Rural Property Rights, Migration, and Welfare in Developing Countries

1 Introduction

It is well-known that the quality of rural property rights plays a certain role in both agricultural and urban development in developing countries. ^[1] This can be attributed to the facts that rural property rights are closely related to rural-urban migration and that rural-urban migration can reduce rural poverty and stimulate the urbanization process.

Nevertheless, how rural property rights influence rural-urban migration is empirically controversial. Conventional wisdom contends that strong rural property rights will increase peasants’ incomes and thus tie them to rural areas, because there exists a positive relationship between rural property rights and agricultural productivity (or the peasants’ incentive to conduct land investment) and exists a negative relationship between agricultural productivity and rural migration. These two relationships have been explored by one strand of empirical studies like Besley (1995), Kabubo-Mariara (2003), Goldstein and Udry (2005), Markussen (2008), and Giles and Mu (2012), etc. However, another strand of empirical literature (e.g., Field 2007; Mullan, Grosjean, and Kontoleon 2011; Chernina, Dower, and Markevich 2013) contends that the protection of rural property rights stimulates rural-urban migration since strong rural property rights mean less expropriation. It is not hard to find that the first strand of literature mainly highlights the positive impact of strengthened rural property rights on increasing peasants’ incomes, while the second strand of literature basically emphasizes the positive impact of secured rural property rights on reducing the peasants’ migration cost. Can these two strands of empirical results be integrated into a unified theoretical framework? Do strengthened rural property rights encourage rural-urban migration, reduce urban unemployment, and raise national welfare? We cannot find answers to these questions from the existing theoretical studies. This inspires us to conduct the present study.

In order to fill the theoretical research gap, we establish two-sector general equilibrium models with a dual economy to investigate how strengthened rural property rights influence rural-urban migration, urban unemployment, and national welfare. The protection of rural property rights in our paper can influence both rural productivity and land rent expropriation, which is similar to Fergusson (2013), but our setting is more general. That is, our setting grasps the essence and is also in accord with the existing two strands of empirical literature. In the basic model where the urban wage rate is exogenously given, the relation between strengthened rural property rights and rural-urban migration, as well as national welfare, is determined by comparisons of two effects exerted by the improvement of rural property rights, i.e., the rent-gaining effect and the productivity-enhancing effect. The former effect reduces the rural-urban migration cost, and the latter effect raises the rural wage rate. If the productivity-enhancing effect dominates the rent-gaining effect, strengthened rural property rights will reduce rural-urban migration and urban unemployment, but raise national welfare. However, if the opposite occurs, strengthened rural property rights will increase rural-urban migration and urban unemployment, but possibly reduce national welfare. Our findings on the relation between strengthened rural property rights and rural-urban migration theoretically support the validity of the aforementioned two strands of empirical literature. What’s more, our results concerning the relation between the quality of rural property rights and national welfare are also in line with Harris and Todaro (1970), and Todaro’s (1997) important idea that agricultural development is crucial for economic development and national welfare. When we extend the basic model by considering the endogenously determined urban minimum wage rate, the urban minimum wage determination mechanism is of great importance in determining the outcomes of the basic model. When we extend the basic model by introducing an urban informal sector, the main findings are almost the same as those in the basic model, but the value of labor’s marginal product of the urban informal sector plays a role in determining the impact of strengthened rural property rights on national welfare. In a word, in the extended models, the economic mechanisms behind our results are much more complex than those in the basic model.

