Household Registration, Land Property Rights, and Differences in Migrants’ Settlement Intentions—A Regression Analysis in the Pearl River Delta

Abstract

Objectives: Few studies have examined the impacts of structural differences in the urban–rural dichotomy under the new household registration policy on migration and settlement behavior. Nevertheless, the rationale for the settlement policy of local governments should be further elucidated and improved. This study aims to analyze the household registration, land property rights, and differences in migrants’ settlement intentions. Methods: This study used migration survey data from the Pearl River Delta and probit regression to fill this gap in the literature. Findings: Because of the long-term effects of the household registration system and their socioeconomic differences, urban-urban migrants and rural–urban migrants differed in their settlement intentions. Furthermore, the new points-based household registration system affected migrants’ settlement intentions. Relative to the rural–urban migrants, urban–urban migrants more easily met the settlement requirements under the points-based system, and they tended to settle in their current cities. By contrast, migrants with farmland in their hometowns tended to settle there. The findings underscore the relevance of adopting perspectives that consider the urban–rural dichotomy and related structural differences to understand migrants’ settlement intentions in China.

1. Introduction

Since the Chinese economic reform, urbanization and industrialization in China have experienced considerable growth. Because the urban–rural differences in China are substantial, megalopolises in various regions, including the Pearl River Delta, encounter large influxes of migrants. China’s floating population is defined as migrants without local household registration (hukou) status and is one of the largest migrant populations in the world. According to the Seventh National Population Census of the People’s Republic of China, domestic migrants in 2020 totaled 376 million people. Large-scale migration causes both theoretical and practical issues. In particular, researchers have long discussed the problem of migrant settlement under the urban–rural dichotomy [1,2]. Structural differences in China’s urban–rural dichotomy comprise two aspects, namely household registration and the land system. Compared with locals, migrants cannot enjoy social welfare benefits, and permanent settlement is difficult. As for land use, although farmland guarantees farmers’ livelihoods, overattachment to one’s farmland renders permanent settlement in other regions more difficult.

Because most migrants in China come from farming villages, researchers have centered on the settlement intentions of rural–urban migrants. Because of agricultural laborers’ identities as farmers and outsiders, permanent urban settlement is rare and many continually migrate from city to city or between cities and their hometowns [3]. The settlement intention discussed in this paper refers to the willingness of the floating population to settle permanently in the immigration place in order to obtain the welfare of the immigration place. Migrants’ intentions to settle in cities are restricted not only by the household registration system but also by conditions such as their concentration in the secondary labor market, characterized by labor-intensive industry, unstable jobs, and low social security [4]. The literature centers on rural–urban settlement intention and its determinants, including individual, generational, social, market, and institutional factors. Few studies have compared the settlement intentions of rural–urban and urban–urban migrants, whose household registrations are in farming villages and cities, respectively [4,5,6,7,8,9,10]. The upward trend in urban–urban migration in China is expected to continue. According to data from a monitoring survey of floating population dynamics, urban–urban migrants accounted for 13.95% and 16.7% of the total migrant population in 2011 and 2018, respectively. However, due to the long-term impact of China’s urban–rural dual structure, there are differences between China’s rural migrants and urban migrants in many aspects, such as individual characteristics, family background, and migration behavior [2]. The influences of these differences on settlement intention merit further exploration.

The urban–rural dual classification system is no longer enough to reflect the reality of settlement evolution in many countries, including developing countries. In recent decades, the important feature of the evolution of human settlements is that the boundary between urban and rural areas is becoming increasingly blurred. In developed countries, urban residents’ reliance on convenient transportation networks to work in urban areas and to live in the suburbs has long been common. This phenomenon is the concrete embodiment of the increasingly blurred boundary between urban and rural areas, and the disappearance of the urban–rural dual pattern. At the same time, the differences between urban and rural settlements, and lifestyles in developed countries tend to be blurred, and the traditional urban–rural, dual-division model does not easily reflect the reality. In the context of China’s urbanization, many studies show that China’s urban–rural boundaries are becoming weakened or are even disappearing, and the lifestyle gap between urban and rural areas is becoming smaller and smaller. However, the path dependence of the impact of urban–rural differences in China may still exist.

China’s first-tier and second-tier cities employ a points-based household registration system through which they obtain the skilled individuals they require for development [11]. The general consensus is that rural–urban migrants wish to settle in cities with their families. This is difficult to achieve because of relevant regulations in the existing household registration system. The elimination of the institutional barriers caused by this system is expected to promote permanent settlement in cities [12,13,14]. Other studies have indicated the opposite—that the percentage of rural–urban migrants who wish to transfer their household registration to cities and permanently settle there is low. The current percentage of migrants who wish to do so with no other conditions is 21.8%. By contrast, only 11.8% of migrants wish to do so, even under the condition that their land ownership contract must be returned to the government. This discrepancy may be related to the deep-seated institutional and cultural differences between cities and countryside [15]. Therefore, further exploration of the long-term impacts of land property rights, welfare, and culture in China under the urban–rural dichotomy of the household registration system is essential.

