How to Promote Balanced and Healthy Development of Residents’ Leisure: Based on the Analysis on the Spatiotemporal Evolution of the Scale Structure of Leisure Consumption of Urban Residents in China

Abstract

This study aimed to find the spatiotemporal evolution rules of the leisure consumption of urban residents in China. Based on the measurement of the marginal propensity of leisure consumption for urban residents over the past 20 years, a mixed regression of leisure consumption and disposable income showed that, the growth rates of education, culture, entertainment, transportation, communication, and healthcare large, and these growth trends remain obvious. The coefficient of variation of leisure consumption shows that, transportation and communication consumptions differ among regions. The differences in education, culture, and entertainment consumption have rapidly converged. The spatial correlation analysis of the Moran index showed that, the regional association degree of leisure consumption is not obvious, but there is a clear mutual pull regarding education, culture and entertainment consumptions. Based on the results, suggestions were proposed to further expand the leisure consumption needs of residents, so as to achieve sustainable and sound development of residents’ leisure.

1. Introduction

Leisure consumption is one of the main forms of residents’ participation in leisure activities. It refers to people’s paid consumption activities, in which they purchase leisure goods or services to meet their leisure needs [1]. Over the past forty years of reform and opening up, the social economy in China has developed rapidly, and the quality of life of China’s residents has changed dramatically. According to the World Bank, China’s GDP in 2017 was US $149,540 million, an increase of more than 80 times since 1978. The improvements in income levels and the changes in the concept of consumption have led to a significant increase in the leisure consumption scale and the optimization of the leisure consumption structure [2]. However, in a certain sense, the social atmosphere of rich leisure remains in part elusive and has deviated from the traditional culture of diligence and thrift in China. The breadth and depth of residents’ leisure consumption remains to be further expanded. In the 1990s, with the full implementation of the five-day work system and the adjustment of holidays, leisure consumption gradually became the new fashion of Chinese residents’ life. Accordingly, Chinese scholars began to pay more attention to leisure consumption research. However, COVID-19 swept the world in 2020. The epidemic restrained the increase in leisure consumption expenditure and widened the gap with economic growth [3,4]. The rigorous role of leisure consumption in the rejuvenation of the economy is even more obvious.

In addition, the major contradiction within China has resulted in an important change, and people’s satisfaction with the need for a better life is seen to be in a more prominent position. However, it is worth emphasizing that a good life can be manifested to a large extent through the leisure consumption of residents. Releasing residents’ leisure consumption potential in an all-round way is an important measure to alleviate major social contradictions and realize the transformation and upgrading of residents’ consumption structure. Under the comprehensive effect of various social and economic factors, the scale of leisure consumption of Chinese residents has continued to expand [5], which is conducive to further optimization of the industrial structure and significant growth of the national economy.

Researchers have paid attention to leisure consumption since 1960s. Becker pointed out that all leisure activities contain some kind of consumption, and all consumption activities contain some kind of leisure [6]. From the existing research results, some scholars defined leisure as a concept of time, theoretically analyzed the relationship between leisure and consumption, and then studied the impact of leisure consumption on economic growth. Others regarded leisure consumption as an individual activity and investigated the characteristics of leisure consumption behavior and its influencing factors [7].

Scholars have mainly discussed this topic in the two fields of economics and sociology. The study of economics has mainly focused on two aspects: one is the decision-making choice of leisure and consumption [8]. In the choice of different leisure activities, there is a certain substitution relationship between time and consumption [9]. The consumption decision-making theory based on time distribution has been deeply studied by scholars. It is a very important focus to bring the opportunity cost of time use into the leisure demand system, and then study the consumption expenditures of residents [10,11,12]. The second issue is testing the maximization of leisure and consumption utility. The non-parametric method has been used to evaluate the utility maximization of leisure and consumption [13]. The optimal portfolio, leisure consumption and retirement options under the utility of the invariant alternative elastic function has been studied [14]. The study of sociology has covered the motivations and preferences of leisure consumption [15,16,17], characteristics and trends [18,19], experience and quality [20,21], the construction of leisure consumption theory models [22], and the investigation of capacity and structure [1,23].