It is worth noting that Fergusson (2013) is based on an agricultural economy with rural-urban migration satisfying Lewis’ equilibrium condition, which indicates that a typical peasant will continue to move to an urban area until the income he gets from rural production is equal to that he obtains from urban production, and meanwhile there is no urban unemployment (see Lewis 1954). In Fergusson’s (2013) model, rural property rights are endogenously determined by the rural elite and the exogenously determined urban wage rate plays an important role in determining the peasants’ migration decision. But in our models, we adopt the modified Harris–Todaro migration equilibrium condition. The benefits of our method are that we can take the whole economy containing both rural and urban areas into account, and at the same time accommodate urban unemployment in our model. Originally, the Harris–Todaro migration equilibrium condition shows that a typical peasant will not stop migrating until the wage rate he gets from the rural production is equal to the expected wage rate (i.e., the probability of getting a job multiplying the corresponding wage rate) he obtains in an urban area (see Harris and Todaro 1970). In our paper, we argue that a typical peasant will stop moving to an urban area when the expected income (e.g., wage, land revenue and expropriated land revenue from other migrants) he obtains in a rural area is equal to the corresponding expected income (e.g., expected wage rate and unexpropriated land revenue) he gets in an urban area. We call this the modified Harris–Todaro rural-urban migration equilibrium condition. On the basis of such a condition, we find that the endogenously determined urban wage rate matters for the rural people’s migration decision.

Another point worthy of mention is the relation between strengthened rural property rights and urban unemployment. In the basic and extended models, we argue that under some conditions strengthened rural property rights can effectively reduce urban unemployment, which more or less indicates that the solution to reduce urban unemployment lies in the rural sector and the development of rural areas. Our results are parallel to existing findings obtained in the Harris–Todaro framework. Beladi and Naqvi (1988) use a two-sector model to find that the improvement in the agricultural production technique can reduce the urban unemployment rate. Gupta (1997b) constructs a three-sector model to show that the rise of the agricultural output price can decrease the urban unemployment rate. Chaudhuri (2000) builds a three-sector model to demonstrate that wage and price subsidies to the rural sector can effectively decrease urban unemployment. Yabuuchi and Beladi (2001) construct a different three-sector model from Chaudhuri (2000) to argue that the wage subsidy to the rural sector will reduce urban unemployment. Chaudhuri and Banerjee (2010) contend that FDI on the agricultural land can decrease urban unemployment through constructing a three-sector model. In addition to these rural development polices, other development polices (e.g., wage and price subsidies to urban sectors, tariff reductions on manufacturing products, international factor mobility policies) can also unambiguously or conditionally reduce urban unemployment in the presence or absence of the Harris–Todaro framework (see Basu 1980; Gupta 1993, 1994, 1995b; Chaudhuri 2006; Chaudhuri, Yabuuchi, and Mukhopadhyay 2006; Chaudhuri 2007; Chaudhuri and Yabuuchi 2010; Gupta and Dutta 2011; Pi and Zhou 2012). Our result concerning urban unemployment complements these findings.