In this paper, we use survey data to discuss the following questions: (1) Under the urban–rural dichotomy, how do the demographic characteristics and settlement intentions of rural–urban migrants and urban–urban migrants differ? (2) Which factors affect the migrants’ settlement intentions? Specifically, how do the new points-based household registration system and land system in the Pearl River Delta megalopolis affect settlement intentions? (3) Which factors affect the settlement intentions of rural–urban migrants and urban–urban migrants? The findings advance the understanding of floating population dynamics and urban–rural differences under the new policies contributing to the dichotomous structure in China. The effectiveness of the current Chinese household registration system reform and of the related policy are also examined.

2. Economic, Cultural, and Institution Factors Influencing Settlement Intention

Neoclassical economics theory holds that migration is a rational decision. Regardless of whether they decide to settle in cities or return to the countryside, migrants seek to maximize their own interests [16,17,18,19]. Market factors such as salary, job stability, job opportunities, and occupation types all affect migrants’ settlement intentions. Demand for skilled workers increases faster than does demand for unskilled workers, and that job stability affects the floating population [20]. In the United States, migrants with university degrees are more likely to return to their countries of birth than are migrants with low education levels. However, economic opportunities increased the possibility that educated people would elect to stay in the United States [21]. Notably, higher educational attainment grants migrants more favorable opportunities in China than in the United States [22]. Elderly migrants from Mexico to the Unites States are more likely to return to their home country. The longer migrants stay in their destination countries, the less likely it is for them to return to their countries of origin [23]. Overall, relatively more studies have addressed the market factor as a fundamental motivation for population settlement. To settle in a city first requires the economic capability to live in one, which makes migrants fully consider the market factor in their decision making [2,24].

The neoclassical perspective considers migrants’ material demands in determining factors influencing their settlement intentions but overlooks their psychological needs. Migrants leave their hometown for an unfamiliar one, such as a megapolis, to make a living, seeking to habituate themselves to its rapid tempo, and longing for acknowledgment. Some researchers have begun to study how social culture affects migrants’ settlement intentions. Migrant networks and social relations may also critically affect decision making in this regard. The lack of a migrant network may result in migrants returning to their hometowns [16,25,26]. As migrants stay longer at the inflow areas and adapt to their new environments, their social status improves, and their settlement intentions strengthen. Some studies have taken the social adaptation perspective to discuss indicators, such as living environment, emotional ties, environmental perception, social integration, and city assessment on settlement intention, observing that their reinforcement can increase the settlement intentions of new-generation migrants [25,26,27].

Theoretical research typically emphasizes economic and sociocultural factors, and overlooks systemic factors. In discussing determinants of migrants’ settlement intentions in China, an emerging economy, institutional factors must also be considered [28,29]. Since the economic reform, industrialization and urbanization in China have accelerated, gradually revealing regional differences. People began migrating to, and settling in, coastal cities characterized by favorable economic development affording abundant opportunities. In other countries, the typical migration process is that migrants permanently settle in a city after they have lived and worked there for some time. [30,31]. In China, however, many migrants are part of the floating population, residing in a certain location for a time before moving to the next, without ever settling. This aspect of migration in China is similar to that of illegal immigrants in the United States and migrant workers in Germany, where immigration and settlement are restricted [32].

Under Chinese government policy, household registrations are classified as urban and rural. To a certain extent, this prevents people with rural registration status from enjoying the various benefits of urban welfare programs. These benefits are generated from cities’ socioeconomic development, which result in high-quality education, convenient transportation, and plentiful job opportunities. The household registration system maintains a chasm between the opportunities of migrants and locals, which constitutes the fundamental reason migrants choose to go in circles, moving from city to city, rather than settling permanently anywhere. Migrants face various problems, including low incomes, poor quality of life, unstable work, and limited social welfare, which causes them to be easily rejected and marginalized by urban society [32]. Consequently, many migrants return to their hometowns at times they deem suitable or circle between the city and the countryside [30,33,34].

The background provided thus far explains why Chinese researchers believe that the household registration system is the most pivotal deterrent to migrant settlement in cities. Some studies have asserted that eliminating the household registration system or granting rural migrants urban registration status will lead to settlement in inflow areas [13,14,35,36,37]. Some studies have discussed how to use reforms in the household registration system to control population settlement, given that registration restriction can explain settlement intention to a certain extent [13,14,35,36,37]. However, as studies became more in depth, some studies indicated that, even if the rules of the household registration system were relaxed or even if the system were eliminated, not as many migrants as expected would choose to settle in cities. Whyte (2010) argued that, as rural–urban migration occurs in China, the urban–rural dichotomy caused by the household registration system would reappear in cities [1]. Although cities have long obtained resources, such as human capital, land, and agricultural by-products from the countryside, their high-quality public services are not accessible to rural–urban migrants. Since the economic reform, the urban–rural dichotomy has gradually become less stark, and some obstacles to the floating population have been gradually eliminated. However, city-centered reforms have also been executed at a faster pace. The government set higher standards for cities with regard to infrastructure and public involvement, resulting in a substantial rise in the complexity of welfare eligibility requirements for city residents. In the countryside, systems and infrastructure for education, health, medicine, and insurance lagged far behind. This long-established household registration system has created great urban–rural differences that may also correspond to differences in migrant settlement intention. Hypothesis 1 is presented as follows:

Hypothesis H1.