In modern society, conspicuous consumption is a type of social behavior and an important representation of individual identity. The level of residents’ income, consumption environments, supply of leisure consumer goods and services, and people’s values and consumption outlook are four important factors affecting leisure consumption [24]. However, as seen from the existing literature, scholars have been more concerned with the specific factors influencing leisure consumption [25]. One is income from different sources and forms. Using data from a German continuous household budget survey, one paper estimated the leisure service derivative demand system and provided a consistent derivative instrument of leisure service expenditure elasticity [26]. Disposable income and face value have an important impact on outdoor leisure consumption [27]. A second factor consists of the amount and allocation of leisure time. By constructing a leisure consumption model with "time constraints", a paper discussed the conditions of the optimal leisure consumption choices of consumers and analyzed the influences of wage rate, non-wage income and working time changes on the leisure consumption choice [22]. Government expenditures and the disposable income of urban residents are important factors that affect the leisure consumption of Chinese residents [28]. The substitution effect plays a leading role. The advantage of economic foundation is weaker than the conflict of time allocation. The degree of busyness restricts consumption upgrading [29]. The third factor consists of the concept and social psychological characteristics of leisure consumption. Social status and self-esteem have impacts on conspicuous leisure consumption [30]. A paper studied the potential influencing factors of Generation Y coffee consumers in South Korea and explored the influences of personal material conditions and social conformity, demonstrating the impacts of psychology and functional quality on leisure consumption [31]. In addition, some scholars have studied the impact of population age structure on leisure consumption and found that children’s dependency ratio has a negative impact on residents’ leisure consumption, while the elderly dependency ratio has a positive effect [32]. Retirement has a significant negative impact on housing consumption [33], which is conducive to the increase in leisure consumption.

Most scholars have mainly conducted theoretical discussions and empirical tests on the externalities of leisure and consumption. The externalities of consumption and leisure affect the long-term equilibrium and optimal growth rate in relatively different ways. The growth and welfare effects of light consumption and leisure externalities are obvious [34]. By extending the previous model, which only considered the externalities of production and consumption, one paper studied the optimal tax policy under the externalities of production, consumption and leisure [35]. However, Chinese scholars have directly studied the relationship between leisure consumption and economic growth. Based on a new empirical test of the quantitative relationship between domestic tourism consumption and the income of urban and rural residents in China, it was found that the conclusions drawn by Zhou and Li [36] were not convincing. There is no long-term co-integration relationship between tourism consumption and the current income of Chinese residents, and the tourism consumption function based on the absolute income hypothesis cannot explain very well the tourism consumption of Chinese residents’ fee behaviors [37]. Cultural consumption not only contributes to the improvement of China’s total consumption level, but it also significantly improves the internal structure of China’s consumption [38]. For leisure consumption of rural migrant workers, both income effect and cultural effect exist [39]. However, considering such factors as geographical location and leisure consumption, people’s leisure opportunities in society are uneven [40]. According to data released by China’s National Bureau of Statistics, the per capita disposable income of urban residents in China increased from ¥5425 in 1998 to ¥33,616 in 2016, and the per capita consumption expenditure increased from ¥4332 to ¥23,079. More specifically, the per capita leisure consumption expenditure rose from ¥1515 in 1998 to ¥9464 in 2016, which is an increase of 6.2 times. However, China has more than one fifth of the world’s population and 34 provincial administrative regions. There could be obvious regional differences in residents’ leisure consumption. In addition, leisure consumption involves many aspects of residents’ lives, and its internal structure has a certain complexity [41]. Therefore, this paper attempts to provide useful suggestions for promoting the balanced and healthy development of residents’ leisure through research on the scale structure and space–time evolution of China’s urban residents’ leisure consumption. The main contributions are reflected in the following two aspects: First, through the collection of long-term statistical data, the evolution trend and regional differences of China’s urban residents’ leisure consumption are investigated from the macro level, and then theoretical suggestions are provided for the national leisure business planning and consumption policy formulation. Second, the tendency and preference of Chinese urban residents’ leisure consumption are explored from the perspective of the internal structure of residents’ leisure consumption, so as to seek and grasp the entry point and focus of leisure consumption guidance and provide countermeasures and suggestions for the optimization of urban industrial structure and the transformation and upgrading of residents’ consumption.