This paper also contributes to the existing literature in other two respects. The first and main aspect is anchored in the study of rural-urban migration. Since the publication of Harris and Todaro’s (1970) classic paper, the Harris–Todaro model has been widely used by many development economists. As this model has very good policy implications, numerous theoretical studies have investigated the impacts exerted by development policies, trade policies, and production factors accumulation in such a framework. ^[2] Here we just briefly introduce some of them. For example, Stiglitz (1974), Bhagwati and Srinivasan (1974), Calvo (1978), Basu (1980), Beladi and Naqvi (1988), Gupta (1995a, 1995b), Beladi and Marjit (1996), and Zenou (2011) explore the economic and social impacts of development policies (e.g., price and wage subsidies, tariff reductions and export promoting policies, international factor mobility and tax policies) and the formation of labor unions. By adding an informal sector to the economy, Fields (1975), Quibria (1988), Grinols (1991), Gupta (1993, 1997b), Chandra and Khan (1993), Chaudhuri (2000), Chaudhuri, Yabuuchi, and Mukhopadhyay (2006) investigate the economic and welfare effects of development policies (e.g., price and wage subsidies, tariff reductions and export promoting polies, international factor mobility and tax polices). Considering distortions of capital markets (i.e., segmentations of urban and rural formal and informal capital markets), Khan and Naqvi (1983), Chao and Yu (1992), and Gupta (1993, 1997b) investigate how reductions of capital market distortions and other development policies influence the whole economy and social welfare. By dividing labor into skilled and unskilled types, Chaudhuri and Yabuuchi (2007), Beladi, Chaudhuri, and Yabuuchi (2008), Beladi, Chakrabarti, and Marjit (2010), Chaudhuri (2008), Pi and Zhou (2012, 2014), and Pan and Zhou (2013) study how development policies and other factors (e.g., public infrastructure provisions, global production, labor market reforms) influence the skilled-unskilled wage gap, urban unemployment and social welfare. Some scholars also reconsider the socioeconomic impacts of development policies and economic growth by incorporating the land market, the special economic zone (e.g., the free-duty zone), and the dualism of agricultural sectors into the Harris–Todaro framework (e.g., Miyagiwa 1993; Gupta 1997a; Krichel and Levine 1999; Yabuuchi and Beladi 2001; Chaudhuri 2006; Chaudhuri and Banerjee 2010; Chaudhuri and Yabuuchi 2010). ^[3] However, just as mentioned above, although empirical studies have emphasized the close relation between rural property rights and rural-urban migration, such a relation is rarely explored within the Harris–Todaro framework. To the best of our knowledge, this paper is the first one to incorporate rural property rights into the Harris-Todaro framework.

The second aspect mainly rests on our theoretical methodology. When the issue related to rural property rights is addressed, the game-theoretical or contractual approach is commonly employed (see, e.g., Baker 2003; Tassel 2004). However, the game-theoretical or contractual approach usually treats the factor prices as given and ignores the interactions among markets to some degree. Such limitations of partial equilibrium analysis cannot tell us the whole story when we discuss the impacts of improved rural property rights. Just as Acemoglu (2010, p. 22) puts it, “Depending on magnitudes of various effects, general equilibrium interactions can offset or even reverse sensible partial equilibrium conclusions …. Economic theory nonetheless provides some guidance in assessing the importance of general equilibrium effects.” To fill the gap, our analytical approach highlights the market interactions and develops full scenarios for the assumed economy. We establish a unified and open theoretical framework with a general equilibrium approach to investigate the issue concerning the quality of rural property rights.

The rest of this paper is organized as follows. The basic theoretical model is established in Section 2. We analyze how the quality of rural property rights influences rural-urban migration and national welfare in Section 3. In Section 4, we extend the basic model and examine the validity of findings obtained in Section 3. Section 5 provides some concluding remarks.

2 The theoretical model

Consider a small open dual economy consisting of two sectors, the urban sector and the rural sector. The outputs of these two sectors are internationally tradable. The urban sector uses labor and capital as factors of production, and the rural sector uses labor and land. Capital and land are sector specific to the urban and rural sectors, respectively. In urban areas, workers are often protected by minimum wage laws or receive a contractual wage rate. If the minimum wage is high enough, urban unemployment will persist (e.g., Calvo 1978; Gupta 1997a; Zenou 2011). In this section, we assume that the urban minimum wage rate is downwards rigid and exogenously given. This assumption will be relaxed in Section 4. In rural areas, labor can be fully employed due to the rural flexible wage rate. It is worth noting that labor in urban areas is composed of the urban initial labor force and the rural-urban migrants. We further assume that all the goods and factor markets are perfectly competitive, and the output of the rural sector is numeraire.

The urban sector’s production function is described by:

where M is the output of the urban sector. LM and KˉM are labor and capital used by the urban sector, respectively. FM is the urban production function: it is strictly concave and linearly homogenous.

The profit maximization conditions of the urban sector yield:

where pM is the output price of the urban sector, wˉ is the urban minimum wage rate, r is the interest rate of capital, and Fji=∂Fi/∂j (i=M, and j=L,K).