Owing to the long-term impacts of the urban–rural dichotomy in China, rural–urban migrants and urban–urban migrants differ in their socioeconomic and cultural characteristics and, therefore, also in their settlement intentions.

In recent years, China has been actively promoting various household registration policies in an effort to resolve the various problems caused by population migration under the urban–rural dichotomy of the household registration system. In early 2010, Zhongshan, located in Guangdong Province, became the first city to implement the Guidelines to the Points-Based Management of the Floating Population. Migrant workers meeting certain criteria may be given local registration status. In June 2010, the Guangdong Province government established the Guidelines for Admitting Migrant Workers into Urban Cities and Towns through the Points System (Trial), marking the province-wide launch of the points-based household registration system and, thereby, providing tens of thousands of migrants the opportunity to obtain local registration status.

The Chinese State Council’s 2014 publication “Opinions about further promoting hukou reform from the State Council” stated that they would fully lift the restrictions on settlement in towns and small cities, systematically lift the restrictions on settlement in medium-sized cities, reasonably determine the conditions for settlement in large cities, strictly control population size in megacities, and implement a points-based settlement system. However, the reform of the household registration system did not break the systemic barriers that constrain free two-way mobility between cities and the countryside. Specifically, the establishment of the points-based household registration system erected a hidden barrier to urban settlement for farming populations. The system may continue to enlarge the urban–rural divide [11]. In the long term, the household registration system provides more favorable conditions for urban–urban migrants than for rural–urban migrants, facilitating their fulfillment of the requirements in the points-based system. Therefore, the existing points-based household registration system may generate a new urban–rural dichotomy between urban–urban migrants and rural–urban migrants. Hypothesis 2 is presented as follows.

Hypothesis H2.

The points-based household registration system affects migrants’ settlement intentions. Compared with rural–urban migrants, urban–urban migrants more easily meet the criteria and are more likely to settle in their current cities.

Another critical aspect of the urban–rural dichotomy is the land system. Although all land in China is under public ownership, urban land belongs to the country, whereas most land in farming villages is collectively owned. Control of the value and use of urban land differs from that of rural land. The Notice by the General Office of the State Council on the Strict Implementation of Laws and Policies on Collectively Owned Rural Construction Land issued in December 2007 mandated that residential land in a farming village can only be distributed among residents of that village; town residents may not purchase land as building foundations for personal homes (hereinafter referred to as house foundation land), farmers’ houses, or illegally constructed houses without the proper governmental certificate in farming villages. This dichotomized land system guarantees land as farmers’ livelihoods but inevitably results in farmers’ overattachment to their land. This is a possible reason for differences in settlement intentions between rural–urban migrants and urban–urban migrants.

A major current problem is that migrants from farming villages must give up their farmland in exchange for obtaining urban registration status. Therefore, gradual household registration reform must be based on a certain signal mechanism. Regarding local policy implementation, the points-based system reform implemented in various regions is primarily based on migrants’ income, education, years lived in the city, and social security insurance payments [2,11]. Migrants with medium-to-low socioeconomic status are overlooked, as are self-employed individuals with substantial incomes but who do not have social security insurance, and floating migrants willing to relinquish their farmland for urban registration status (which may free up more farmland and house foundation land for farming village residents). Therefore, to study the problem of migrant settlement in cities, whether people are willing to surrender their land for urban registration status should also be considered.

In the Third Plenary Session of the 18th Communist Party of China Central Committee, a mortgage guarantee scheme conferring rights to contract management and rural housing property was proposed, marking a great step in land system reform in farming villages. Land system reform is related to whether rural–urban migrants can settle in cities by obtaining benefits from their land in their hometowns. It is also related to whether city governments can use migrants’ land in the farming villages (or land indicators) to obtain certain economic resources with which they can provide public services facilitating migrant settlement. However, studies have stated that, owing to factors such as risk avoidance of unemployment in the city, increases in farming village land value, farmland contracting rights, rights of use concerning house foundation land, and the traditional belief that one’s land is one’s life, migrants are often unwilling to surrender their land [33,38,39]. Therefore, hypothesis 3 is presented.

Hypothesis H3.

Rural–urban migrants are more inclined to return to, and settle in, their hometowns if they own house foundation land and farmland there. Migrants who are willing to relinquish rights to their farmland in their hometowns in order to settle in the city are more inclined to settle in the destination city.

3. Variables and Methods

3.1. Research Area and Survey Data

The data for this study were derived from a joint project survey involving quantitative questionnaires and qualitative interviews conducted in the Pearl River Delta Metropolitan Region in Guangdong Province from May to October, 2017. We conducted a survey in six cities in the core area of the Pearl River Delta Metropolitan Region, and included scholars and graduate students from various universities and research institutes. According to the size of the city population, net inflow of population, and operability of the survey, the research group roughly allocated the number of questionnaires for each city, as follows: Guangzhou, 560; Dongguan, 499; Foshan, 384; Zhongshan, 317; Zhuhai, 293; and Shenzhen, 363. In each city, multistage stratified sampling was used to select individuals. Crucially, the number and distribution of the sample in each city and district were controlled according to the total number and distribution of the migrant population (Figure 1). A total of 2468 questionnaires was prepared for distribution. After excluding discrepancies and incomprehensible answers, our dataset comprised 2416 cases for analysis, with an effective rate of 97.8%.

