1. Introduction

Happiness, as a “Hidden national wealth” , is an inalienable right enjoyed by all. Although the economic research on the effect of residents’ well-being started late, the related research results are still quite abundant [1]. In many economic studies, the relationship between income and resident happiness is the eternal topic of discussion. Easterlin found that the increase of income level in a country does not necessarily increase the average well-being of its residents. In some periods, there may even be a contradiction between income level and residents’ well-being, this phenomenon is known as the “Easterlin paradox”. Although not all scholars agree that the “Easterlin paradox” exists [2] , but “Easterlin paradox” has laid a foundation for some scholars to study the relationship between income and residents’ happiness from the perspective of relative deprivation. For example, relevant studies have found that after controlling the individual income level, the relative income level of the surrounding population will have a negative impact on residents’ well-being [3]; other studies have also confirmed that the more unequal the individual income, greater differences in well-being [4].

In addition to income, unemployment, consumption and inflation are also important factors affecting residents’ happiness. In the study of the effect of unemployment on residents’ well-being, most scholars think that unemployment will cause the loss of residents’ well-being. For example, studies have shown that unemployment significantly reduces the mental health of the unemployed [5]. Inflation has a negative impact on the well-being of residents, mainly because inflation and inflation expectations will lead to the deterioration of income distribution among residents and reduce their living standards [6]. On one level, consumption may be more reflective of an individual’s true level of well-being than income.

Different from the above studies, this paper focuses on the establishment of this employment system with Chinese characteristics in a society with an underdeveloped labor market and a long-term existence of a large number of informal employments, does it have a significant protective effect on residents’ well-being, with a besieged effect on happiness? Based on the above analysis, using the data of 2018 Chinese family tracing survey, this paper tests the happiness siege effect. Different from previous studies on the relationship between residents’ well-being and the welfare differences between contract workers and employed residents, civil servants and employees of enterprises, the innovation of this paper is to analyze the relationship between compilation and residents’ well-being directly by using micro-data. The results of this study not only enrich the research on the determinants of our residents’ well-being, at the same time, it provides a new perspective to understand the long-standing “Establishment fever” in China’s employment market and the phenomenon that “Establishment fever” presents different characteristics in different groups.

2. Data, variables, and models

2.1. data sources

The data for this study come from the China Family Panel Study (CFPS) conducted by the China Centre for Social Sciences at Peking University. The CFPS project aims to track the evolution of China’s society, economy, demographics, education and health. As a large number of samples were lost, this study mainly uses the recently published 2018 CFPS tracking survey data. In addition, according to the research needs, data preprocessing was done for this study, and a total of 14,682 valid samples were obtained after removing missing observations of key variables.

2.2. variable description and description

The interpreted variable. The main aim of this paper is to study the happiness siege effect, so we take the individual’s “Subjective well-being” as the interpreted variable. The CFPS survey asks the question “How happy do you think you are?” and measures overall well-being by giving respondents an overall assessment of their overall well-being and a firm rating. A score of 0 indicates the lowest level of happiness and a score of 10 indicates the highest level of happiness.

Core explanatory variables. This paper mainly discusses the mechanism of the influence of compilation on well-being, so whether compilation is the core explanatory variable is chosen. The CFPS question about whether an individual is organized is “Are you organized in the workplace?” When the respondent answers “Yes,” the respondent is organized and assigned a value of 1; when the respondent answers “No,” the respondent is unorganized and assigned a value of 0.

Control variables. There are other factors that affect residents’ happiness besides whether they are organized or not. Referring to the research of relevant scholars [7], this paper tries to control the individual characteristics and family characteristics which are closely related to residents’ happiness [8]. Specifically, the individual characteristics variables include: age, age square, gender, education, marriage, self-rated health, household registration. Family size was selected as the family characteristic variable.

Table 1. Descriptive statistics of variables.