In this paper, the key variable is the quality of rural property rights. Following Fergusson (2013), the quality of rural property rights is denoted by a parameter μ, where μ∈[0,1]. When μ=0, there is no rural property rights protection at all. When μ=1, the rural property rights protection is complete. The larger the value of μ, the better the quality of rural property rights. In this paper, we focus our attention on the case where rural property rights are incomplete, that is, μ∈(0,1). In accord with the above-mentioned empirical literature, the quality of rural property rights has two impacts, one impact on the rural productivity and the other impact on land rent expropriation. These two impacts will be demonstrated in details as below.

First, let us take a close look at the relation between the quality of rural property rights and rural production. In our assumed economy, rural production not only relies on the production inputs but also on the quality of rural property rights. μ influences rural production by way of changing the rural productivity. The larger the value of μ, the higher the rural productivity. Therefore, the rural sector can produce more output if rural property rights are better protected.

The production activities are described by:

where R is the output of the urban and rural sector. LR and Tˉ are labor and land used by the rural sector, respectively. FR is the rural production function: it is strictly concave and linearly homogenous with respect to production factors. A represents rural productivity, which is a function of the quality of rural property rights (i.e., μ). A(μ) is a strictly concave function, indicating that strengthened rural property rights improve rural productivity at a decreasing marginal ratio. We further assume that A(0)=0 and A(1)=1. Our setting of A(μ) follows the spirit of Fergusson (2013), but here our expression (i.e., eq. [4]) is more general in the sense that we do not explicitly show the relation between the quality of rural property rights and rural production factors. The causal mechanisms behind A(μ) have been considered by Feder and Noronha (1987), Feder and Feeny (1991), and Migot-Adholla et al. (1991). Besides, eq. [4] also captures the essence of the first strand of the empirical literature, that is, better protection of rural property rights indicates higher rural productivity.

The profit maximization conditions of the rural sector yield:

where wR is the flexible rural wage rate, τ is the rental rate of land, and Fji=∂Fi/∂j (i=R, and j=L,T).

We now consider how the quality of rural property rights influences land rent expropriation. Our setting of the relationship between rural property rights and land rent expropriation also follows the spirit of Fergusson (2013), but we incorporate the relationship into the Harris–Todaro rural-urban migration equilibrium condition. All of the details will be shown below.

We assume that each rural household is endowed with one unit of labor. For simplicity, each rural household is represented by one peasant, either working in agricultural production or migrating to an urban area. The initial rural labor endowment of all the rural households is denoted as LˉR. The number of rural-urban migrants is denoted as LTR. Furthermore, we assume that LUN is the number of urban unemployed and LˉU is the initial urban labor endowment. Different from Fergusson (2013), who assumes that the rural elite owns most of the land and each peasant has the same amount of land, here we suppose that all the land belongs to rural households, and they equally own a certain amount of land. Specifically, each family has the same land endowment Tˉ/LˉR. Although such an assumption is realistic in some developing countries (e.g., China and Vietnam) and unrealistic in others (e.g., some Latin American countries), the focus of our paper is on migration and rural property rights instead of the political economy of rural property rights, and such a treatment helps us to capture the main mechanisms in the purest way. Our assumption is beneficial to simplify analyses and concentrate on the main effects. With the process of urbanization and industrialization, some rural workers will transfer from rural areas to urban areas. If a peasant chooses to stay in a rural area, he can not only get his labor income wR, but also get his own land rent income τ(Tˉ/LˉR). In addition, due to weak rural property rights, the peasant staying in a rural area can also get a chance to expropriate other migrant families’ land rent income (i.e., τ(1−μ)(Tˉ/LˉR)LTR). We assume that each peasant has equal probability to expropriate the other migrants’ land rent incomes and hence he can get τ(1−μ)(Tˉ/LˉR)(LTR/LR). Thus, the expected income of a typical peasant staying in a rural area is

