of laborers have transitioned to urban sectors or local non-farm industries. This diminished agricultural dependence renders land certification less relevant for livelihood strategies.

2. Based on the perspective of intergenerational differences.

Following the classification criteria established by Yin et al. [43], this study categorizes labor forces into three age groups: young laborers (average age ≤ 40 years), middle-aged laborers (41–60 years), and elderly laborers (>60 years). As shown in

Table 11. Heterogeneity analysis 2.

	Intergenerational Differences
	Young Laborers	Middle-Aged Laborers	Elderly Laborers
Land titling	0.025 (0.033)	0.021 ** (0.009)	−0.074 ** (0.035)
Control variable	Yes	Yes	Yes
Year fixed effect	Yes	Yes	Yes
Individual fixed effect	Yes	Yes	Yes
Observations	366	4252	158
Adjusted R-square	0.503	0.554	0.714

Notes: ** represent the significance level of 5%, respectively; the robust standard error of clustering at the household level is in parentheses.

, the implementation of land titling policy exerts differential impacts on rural laborer migration intentions across age cohorts. Specifically, the policy positively stimulates migration intentions among both young and middle-aged laborers, with statistically significant effects observed for the latter group. Conversely, it significantly suppresses migration intentions among elderly laborers. These findings can be attributed to the higher physical capacity and educational attainment of young and middle-aged laborers, who generally exhibit stronger preferences for urban employment in non-agricultural sectors to pursue higher incomes. In contrast, elderly laborers demonstrate deeper emotional attachment to farmland (“land affinity”) and greater agricultural experience [44]. The policy implementation reinforces their inclination toward agricultural production in rural areas rather than urban migration, reflecting the “leaving rural areas without abandoning farmland” phenomenon prevalent among elderly laborers.

4. Discussion

This study employs the CLDS database and a difference-in-differences (DID) model to empirically examine the impact of land titling on rural labor mobility intentions in China. The findings reveal that land titling significantly enhances rural laborers’ willingness to migrate, though this effect exhibits heterogeneity across age cohorts and villages with varying degrees of occupational diversification. Prior research predominantly focused on migration behaviors, non-agricultural employment patterns, and the urbanization status of migrant workers, leaving a critical gap in understanding the psychological determinants of mobility intentions. By addressing this gap, our work provides theoretical foundations and practical insights for optimizing resource allocation and promoting urban–rural integration.

Notably, while the mechanisms are explored through land reallocation and agricultural machinery investments, land titling likely influences mobility intentions through additional pathways beyond these two channels. Due to space constraints and data availability limitations, we prioritize these two mechanisms based on their substantial explanatory power and empirical salience. Future investigations should extend to broader implications of land titling, including its effects on food security, farmer welfare, and land scale operations. These dimensions represent crucial research frontiers for understanding the multifaceted impacts of rural land institutional reforms.

5. Conclusions

Enhancing the mobility of rural laborer not only plays a pivotal role in promoting urbanization and industrialization, but also exerts profound impacts on agriculture, rural areas, and farmers. Existing scholarship on land titling has predominantly focused on its tangible impacts on rural laborer mobility patterns and land market dynamics [45,46], often overlooking the nuanced psychological mechanisms shaping migration intentions. Utilizing balanced panel data from two waves of the China Labor-force Dynamics Survey (CLDS), this study employs a difference-in-differences (DID) model to empirically analyze the policy effects of land titling on rural laborers’ mobility intentions. The study finds that land titling significantly enhances rural laborers’ willingness to migrate, and this effect remains robust after controlling for individual fixed effects, year fixed effects, and household-level clustered robust standard errors. Under the land titling policy, the frequency of land reallocation within villages is likely reduced, which elevates farmers’ expectations of tenure stability and thereby strengthens their migration willingness. Furthermore, the implementation of land titling may encourage agricultural machinery investments, facilitating laborer substitution through mechanization and creating opportunities for laborer mobility. Further analysis reveals that the policy effects are more pronounced among middle-aged laborers and laborers in villages where agriculture dominates employment.

The research findings yield significant policy implications for optimizing rural land governance and fostering sustainable urbanization. This study provides empirical evidence and theoretical support for the government to promote the construction industry of new urbanization through this policy, and also provides reference for further deepening the reform of the land system in the future.

First, consolidating the positive externalities generated by farmland tenure confirmation remains imperative. By alleviating farmers’ apprehensions about land insecurity during urbanization transitions, policymakers can enhance the efficacy of new urbanization initiatives. Strengthening the legal enforceability of land contracting management rights certificates will provide robust safeguards for farmers’ legitimate land interests, thereby reducing land dispossession risks and amplifying tenure security’s role in facilitating rural laborer mobility. This institutional reinforcement aligns with China’s strategic objectives to maintain cultivable land above 120 million hectares while promoting agricultural modernization.

Second, prioritizing long-term land rights stability emerges as a cornerstone for sustainable rural development. The enhanced labor mobility under tenure confirmation primarily derives from farmers’ perceived reduction in land-related uncertainties, enabling dual income streams from off-farm employment and land rental markets. Frequent administrative land reallocations undermine this stability-driven effect by disrupting intertemporal decision-making, necessitating institutionalized mechanisms for contractual duration guarantees and transparent renewal procedures. Such measures would preserve the integrity of tenure security’s motivational impacts on labor reallocation.

Third, accelerating agricultural mechanization through targeted technological interventions proves critical for structural transformation. Machinery substitution effects not only optimize factor allocation, but also catalyze service-scale agriculture development through labor intensity reduction and operational efficiency gains. Grain-producing villages particularly require strategic investments in smart farming technologies and mechanized cooperatives to overcome scale constraints, a policy direction consistent with China’s rural revitalization priorities. Concurrent modernization of complementary infrastructure and digital extension services could amplify these productivity enhancements.

Lastly, the heterogeneous policy effects of tenure confirmation across village characteristics and individual attributes necessitate differentiated strategies. For villages with substantial non-agricultural employment, infrastructure development and rural specialty industry cultivation should be prioritized. Concurrently, enhancing vocational training and entrepreneurial support for young laborers will better align human capital with evolving rural economic structures. These tailored approaches will optimize policy synergies while balancing stakeholder interests across diverse socioeconomic contexts.