All the interviewees were members of the floating population who did not have registered households in the urban area of the city (had households registered in other counties or cities, whether rural or urban) and who worked and lived in the research area for more than 3 months. The survey mainly combined quota sampling, snowball sampling introduced by acquaintances, random sampling, and stratified sampling methods to investigate the floating population in various regional government units, factories, parks, supermarkets, and streets. Investigators conducted face-to-face interviews. Each interview lasted approximately 50 min. The survey mainly collected information regarding the personal and family situations of the floating population, including the household registration, employment, labor contract, property, and education of children. To improve the reliability and robustness of the questionnaire, not more than 15 questionnaires were usually given at a specific area. Moreover, we provided remuneration to the respondents and registered their telephone numbers for verification. We primarily used descriptive statistics and a logistic regression model to analyze the data from the 2416 cases.

3.2. Dependent and Independent Variables

3.2.1. Dependent Variable Selection

The dependent variable, migrants’ settlement intentions, was assessed using migrants’ permanent settlement plans. In the survey questionnaire, we designed the question “What is your plan for permanent settlement in the future?” In the probit regression model, those who selected “To permanently settle in my current city” were set as 1, whereas those who selected “Unsure or other cities” and “Return to where I came from” were set as 0.

3.2.2. Independent Variable Selection

Reference to the literature revealed that the main factors affecting migrants’ settlement intentions are economic, sociocultural, and institutional [40,41,42,43,44,45]. After further reference to relevant studies and examination of the key problems we wished to address, we classified the influencing factors into individual economic and skill-related factors, family and social factors, and institutional factors.

A.

Individual economic and skill-related factors

Individual differences are a major reason for differences in settlement intention. No consensus regarding the influence of age on settlement intention has been reached. Overall, however, young people have stronger intentions to settle in cities. Females more commonly settle in cities than do males [46,47]. According to a study on the settlement intentions of people of different occupations, investor-type migrants have the strongest city settlement intention, followed by intellectual-type migrants, and labor-type migrants [4]. The market performs selection according to its labor capital needs. Migrants move to cities with more job opportunities and higher wages [48,49,50]. On the basis of the literature, the current study adopted gender, age, occupation, educational attainment, and individual income as variables for assessing the impacts of individual economic and skill-related factors on settlement intention.

B.

Family and social factors

Chinese people emphasize familial harmony and typically prefer to marry before establishing their careers. Because family is valued, family factors have a critical impact on settlement intention [51]. This study assumed that the degree to which a family has all its members in the same city exerts considerable impacts on settlement intention. Social factors are critical perspectives for analyzing migrants’ settlement intentions. Some studies have included variables such as migrant networks in the destination city, the frequency of interaction with locals, environmental perception, and identification with the local society [44,47]. Therefore, this study used indicators such as the percentage of family members living in the same city (Proportion of family members living in the same city), social interaction (degree and frequency, Do you have social interaction with local people?), ability to speak local dialects (Mastery of local dialect), degree of identification with the city (Do you like residence city), frequency of park visits (How often do you go to the park), convenience of transportation (determined by the commute time, Distance from residence to work place by car) to assess the impacts of family and society on settlement intention.

C.

Institutional factors

The high concentration of migrants in large cities increases the management stress of city governments. Thus, Chinese megacities adopted the household registration system to limit the population inflow. The most widely used household registration policy is the points-based system [2]. In this study, we hypothesize that this system is considered a factor influencing migrants’ settlement intentions and use the variable of hukou acquisition, which relates to whether they meet the hukou threshold conditions in destination cities.

Having a guarantee for the livelihoods of Chinese farmers in the form of house foundation land and land ownership is regarded as the major factor connecting rural–urban migrants and farming villages. Weakening this connection may strengthen settlement intention [33,39]. This study examined the impacts of rural–urban migrants having house foundation land and land ownership on migrants’ settlement intentions. To quantify the impacts of land as an institutional factor on settlement intention, factors such as farmland ownership (Whether have own arable land) and house foundation land ownership (Whether have homestead land) were considered, as were willingness to relinquish rights in hometowns (Exchange welfare in hometown for household registration), land contracting rights (Exchange welfare in hometown for household registration), or house foundation land (Exchange homestead land for household registration and urban welfare) in exchange for urban registration status (

Table 1. Configuration of independent variables.

Category	Independent Variables	Configuration of Independent Variables
Individual economic and skill-related factors	Gender	0— Female, 1—Male
	Age	0—Born before 1979, 1—1980 and beyond
	Educational attainment	0—Primary school and below, 1—Junior and senior high school, 2—College and above
	Occupation	0—(Non-skilled or low-skilled) migrant workers, 1—Self-employed, private entrepreneurs, and professional technicians.
	House value	1—(≤200), 2—(200–500), 3—(500–1000), 4—(>1000) (Unit: 1000 yuan)
	Individual income	1—(≤30), 2—(30–60), 3—(60–100), 4—(>100) (Unit: 1000 yuan)
Family and social factors	Proportion of family members living in distance city	1—(≤25%), 2—(25–50%), 3—(50–75%), 4—(75–100%)
	Do you have social interaction with local people?	1—basically not, 2—not very often, 3—sometimes, 4—quite often, 5—often
	Mastery of local dialect	1—can’t understand, 2—can understand a little, 3—can understand but can’t speak, 4—can speak
	Do you like residence city	1—feel excluded, 2—have not decided, 3—like it
	How often do you go to the park	1—once a month, 2—once a week, 3—at least once a day
	Distance from residence to work place by car	1—less than 30 min, 2—more than 30 min
Institutional factors	Hukou acquisition (whether meet the hukou threshold conditions in destination cities)	0—No, 1—Yes
	Whether have own arable land	0—No, 1—Yes
	Whether have homestead land	0—No, 1—Yes
	Exchange welfare in hometown for household registration	0—No, 1—Yes
	Exchange arable land for household registration and urban welfare	0—No, 1—Yes
	Exchange homestead land for household registration and urban welfare	0—No, 1—Yes