But if the peasant moves to an urban area, he can earn an expected wage rate (i.e., (LM/(LM+LUN))wˉ) and an unexpropriated land rent income (i.e., τμ(Tˉ/LˉR)). ^[4] Therefore, if the peasant migrates to an urban area, he can earn

In the spirit of Harris and Todaro (1970), the rural-urban migration equilibrium condition in our paper can be described as follows: a typical peasant will stop moving to an urban area when the expected income of staying in a rural area (i.e., wR+τ(Tˉ/LˉR)+τ(1−μ)(Tˉ/LˉR)(LTR/LR)) is equal to that of working in an urban area (i.e., (LM/(LM+LUN))wˉ+τμ(Tˉ/LˉR)). Here, we call it the modified Harris–Todaro rural-urban migration equilibrium condition. Formally, it can be described as:

In Fergusson (2013), rural-urban migration satisfies the Lewis’ equilibrium condition. That is, a peasant will stop migrating when the income he obtains from agricultural production is equal to that he earns in an urban area. There is no unemployment in urban areas and one can get a job in urban areas for sure if he chooses to migrate. However, in our paper, the rural-urban migration equilibrium satisfies the modified Harris–Todaro equilibrium condition. This is one of the most significant differences between this paper and Fergusson (2013).

To close the model, the labor market-clearing condition is stated as: ^[5]

In our model, capital and land are sector-specific factors to the urban and rural sectors. For the sake of simplicity, we do not demonstrate the market clearing conditions for the capital and land markets.

So far, the theoretical model has been built. Equations [1]–[9] determine nine endogenous variables, M, R, LM, LUN, LR, LTR, r, wR, and τ. μ is the exogenous variable. Others are parameters.

3 The general equilibrium analysis

Before conducting the comparative static analysis, firstly we would like to totally differentiate eqs [5]–[7]. This step will strengthen our understanding of how the quality of rural property rights influences rural-urban migration.

By totally differentiating eqs [5]–[7] and combining with the help of the remaining equations, we can get:

where Fjli=∂2Fi/∂j∂l (i=M,R, j=L,K,T, and l=L,K,T), and A′=dA/dμ.

The established economic system can be reduced to a subsystem, namely eqs [10]–[12], which explicitly show how the protection of rural property rights influences rural-urban migration. On the one hand, according to eqs [10] and [11], the improvement of rural property rights raises the marginal productivities of labor and land (i.e., A′FLR and A′FTR), therefore influencing rural-urban migration. On the other hand, from the migration equilibrium condition (i.e., eq. [12]), the protection of rural property rights also exerts an impact on the migration decision by way of changing the migration benefit in terms of the potential land rent income (i.e., τ(Tˉ/LˉR)(LTR/LR)+τ(Tˉ/LˉR)).

It is not hard to get the following equations system:

The determinant of the coefficient matrix of the above equations system is denoted as Δ. It is easy to find that

In order to intuitively understand the impacts of the quality of rural property rights, we try to demonstrate the propositions and corollaries without using mathematical notation in this section and the following section. However, when providing the economic intuition behind these propositions and corollaries, we use the simplest mathematical notation in order to avoid unnecessary misunderstandings.

Proposition 1 is established to show how the quality of rural property rights influences rural-urban migration.

Proposition 1 implies that the quality of rural property rights exerts an ambiguous impact on rural-urban migration. On the one hand, strong rural property rights will raise the migrant’s land rent income (i.e., τ(Tˉ/LˉR)) and at the same time decrease land rent expropriation (i.e., τ(Tˉ/LˉR)(LTR/LR)). This will stimulate a peasant to migrate to an urban area (i.e., (1/Δ)(τ(T¯/L¯R)(LTR/LR)+τ(T¯/L¯R)). We call this effect (i.e., τ(Tˉ/LˉR)(LTR/LR)+τ(Tˉ/LˉR)) the rent-gaining effect. On the other hand, strong rural property rights will also raise the peasant’s income in a rural area. This is because strengthened rural property rights will not only raise the marginal product of labor (i.e., A′FLR), but will also increase the net land rent income of a peasant staying in a rural area (i.e., ((1−μ)Tˉ/LR)FTR). This will cause a peasant to remain in a rural area (i.e., −(1/Δ)A′(FLR+(((1−μ)Tˉ)/LR)FTR)<0). We call this effect (i.e., A′(FLR+(((1−μ)Tˉ)/LR)FTR)) the productivity-enhancing effect. The final migration decision depends on which effect dominates. If the former effect dominates the latter, strengthened rural property rights will increase rural-urban migration. However, if the former effect is dominated by the latter, the opposite will occur.