2. Research Methods

2.1. Marginal Propensity to Consume and Income Elasticity

The marginal propensity to consume is the ratio of consumption increase and decrease to the alteration of disposable income. It reflects the consumption change when each unit of disposable income increases or decreases. Income elasticity refers to the extent of the demand quantity changes due to the variety of consumer income, under the condition that prices and other factors remain unchanged. To a certain extent, the marginal propensity to consume and income elasticity can reflect the changing trends in urban residents’ leisure consumption structures [42]. The evaluation models are set separately as follows:

$$c_{it}^k={\alpha }_1+{\alpha }_2{y}_{it}+{\epsilon }_{it}$$

(1)

$$\ln c_{it}^k={\beta }_1+{\beta }_2\ln y_{it}+{\mu }_{it}$$

(2)

where $c_{it}^k$ is the actual expenditure of various leisure consumptions of urban residents ($k$ = 1, 2, …, 5 represent various aspects of leisure consumption, such as daily necessities and services, transportation and communication, education, culture and entertainment, medical needs and healthcare, and other supplies and services). $y_{it}$ is the per capita disposable income of urban residents. $i$ is the province and city. $t$ is the year.

2.2. Coefficient of Variation (CV)

This study introduced the CV to analyze the spatial balance of the leisure consumption structure. Known as the discrete coefficient, the CV can also reflect the spatial distribution of the leisure consumption structure of urban residents to a certain extent in China [43]. The larger that the CV value is, the more uneven that the spatial distribution is. The calculated formula is as follows:

$$CV=\frac{1}{\overline{x}}\sqrt{{\displaystyle \sum _{i=1}^n{\left(x_i-\overline{x}\right)}^2/n}}$$

(3)

In the formula, $x_i$ refers to the leisure consumption expenditures for urban residents’ daily necessities and services, transportation and communication, education, culture and entertainment, medical care, other supplies and services in all provinces and cities. ${\overline{x}}_i$ refers to the average leisure consumption expenditures of urban residents in all provinces and cities.

2.3. Global Moran’s Index (Moran’s I)

The spatial correlation of the leisure consumption structure was analyzed using Moran’s I. Moran’s I is a comprehensive evaluation index for the spatial autocorrelation degree and is used to measure the interdependence degree of attributes in space [43].

The calculated formula is as follows [44]:

$$Moran’sI=n{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{W}_{ij}\left(x_i-\overline{x}\right)\left(x_j-\overline{x}\right)}}/{\displaystyle \sum _{i=1}^n{\displaystyle \sum _{j=1}^n{W}_{ij}{\displaystyle \sum _{i=1}^n{\left(x_i-\overline{x}\right)}^2}}}$$

(4)

In the formula, $n$ represents the number of provinces and cities. $W_{ij}$ is the spatial weight matrix that is constructed according to the actual situation of each province and city. $x_i$ and $x_j$ represent the attribute values of spatial units $i$ and $j$, respectively. $\overline{x}$ represents the average of all spatial data of urban residents’ leisure consumption structure. The closer that the Moran index is to 0, the weaker that the spatial autocorrelation is. In contrast, the closer that the Moran index is to 1 or −1, the stronger that the spatial autocorrelation is.

3. Empirical Analysis

The data used in this study were mainly from the "China Statistical Yearbook" on the official website of the China’s National Bureau of Statistics from 1999 to 2017. The main variables involved were "per capita disposable income of urban residents" and "per capita consumption expenditure of urban residents by region". Among these data, per capita consumption expenditure includes eight categories of consumption indicators. Leisure consumption covers the consumption of food and clothing to meet physiological needs, sports and tourism to meet the needs of health and entertainment, and knowledge and skills learning activities to meet the needs of development and spiritual needs. Therefore, this study took the consumption of education, culture, entertainment, healthcare consumption, transportation and communication consumption, the consumption of daily necessities and services, and the consumption of other supplies and services as the measurement of leisure consumption. The Engel coefficient represents the proportion of food expenditure out of total consumption expenditures, calculated according to the ratio of food consumption expenditure to total consumption expenditure.