).

3.3. Models and Methods

First, we analyzed all factors affecting migrants’ settlement intentions (model 1). Second, because the individual factors, living environments, and social networks of urban–urban migrants and rural–urban migrants differed, and because those two groups were highly heterogeneous, this study further analyzed the factors influencing settlement intention in urban–urban migrants (model 2) and rural–urban migrants (model 3). Due to the land system factor under the urban–rural dichotomy, urban–urban migrants do not have house foundation land and farmland in their hometowns. To facilitate the comparison of the factors influencing the settlement intentions of rural–urban and urban–urban migrants, in models 2 and 3, under land and institutional factors, we only considered the variable of whether the criteria for the points-based household registration system were met.

Before conducting regression analysis on the factors influencing settlement intention, the factors were subjected to collinearity analysis. No collinearity was detected, and the chi-squared value of 0.932 was significant, indicating that probit regression analysis could be performed. The model is as follows:

$$\mathrm{P}\left(Y_i=1|X_i\right)=\mathrm{G}\left(\mathrm{X}\mathsf{\beta }\right)=\mathrm{G} \left({\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{X}}_1+{\mathsf{\beta }}_2{\mathrm{X}}_2+\dots +{\mathsf{\beta }}_n{\mathrm{X}}_n\right)$$

(1)

where $Y_i$= 1 means the population is willing to settle in the city;$Y_i$= 0 means the population is unwilling to do so; ${\mathsf{\beta }}_0$ is a multiple regression constant; ${\mathsf{\beta }}_1,{ \mathsf{\beta }}_2,$…${\mathsf{\beta }}_n$ are partial regression coefficients; and ${\mathrm{X}}_1,{ \mathrm{X}}_2,$…${\mathrm{X}}_n$ are explanatory variables. The evaluation of the performance of the regression models is shown as

Table 2. Chi-square test.

	Chi Square	Freedom	Significance
Model 1	1799.03	2084	0.999
Model 2	1667.385	1666	0.486
Model 3	436.772	432	0.427

and

Table 3. Estimated natural response rate.

	Total Number	Response Number	Standard Error
Model 1	2097	632	0.033
Model 2	1641	424	0.037
Model 3	416	199	0.163

:

4. Results

4.1. Descriptive Statistics

We categorized the interviewees according to their household registration characteristics, asking them “Among your family members, does anyone have urban registration status?” Those who responded affirmatively and negatively were defined as urban–urban migrants and rural–urban migrants, respectively. Among the interviewers, 1912 were rural–urban migrants, and 504 were urban–urban migrants.

Regarding individual economic and skill-related factors, the rural–urban migrants and urban–urban migrants did not differ significantly in gender, age, or marital status. The urban–urban migrants (47.3%) had far greater intentions to settle in cities than did rural–urban migrants (26%). In both populations, the majority was unwilling to settle in cities. Overall, the rural–urban migrants had lower educational attainments, individual incomes, and total property values. For example, 44.9% of urban–urban migrants went to university, substantially more than the 18.4% of rural–urban migrants (

Table 4. Settlement intention and individual characteristics of rural–urban migrants and urban–urban migrants.

Variable Category		The Proportion of Rural–Urban Migrants (%)	The Proportion of Urban–Urban Migrants (%)
Settlement intention	Go back hometown or unsure	74	52.7
	Settlement in resident city	26	47.3
Gender	Female	45.4	47.8
	Male	54.6	52.2
Age	Born before 1979	35.5	33.9
	1980 and beyond	64.5	66.1
Marital status	Unmarried	33.9	32.7
	Married or divorced/widowed	66.1	67.3
Educational attainment	Primary school and below	16.3	0.6
	Junior and senior high school	65.3	49.1
	College and above	18.4	44.9
Occupation	(Non-skilled or low-skilled) migrant workers	61	36.9
	Self-employed, private entrepreneurs, and professional technicians	39	63.1
House value	≤200	53.9	26.8
	200–500	22.5	23.4
	500–1000	13.8	18.1
	>1000	9.8	31.7
Individual income	≤30	36.2	24
	30–60	46.9	41.7
	60–100	12	21.4
	>100	4.9	12.9

).

Regarding family and social characteristics, the percentage of family members living in cities was lower among the rural–urban migrants. Differences in social interaction were even greater. The urban–urban migrants socialized more with locals than did the rural–urban migrants. When urban–urban migrants were asked whether they socialized with locals, 51.3% responded with “often” or “somewhat often”. By contrast, only 34.4% of the rural–urban migrants indicated that they socialized with locals (

Table 5. Family and social characteristics of rural–urban migrants and urban–urban migrants.