Model setting. Referring to the happiness determination model established by scholars, the basic model of this study is as follows:

\( {Happiness_{ij}}={α_{0}}+{α_{1}}{Establish_{ij}}+{βX_{ij}}+{γC_{j}}+{μP_{ij}}+{ε_{ij}}\ \ \ (1) \)

In formula (1) , \( {Happiness_{ij}} \) was the interpreted variable, expressed using the subjective well-being of the i individual in the j community; Establishij was the core explanatory variable, using community j, whether the individual i in Community j has a compiled representation; \( {X_{ij}} \) represents a set of individual and family characteristics that influence individual i happiness in Community j, \( {C_{j}} \) is the characteristic factor of the community, \( {P_{ij}} \) is the virtual variable of the province, and \( {ε_{ij}} \) is the random disturbance term of the model. \( {α_{0}} \) is a constant term, \( {α_{1}} \) is the influence coefficient of the happiness siege effect, β､µ､γ is the influence coefficient of different control variables.

3. Empirical results and analysis

3.1. To compile the baseline regression analysis of happiness siege effect

Table 2 reports the OLS and Order Logit regression results used to preliminarily explore the compilation of the happiness siege effect. The (I)-(III) columns were the OLS regression results of the effect on residents’ well-being after controlling for different variables, such as individual characteristics and family characteristics, respectively, columns (V)-(VII) are the results of the Order Logit regression with no control variables and the effect on residents’ well-being after controlling for different variables. The OLS regression results showed that the symbols and significance of the estimated coefficients were still highly consistent, and both were positively significant at the 1% statistical level. This suggests that people who are registered have significantly higher subjective well-being than those who are not. The results of the Order Logit regression also showed that the subjective well-being of the residents with a list was significantly higher than that of the residents without a list. The above results all confirm that the establishment can significantly improve the residents’ well-being, that is, the establishment has a siege effect on the residents’ well-being.

The results of control variables estimation show that the well-being of residents with older age is lower, which may be due to the gradual decline of individual physical function with older age, which will negatively affect the residents’ well-being. There is a significant positive correlation between age square and residents’ well-being, indicating that the relationship between residents’ well-being and age is not a single linear relationship. Education significantly positively affected residents’ subjective well-being, which may be related to the income level, working environment and social status of residents with higher education. The subjective well-being of residents in marriage is significantly higher, indicating that marriage is an important protective factor of residents’ subjective well-being. Self-rated health significantly positively affected residents’ subjective well-being. The larger the family size, the better the subjective well-being of residents, which may be because the larger the family size, through mutual help can minimize the economic burden of individual residents.

Table 2. The baseline regression analysis of happiness siege effect.

3.2. heterogeneity analysis

The regression results in table 3 show that the regression coefficients based on the estimates of both male and female samples were positive and passed the significance test at the 1% level. The estimated results also show that the regression coefficients based on the female sample estimates are significantly larger, indicating the compilation is more beneficial to women’s happiness. This may be due to the fact that the establishment is able to better meet women’s expectations of a relatively better remuneration for work.Table 3 also shows the estimated results of the impact of compilation on the well-being of residents with different household registration. On the whole, the establishment has a significant positive impact on the subjective well-being of both non-agricultural and agricultural residents, and the establishment has a greater impact on the subjective well-being of agricultural residents. The reason may be that they are more conservative, and more inclined to work in a relatively high income and stable system.

Table 3. The impact of the establishment on the happiness of residents of different genders and household registration.

4. Conclusions and policy recommendations

Based on the 2018 China family follow-up survey (CFPS) data, this paper discusses and tests the encirclement effect of compilation on residents’ well-being and its possible mechanism. The findings are as follows: first, under the control of individual, family, community and provincial characteristics, the subjective well-being of the residents is significantly higher.In terms of group differences, compared with male and rural residents, the effect of happiness encirclement is more obvious in women and urban residents.In terms of practical significance, the conclusions of this study provide two important inspirations for regulating the labor market and improving the welfare of residents, the function of special welfare label should be weakened. It is suggested to regulate the post promotion, welfare level and working hours of residents in work and non-work in the form of laws and regulations in the short term. Second, we should vigorously promote cultural and ethical progress and create a good social climate. It is suggested that special attention should be paid to reducing the consciousness of power and service of the residents, and to strengthening the sense of equality of the public, so as to weaken the social influence of the residents and guarantee the welfare level of the public.