3.1. Characteristics of Total Change

Leisure consumption as a whole showed an increasing trend, but the growth rate was slower than those for income and total consumption. Since the reform and opening up, China’s economy has maintained a rapid growth momentum, and it surpassed Japan as the world’s second-largest economy after the United States in 2010. According to data released by the World Bank, China’s per capita GDP has increased 56.8 times over the past 40 years, including 10.7 times from 1998 ($829) to 2016 ($8866). Correspondingly, per capita disposable income and consumption expenditure have also changed dramatically. It is not difficult to see that the disposable income of urban residents grew faster than consumption expenditures from 1998 to 2016, while the growth rate of consumption expenditures was significantly greater than that of leisure consumption. From the trend of leisure consumption, we observed that following. (1) The per capita leisure consumption exceeded ¥3000 in 2004. Since the beginning of the new century, China has proposed the expansion of the macroeconomic strategy of domestic demand, and the holiday economy has played an important role in promoting leisure consumption. However, the severe acute respiratory syndrome (SARS) outbreak in 2003 undoubtedly suppressed the strong consumer market. With the gradual decline in the negative impact of the SARS outbreak, residents’ consumption, especially leisure consumption, has revived. (2) In 2009, per capita leisure consumption exceeded ¥5000. On the one hand, in 2007, the 17th National Congress of the Communist Party of China once again made it clear that economic growth must be driven by the coordination of consumption, then investments, and finally exports, instead of by investments, exports and consumption, which was the order that had greatly stimulated the rapid growth of residents’ consumption. On the other hand, in 2009, the State Council issued the "Opinions on Accelerating the Development of Tourism", and tourism once again became a new force and leader in the leisure consumption market. Additionally, with the gradual regression of the 2008 global financial crisis, China’s economic situation began to improve, and the income level of residents increased rapidly, providing a solid base for leisure consumption. (3) In 2013, per capita leisure consumption exceeded ¥8000. In 2012, the 18th National Party Congress once again stressed the firm grasping of expanding domestic demand as the strategic basis by accelerating the establishment of a long-term mechanism to expand consumer demand and increasing the potential for residents’ consumption. The adjustment and guidance of the national macroeconomic strategy remain important driving forces for the rapid growth of leisure consumption.

Leisure consumption accounts for a relatively stable proportion of disposable income and total consumption. From 1998 to 2016, the per capita leisure consumption, per capita disposable income and growth rate of per capita consumption were different. Despite this difference, the leisure consumption proportion of disposable income and total consumption remained within a certain range. The analysis found that the per capita leisure consumption proportion of urban residents’ disposable income was maintained at between 25% and 35% due to the following three causes. First, starting in 2000, there was a significant increase in the leisure consumption proportion of disposable income. In 2002, the percentage was as high as 33%. On the one hand, one reason for this increase is that after China implemented the Golden Week system in 1999, a tourism boom swept across the country, and the long holiday system became an important method of stimulating domestic demand and promoting consumption. On the other hand, in 2001, the State Council issued the "Notice on Further Accelerating the Development of Tourism". Tourism as a comprehensive industry had become an important engine for stimulating residents’ leisure consumption [45]. Second, the leisure consumption proportion of disposable income gradually decreased since approximately 2008. A possible reason for this decrease is that the growth rate of leisure consumption has been far behind the growth rate of per capita GDP. The formation of the leisure consumption concept requires a relatively long period of time, and there has been some inertia in leisure consumption behavior [46] causing the growth of leisure consumption to advance gradually in due order. In contrast, the favorable environment for China’s social and economic development has led to a rapid increase in per capita GDP. According to data released by the World Bank, China’s per capita GDP reached $1000 in 2001 and $2000 in 2006, and it exceeded $3000 by 2008. The growth rate of per capita disposable income was also great. In 1998, per capita disposable income was ¥5425. In 2005, it exceeded ¥10,000. In 2008, per capita disposable income already exceeded ¥15,000. Subsequently, its growth rate was even faster. Third, in 2014, the proportion of leisure consumption out of disposable income underwent a significant decrease and then slowly recovered. This decrease in the proportion of per capita leisure consumption was the main reason for the sharp decline in per capita disposable income. In addition, the proportion of per capita leisure consumption out of total consumption was basically maintained at approximately 35% to 45%, similar to the change in the ratio of per capita leisure consumption to disposable income.

The growth rate of per capita leisure consumption has slowed and has gradually synchronized with the growth of income and total consumption. From 1998 to 2016, the per capita leisure consumption of urban residents in China increased each year, but the growth rate demonstrated a slowing trend. As indicated in

Table 1. Growth rate of per capita leisure consumption, disposable income and total consumption.