Variable Category		The Proportion of Rural–Urban Migrants (%)	The Proportion of Urban–Urban Migrants (%)
Proportion of family members living in resident city	≤25%	24.1	18.7
	25–50%	32	28.6
	50–75%	14.7	17.5
	75–100%	29.1	35.3
Do you have social interaction with local people?	Basically not	27.6	14.5
	Not very often	22.5	18.3
	Sometimes	15.5	15.9
	Quite often	11.9	18.8
	Often	22.5	32.5
Mastery of local dialect	Can’t understand	28.4	20.4
	Can understand a little	21.7	20.2
	Can understand but can’t speak	19.4	24.2
	Can speak	30.6	35.1
Do you like residence city	Feel excluded	1.9	0.8
	Have not feel	41.3	30.2
	Like it	56.8	69
How often do you go to the park	Once a month	73.2	70.5
	Once a week	19.7	21.5
	At least once a day	7.1	8
Is it convenient for you to see a doctor	Inconvenient	12	11.5
	Medium	19.3	20.2
	Convenient	68.7	68.3
Distance from residence to work place by car	Less than 30 min	95.3	87.4
	More than 30 min	4.7	12.6

). Overall, 69% of the urban–urban migrants identified with the city, compared with only 56.8% for the rural–urban migrants.

Regarding institutional factors, settlement intention was substantially affected by three factors, namely whether they satisfied the criteria for household registration, whether they owned house foundation land in their hometown, and whether they had farmland in their hometown. Nearly 75% of participants who did not meet the criteria indicated that they would not settle in cities. Among those meeting the criteria, only 60% chose to settle in cities. Overall, 75% and 71% of participants who had farmland and house foundation land in their hometowns were unwilling to settle in cities, respectively.

4.2. Analysis of Factors Influencing Settlement Intention

4.2.1. Factors Influencing Settlement Intention in All Migrants

Factors influencing settlement intention in all migrants were analyzed (

Table 6. Comparison of settlement intention among migrants with different characteristics related to household registration and land ownership.

		Unwilling/Not Sure		Willing		Total Number (n)
		Number (n)	Proportion (%)	Number (n)	Proportion (%)
Hukou acquisition	No	1518	74.8	511	25.2	2029
	Yes	143	40	214	60	357
Whether have own arable land	No	473	58.7	332	41.3	806
	Yes	1189	75	396	25	1585
Whether have homestead land	No	167	56.2	130	43.8	297
	Yes	1493	71.2	586	28.2	2079
Exchange welfare in hometown for household registration	No	1143	81.9	252	18.1	1395
	Yes	515	52	475	48	990
Exchange arable land for household registration and urban welfare	No	1382	77.9	393	22.1	1775
	Yes	276	46.1	323	53.9	599
Exchange homestead land for household registration and urban welfare	No	1421	77.3	418	22.7	1839
	Yes	235	44	300	56	535

). Significance was reached in some variables under individual economic and skill-related factors, family and social factors, and institutional factors, namely birth year, educational attainment, occupation, percentage of family members living in the same city, social interaction, mastery of the local dialect, degree of city identification, own arable land ownership, homestead land ownership, whether the criteria for urban registration status were met, willingness to relinquish rights in one’s hometown in exchange for urban registration status, and willingness to surrender one’s house foundation land in exchange for urban registration status.

Among the variables under the individual and family factors, women had a higher probability of settling in the city than did men, in line with the results of another study [47]. The percentage of people born after 1980 who settled in the city exceeded that of people born before 1980, possibly because most people born after 1980 had not engaged in agricultural activities. If they returned to their farming village hometowns, they might lose their livelihoods. They were also more adapted to city life. As for people born before 1980, they might hold the traditional Chinese belief that one should return to one’s roots.

Educational attainment is a critical reflection of human capital. More highly educated people face fewer challenges in economic and social integration in city life [31,52]. Specifically, educational attainment was crucial for the present rural–urban migrants, whose educational attainment was low overall. A rise in income standards increased the feasibility of settling in cities. Due to the expense of city life, income was connected to quality of life. Therefore, the high-income group was more likely to settle in cities [2].

Since ancient times, Chinese people have attached great importance to the concept of home. Family bonds constitute essential emotional connections. The more family members the migrants had in the destination city, the stronger their emotional connection with that place, corresponding to a significantly stronger settlement intention [27]. Moving the entire family to a destination reduces the number of visits migrants make to their hometowns, thereby weakening their ties to those locations. As a result, the probability of settling in the city increases.

As for variables under social factors, social interaction and dialect proficiency represented the degree of social integration. Specifically, high-frequency social interaction and high dialect proficiency represented a high degree of social integration and, therefore, a high probability of settlement. Furthermore, high city identification indicated strong settlement intention.