Our theoretical results support the findings of both the first and the second strands of empirical literature. If the protection of rural property rights only means the protection of agricultural output and the rise of the rural productivity, that is, there is no τ(Tˉ/LˉR)(LTR/LR)+τ(Tˉ/LˉR), then strengthened rural property rights will tie peasants to rural areas, which supports the results obtained by the first strand of empirical literature (e.g., Kabubo-Mariara 2003; Giles and Mu 2012). However, if the protection of rural property rights only results in less land rent expropriation, that is, A′=0, then strengthened rural property rights will increase rural-urban migration, which confirms the findings of the second strand of empirical literature (e.g., Field 2007; Mullan, Grosjean, and Kontoleon 2011; Chernina, Dower, and Markevich 2013).

Based on Proposition 1, we use Corollary 1 to summarize the impact of strengthened rural property rights on urban unemployment.

Combining Proposition 1 with Corollary 1, we see that changes in urban unemployment and rural-urban migration move in the same direction. That is to say, if strengthened rural property rights stimulate peasants to migrate to urban areas, increased migration will unambiguously raise urban unemployment. But if strengthened rural property rights tie peasants to agricultural activities, the number of rural-urban migrants will decline and urban unemployment will shrink. The economic explanation for above demonstrations can be attributed to the fact that the urban sector has a relatively constant scale. One may wonder whether the positive relation between the number of rural-urban migrants and the scale of urban unemployment can be empirically supported. The answer is yes. For example, Todaro (1997) and Potts (2000) empirically confirm that in many African countries, increased urban unemployment can be largely attributed to the growing number of rural-urban migrants.

Now, we can use Figure 1 to intuitively show the contents of Proposition 1 and Corollary 1. For simplicity, we only use straight lines to show the relations among variables. ^[6]LM is determined by eq. [2], which is independent of μ. Replacing wR and LUN in eq. [7] by eqs [5] and [8], we can get the relation between LTR and τ shown by eq. [7], which is depicted by Line a. Line b represents the relation between LTR and τ described by eq. [6]. Line c indicates the relation between LTR and LUN implied by eq. [8]. o is the original point. τ∗, LTR∗, and LUN∗ are set as the initial equilibrium values. Now, consider the case where the quality of rural property rights is improved, i.e., μ increases. By eq. [6], an increase in μ will make Line b move leftwards to b′. However, an increase of μ may either make Line a move leftwards or rightwards to a′. When

that is, the rent-gaining effect dominates the productivity-enhancing effect (denoted as Condition I), Line a will largely move leftwards. Combined with the movement of Line b, such a movement of Line a will lead to an increase in LTR. By Line c, the amount of LUN also increases. This demonstrates the first parts of Proposition 1 and Corollary 1. When

that is, the productivity-enhancing effect dominates the rent-gaining effect (denoted as Condition II), Line a will slightly move leftwards or move rightwards, ^[7] which leads to a reduction of LTR. By Line c, LUN decreases. This demonstrates the second parts of Proposition 1 and Corollary 1.

Figure 1:

Impacts of strengthened rural property rights on migration and unemployment.

Now, let us focus our attention on how the quality of rural property rights generates an impact on national welfare.