Year	Leisure Consumption	Disposable Income	Total Consumption
1999	15.3	7.9	6.6
2000	16.7	7.3	8.3
2001	8.6	9.2	6.2
2002	14.9	12.3	13.6
2003	8.4	10	8
2004	10.7	11.2	10.3
2005	12	11.4	10.6
2006	10.5	12.1	9.5
2007	15	17.2	15
2008	7.5	14.5	12.5
2009	12.9	8.8	9.1
2010	11.7	11.3	9.8
2011	11.6	14.1	12.5
2012	11.4	12.6	10
2013	10.1	7.7	10.9
2014	−2.6	9	8
2015	9.6	8.2	7.1
2016	10	7.8	7.9
Average of 1999~2002	13.9	9.2	8.7
Average of 2003~2007	11.3	12.4	10.7
Average of 2008~2012	11	12.3	10.8
Average of 2013~2016	6.8	8.2	8.5

, the growth rate of per capita leisure consumption in 1999 and 2000 remained greater that 15% but fell to less than 10% in 2001, 2003, 2008 and 2014; it even underwent negative growth in 2014. To examine the growth rate of per capita leisure consumption more clearly, the inspection cycle was divided into four stages according to the principle of symmetry of year intervals. It should be noted that in the beginning of each stage, the growth rate of per capita leisure consumption is exactly at the fall point of the ratio. The average annual growth rate of the per capita leisure consumption was 13.9% between 1999 and 2002, 11.3% between 2003 and 2007, 11% between 2008 and 2012, and 6.8% between 2013 and 2016.

Comparing the growth rate of per capita leisure consumption with that of disposable income, we found that there was a large gap between 1990 and 2000, in which the growth rate of per capita leisure consumption was much greater than the growth rate of per capita disposable income. There was also a clear gap between 2008 and 2014, in which the growth rate of per capita leisure consumption was much smaller than the growth rate of the disposable income. Otherwise, except for 2007 and 2009, the deviation between the two rates was within three percentage points. According to the stage division principle mentioned above, in comparing the growth rates of leisure consumption and disposable income per capita, we found that there was a large gap between 1999 and 2002, with a difference of nearly five percentage points. The gaps between 2003–2007 and 2008–2012 were small, since the growth rate of per capita leisure consumption declined, and the growth rate of per capita disposable income increased. Both rates decreased from 2013 to 2016 and retained small deviations. Comparing the growth rates of per capita leisure consumption and total consumption, a pattern emerged similar to that of the growth rate change of per capita leisure consumption and disposable income. This outcome reflects that the growth rate of total consumption and disposable income is basically the same, while that of leisure consumption is more susceptible to external factors, bringing about different changes.

3.2. Structural Evolution Analysis

The leisure consumption structure refers to the proportion of various types of leisure consumption in the total leisure consumption and the proportional relationship between the various types of leisure consumption. Concerning the content of leisure consumption, although education, culture and entertainment are the core of residents’ leisure consumption, the proportional relationship of this type of consumption with the consumption of transportation, medical needs and healthcare, daily necessities and services, and other supplies and services can still explain the reasonable level of leisure consumption structure in a certain sense.

First, the changes in the family-level Engel coefficient values should be examined. The Engel coefficient is the proportion of food expenditures to total expenditures. The smaller that the coefficient value is, the higher that the standard of living is. Conversely, the larger the coefficient value is, the lower that the standard of living is. According to the Food and Agriculture Organization of the United Nations (FAO) classification criteria, an Engel coefficient greater than 0.6 indicates poverty, a coefficient between 0.5 and 0.59 indicates comparative prosperity, a coefficient between 0.4 and 0.49 indicates being well off, a coefficient between 0.3 and 0.39 indicates being wealthy, and a coefficient less than 0.3 indicates being the wealthiest. Over the past 40 years of reform and opening up, with the continuous development of China’s economic levels and the income levels of residents, the Engel coefficient has shown a downward trend. In 2000, the Engel coefficient of China’s urban residents reached the well-off stage, and it reached the wealthiest stage by 2016. The continuous decline of the Engel coefficient of urban residents indicates that the expenditure proportions for clothing, residence and leisure consumption continue to increase. More importantly, with the improvement of social civilization and changes in consumer attitudes, the proportions of various leisure consumptions have also changed and adjusted, and the leisure consumption structure has been continuously optimized.

Second, the scales of various types of leisure consumptions have increased rapidly, specifically those of education, culture, entertainment, transportation, communication, and healthcare consumption. The education, culture and entertainment consumption of urban residents first reached ¥1000 in 2004 and exceeded ¥2000 in 2012. In 2016, transportation and communication consumption reached more than ¥3000, while the consumption of healthcare, daily necessities and services remained at approximately ¥1500. Compared with 1998, by 2016, urban residents’ traffic and communication consumption increased by 12.3 times; education, culture and entertainment consumption increased by 5.3 times; and medical needs and health consumption increased by 8 times. From the changing trends, it can be seen that transportation, communication, education, culture and entertainment consumption grew the fastest, followed by healthcare, daily necessities and services, while the consumption of other supplies and services grew at a slower rate. Additionally, the growth rates of transportation and communication consumption were faster than those of education, culture and entertainment consumption. A possible reason for this outcome is that, with the accelerated pace of people’s work lives and the rapid development of the Internet and communication technologies, the breadth and depth of social interactions among urban residents have expanded significantly, and consumption spending in this area has increased dramatically. However, the growth rate of healthcare, daily necessities and services indicates an inversely proportional relationship.