Regarding variables under institutional factors, the percentage of migrants with farmland in their hometowns who intended to settle in cities was 15.4% lower than that of their landless counterparts. In other words, migrants with farmland in their hometowns were more inclined to settle there. Compared with people who did not satisfy the criteria for urban registration status, those who did had a 51.9% higher probability of settling in cities. They also had more favorable socioeconomic conditions and skill attainment, and tended to settle in cities. Compared with those who were unwilling to give up their rights in their hometowns, the migrants willing to do so were 28.9% more likely to settle in cities. Compared with those who were unwilling to give up their house foundation land, the migrants willing to do so were 44.7% more likely to settle in cities. Overall, the migrants who were willing to give up their rights in their hometowns and their house foundation land to settle in cities adapted better to city life and had stronger settlement intentions.

The continuous reform of the hukou system over the past decade has dramatically lowered the institutional threshold of the floating population for settling down permanently [2]. Migrants who hold an urban hukou or whose hukou is registered within the Guangdong province can adapt to urban life and exhibit a relatively high degree of integration, recognition, and support for local life in the destination cities. Their willingness to settle down is correspondingly stronger than that of their counterparts [47]. The background to this study is that the urban–rural dichotomy has been enforced for several decades. This explains the considerable differences between the rural–urban migrants and urban–urban migrants in individual, family, and social factors (Table 4 and Table 5). Those variables exerted significant effects on their settlement intentions, supporting hypothesis 1. Under general circumstances, migrants meeting the criteria of the points-based household registration system were more inclined to settle in cities. Urban–urban migrants had better socioeconomic statuses and skill attainment and, therefore, met the requirements more easily. Overall, the long-term impacts of the household registration system, the points-based household registration system, and the new land system caused differences in the settlement intention of urban–urban migrants and rural–urban migrants. Thus, hypotheses 2 and 3 are supported.

4.2.2. Factors Influencing the Settlement Intention of Rural–Urban Migrants and Urban–Urban Migrants

To further examine the factors influencing the settlement intentions of rural–urban migrants and urban–urban migrants, probit regression models 2 and 3 were established, the results of which are listed in

Table 7. Probit regression analysis of reasons underlying migrants’ settlement intentions (model 1).

Independent Variables	Standard Deviation	Z Score	Coefficient
Gender	0.064	0.192	0.012
Age	0.075	2.796	0.21 **
Educational attainment	0.064	2.035	0.13 *
Occupation	0.069	3.187	0.22 ***
House value	0.03	1.695	0.051
Individual income	0.039	0.006	0.000
Proportion of family members living in the same city	0.027	1.99	0.054 *
Social interaction	0.023	3.568	0.082 ***
Mastery of local dialect	0.029	2.186	0.064 *
Do you like residence city	0.062	4.218	0.262 ***
How often do you go to the park	0.049	1.817	0.09
Distance from residence to work place by car	0.124	0.903	0.112
Hukou acquisition	0.088	4.354	0.383 ***
Whether have own arable land	0.071	−2.301	−0.164 *
Whether have homestead land	0.101	1.964	0.199 *
Exchange welfare in hometown for household registration	0.083	2.545	0.21 *
Exchange arable land for household registration and urban welfare	0.165	−0.976	−0.161
Exchange homestead land for household registration and urban welfare	0.165	2.827	0.465 **

Note: ***, **, and * denote significance at 1%, 5%, and 10%, respectively.

. Variables that significantly affected the settlement intentions of both rural–urban migrants and urban–urban migrants included the percentage of family members living in the same city, social interaction, degree of city identification, and whether the criteria for urban registration status were met. Regarding institutional factors, the satisfaction of these criteria significantly affected both types of migrants. Specifically, for the urban–urban migrants, the probability that people meeting the criteria would eventually settle in cities was 93.1% higher than that of those not meeting the criteria.

The variables exerting significant impacts on rural–urban migrants’ settlement intentions included year of birth, educational attainment, occupation, total property value, social interaction, dialect proficiency, and transportation convenience. The rural–urban migrants born after 1980 had a 30.3% higher probability of settling in cities than those born before 1980. A high value of the owned property resulted in strong settlement intention in cities in rural–urban migrants. Under the background of the urban–rural dichotomy, rural–urban migrants constitute a disadvantaged group in society. Because of reasons such as low educational attainment, some rural–urban migrants struggle to adapt to urban life. For rural–urban migrants, high social integration and dialect proficiency serve to enhance adaptability to life in the destination city, thereby generating high levels of social and city identification, and potentially strengthening settlement intention.

Variables significantly affecting urban–urban migrants’ settlement intentions included gender, the percentage of family members living in the same city, social interaction, city identification, and park visit frequency. Gender had a certain degree of influence on the settlement intentions of urban–urban migrants. The probability of urban settlement for women was 28.4% higher than that of men. When urban–urban migrants decide to settle permanently in a city, they might pay particular attention to family members being in the same city, social relations between different cities, and city identification. Due to the gap in job opportunities being smaller than that between urban and rural regions, variables such as educational attainment, occupation, and income were not significant.

Overall, several variables reflecting individual, family, and social characteristics exerted significant impacts on the settlement intentions of rural–urban migrants and urban–urban migrants. In China, the household registration system strictly divides residents into urban and rural residents, thereby resulting in notable differences between the two groups in terms of education, social security, employment opportunities, social welfare, financial conditions, and living habits. These differences may, to a considerable extent, determine whether the floating population can adapt to urban life and be integrated into the urban society by establishing social networks [2,35]. Thus, hypothesis 1 is validated(

Table 8. Probit regression analysis of settlement intentions in city–city and rural–urban migrants.