The balance of the national expenditure and income is stated as:

where E is the minimum expenditure function, and U is the social utility level which depicts national welfare. Investigating national welfare by considering the balance of the national expenditure and income is widely used by many development economists (e.g., Grinols 1991; Chandra and Khan 1993; Gupta 1997a).

Now we can obtain Proposition 2.

The economic mechanism behind Proposition 2 is as follows. On the one hand, strengthened rural property rights will directly increase rural production output (i.e., A′FR). On the other hand, it will also indirectly affect national welfare by exerting an impact on rural-urban migration. If strengthened rural property rights decrease rural-urban migration (i.e., dLTR/dμ<0), then from eq. [8] we know that more peasants will stay in rural areas, which definitely increases rural output (i.e., −AFLR(dLTR/dμ)>0). In this situation, strengthened rural property rights will increase national welfare. However, if strengthened rural property rights stimulate rural-urban migration to a certain degree (i.e., dLTR/dμ>A′FR/(AFLR)), that is, the increase in rural output (i.e., A′FR) is less than the loss of rural output due to increased rural-urban migration, then national welfare will decrease.

Proposition 2 implies that the change in national welfare is determined by the change in urban unemployment (i.e., rural employment). The reason behind this is that the urban sector has a fixed size in the basic model. Less urban unemployment (i.e., more rural employment) expands the rural sector, increasing agricultural output and national welfare. Proposition 2 highlights the role that the rural sector plays in economic development. With the help of such a simple model, our argument is parallel to Jorgenson (1967) and Harris and Todaro (1970) who also contend that rural development is of great importance for economic development in a dual economy.

We should point out that such a finding highly depends on the fixed size of the urban sector. However, when the urban sector size is flexible, the impacts of the quality of rural property rights on national welfare are more complex, which will be explored in Section 4.

4 Extension and discussion

Section 4 extends Sections 2 and 3. We will make the following two extensions. The first one is to relax the assumption of the urban exogenous minimum wage rate by endogenizing it. Such an extension is made to further investigate the role that the urban sector plays. The second one is to introduce an urban informal sector to the basic model, since such a sector plays a crucial role in economic development of developing countries. ^[8]

5 Concluding remarks

This paper provides a unified theoretical framework and employs the general equilibrium approach to investigate the impacts of the quality of rural property rights on rural-urban migration and national welfare in developing countries. We pay much attention to the empirical debate concerning how the quality of rural property rights influences the rural-urban migrants. In the basic model where the urban wage rate is exogenously given, the impacts of strengthened rural property rights on rural-urban migration and national welfare are determined by the rent-gaining effect and the productivity-enhancing effect. Our theoretical results are in line with the existing empirical literature. For the sake of generality and completeness, we extend the basic model. When we extend the basic model by considering the endogenously determined urban minimum wage rate, the urban minimum wage determination mechanism is also of great importance in determining the findings obtained by the basic model. When we extend the basic model by introducing an urban informal sector, the value of labor’s marginal product of the urban informal sector is crucial for determining the impact of strengthened rural property rights on national welfare.

Future studies can extend our analysis as follows. Firstly, the urban informal sector could be vertically integrated with the urban formal sector. However, this paper does not take it into account. Secondly, the global increase in skilled-unskilled wage inequality in developing countries has captured the attention of many economists. Will strengthened rural property right alleviate or worsen the wage inequality? Thirdly, the studies of rural property rights are mostly analyzed by the empirical or game-theoretical or contractual approach. Much theoretical work could be done through the employment of the general equilibrium approach. Fourthly, this paper assumes that when a peasant migrates, his family’s land rent may be expropriated. But in some situations, the family members can effectively protect their land from expropriation by changing their migration choices. ^[14] Hence, relaxing our assumption by considering changing the choice of migration decisions is worthy of being investigated in the future research. Fifthly, it may be important to relate the structural transformation within a country (i.e., contraction of the rural sector and the expansion of the urban sector) to the wage convergence between the two sectors.