Independent Variables	Model 2 (Rural–Urban Migrants)			Model 3 (City–City Migrants)
	Standard Deviation	Z Score	Coefficient	Standard Deviation	Z Score	Coefficient
Gender	0.075	−1.683	-0.126	0.139	−2.04	−0.284 *
Age	0.088	3.438	0.303 ***	0.165	0.633	0.104
Educational attainment	0.074	0.7	0.052 *	0.132	0.63	0.083
Occupation	0.078	4.176	0.326 ***	0.146	−1.515	−0.221
House value	0.035	3.437	0.121 ***	0.059	-0.008	0.000
Individual income	0.047	0.891	0.042	0.078	0.971	0.076
Proportion of family members living in the city	0.032	5.465	0.173 ***	0.062	2.974	0.184 ***
Social interaction	0.026	3.107	0.08 **	0.056	2.359	0.131 *
Mastery of local dialect	0.033	3.578	0.119 ***	0.069	0.157	0.011
Do you like residence city	0.071	7.046	0.503 ***	0.154	4.619	0.713 ***
How often do you go to the park	0.058	1.04	0.061	0.11	2.859	0.316 **
Distance from residence to work place by car	0.159	2.292	0.363 *	0.21	1.004	0.211
Hukou acquisition	0.11	2.66	0.294 **	0.155	6.001	0.931 ***

Note: ***, **, and * denote significance at 1%, 5%, and 10%, respectively.

).

5. Discussion and Conclusions

Migrants’ settlement intentions are formed by multiple factors [4,5,6,7,8,9,10]. Investigating the institutional factors underlying the settlement intentions of migrants in China, an emerging market, serves as a valuable reference for research on international migration theory. The long-established Chinese household registration system may constitute a critical determinant of differences in settlement intentions among rural–urban migrants and urban–urban migrants. In this study, we discussed the factors influencing these intentions with regard to individual differences in socioeconomic status and skill level, family and social characteristics, and institutional characteristics. One focus was on the impacts of institutional factors on the settlement intentions of the entire sample. We have several findings in our research:

Firstly, the urban–rural dichotomy has exerted long-term impacts on migrants’ settlement intentions. Some studies have argued that migrants in China would choose to settle permanently in cities if the household registration restrictions were lifted [12,14,36,37]. However, only 30.13% of the interviewees intended to settle in their destination cities. The number of migrants who intended to move back to their hometowns was higher than that reported in relevant studies. Specifically, rural–urban migrants were not strongly inclined to settle in cities. The principal reason was that rural–urban migrants differ substantially from urban–urban migrants in individual, family, and social characteristics because of the urban–rural dichotomy, because the household registration system strictly divides residents into urban and rural residents in China [2,35]. These differences may, to a considerable extent, determine whether the floating population can adapt to urban life and be integrated into the urban society by establishing social networks. This study unveiled that migrants’ individual, family, and social characteristics all exhibited significant influence on their settlement intentions, and rural–urban migrants and urban–urban migrants had different settlement intentions.

Secondly, the points-based system implemented in the new Pearl River Delta megapolis has relaxed restrictions on household registrations and affected migrants’ settlement intentions to a certain extent. However, according to the results on path dependence regarding the impacts of the urban–rural dichotomy, urban–urban migrants had higher educational attainment and incomes than did the rural–urban migrants, and they more easily satisfied the criteria for urban registration status. In addition, since the beginning of the 21st century, China’s city–county relationships have shifted from exploitative to protective. Moreover, cities’ reliance on farming villages decreased. Under the economic radiation effect of the cities, farming villages received large amounts of resources from cities. At around the same time, social insurance for farming village residents was established. Although the benefits this population receives differ greatly from those enjoyed by city residents, the value of rural registration status has increased, reducing the urban–rural divide to a certain extent. Furthermore, the effects of people with rural registration status having urban registration status were reduced.

Thirdly, migrants with farmland were less likely to settle in cities than were their landless counterparts. Land system factors have a certain impact on immigrants’ willingness to settle down. The main reason is that farmers can obtain a basic guarantee of a livelihood through farming. Also, as the country urbanizes, the land of farming villages is being expropriated, for which farmers can receive some compensation. Ownership of house foundation land did not significantly affect settlement intention. This can be explained by the fact that urbanization in China began relatively late, and the country is still in the middle-to-late phase of urbanization; therefore, most rural–urban migrants own house foundation land.

The government should establish policies for settlement, housing, and social security according to migrants’ registration status. A solid higher education system and the availability of high-quality jobs can significantly strengthen the settlement intentions of rural–urban migrants. Therefore, we suggest that the government increase its input in education to increase higher education enrollment rates, specifically in the farming village population. To promote urbanization, policies should include measures that expand migrants’ opportunities to obtain jobs and social security, such as affordable apartment rentals that rural–urban migrants and urban–urban migrants have equal opportunities to obtain. Other measures can center on children’s education and social welfare. Specifically, for the rural–urban migrants with less favorable family conditions and lower socioeconomic status, as well as for those willing to relinquish their farmland and house foundation land in exchange for urban registration status, the government should implement policies supporting urban employment and settlement. This can facilitate the upward mobility of migrants, contributing to social equity and harmony.