Third, the proportions of various types of leisure consumptions within both the total consumption and the overall leisure consumption have either increased or decreased, as illustrated in

Table 2. Proportion of leisure consumption in total consumption and leisure consumption.

Year	Daily Necessities and Services		Transportation & Communication		Education, Culture & Entertainment		Healthcare		Other Supplies & Service
	Total Consumption	Leisure Consumption	Total Consumption	Leisure Consumption	Total Consumption	Leisure Consumption	Total Consumption	Leisure Consumption	Total Consumption	Leisure Consumption
1998	8.24	23.55	5.94	16.97	11.53	32.95	4.74	13.54	4.55	13.00
1999	8.57	22.63	6.73	17.77	12.28	32.45	5.32	14.05	4.96	13.09
2000	8.79	21.55	7.90	19.38	12.56	30.79	6.36	15.60	5.17	12.68
2001	8.27	19.83	8.61	20.65	13.00	31.17	6.47	15.51	5.35	12.84
2002	6.45	15.28	10.38	24.62	14.96	35.48	7.13	16.91	3.25	7.70
2003	6.30	14.88	11.08	26.16	14.35	33.89	7.31	17.26	3.30	7.80
2004	5.67	13.35	11.75	27.64	14.38	33.84	7.35	17.30	3.34	7.87
2005	5.62	13.06	12.55	29.15	13.82	32.10	7.56	17.57	3.50	8.12
2006	5.73	13.19	13.19	30.36	13.83	31.84	7.14	16.42	3.56	8.19
2007	6.02	13.85	13.58	31.24	13.29	30.59	6.99	16.09	3.58	8.23
2008	6.15	14.81	12.60	30.33	12.08	29.07	6.99	16.83	3.72	8.95
2009	6.42	14.92	13.72	31.91	12.01	27.93	6.98	16.24	3.87	8.99
2010	6.74	15.42	14.73	33.68	12.08	27.63	6.47	14.80	3.71	8.47
2011	6.75	15.56	14.18	32.70	12.21	28.16	6.39	14.74	3.83	8.84
2012	6.69	15.23	14.73	33.52	12.20	27.76	6.38	14.52	3.94	8.97
2013	6.57	15.07	14.80	33.94	12.41	28.45	6.05	13.87	3.78	8.67
2014	6.18	15.71	13.21	33.59	10.73	27.29	6.54	16.63	2.67	6.79
2015	6.11	15.18	13.53	33.65	11.14	27.69	6.75	16.77	2.70	6.71
2016	6.18	15.08	13.75	33.54	11.43	27.87	7.07	17.23	2.58	6.28

. For example, the proportions of transportation and communication in both total consumption and leisure consumption have basically been on the rise, especially since they have accounted for approximately 1/3 of the overall leisure consumption since 2010. The proportion of education, culture, entertainment and healthcare consumption first increased and then decreased, but there has been a slow increase since 2014. The consumption of daily necessities, services and other supplies and services appears to be declining in the long term. More specifically, the proportion of daily necessities and services as part of total consumption has remained at approximately 6% since 2007, while the category’s proportion of the overall leisure consumption has remained at approximately 15% since 2010. Other supplies and services accounted for 2% and 6% of the total consumption and leisure consumption in 2014, respectively.

From the proportion change rates of the various types of leisure consumption as part of total consumption, the following can be seen. (1) The proportions of leisure consumption have changed greatly since approximately 2002. These changes have been mainly due to the introduction of the Golden Week system, the emergence of the holiday economy and the state’s strong support for the development of tourism, which have caused the consumption structure of urban residents to change quietly. (2) From 2002 to 2013, after a slow decline, education, culture and entertainment consumption were basically maintained at a certain level. The consumption proportions of healthcare, daily necessities and services, and other supplies and services have been relatively small. (3) Since 2014, the consumption proportions of education, culture, entertainment, transportation, and healthcare have been on the rise, while the proportions of household goods and services and other supplies and services have declined slightly.

From the proportional change rate of various types of leisure consumption out of total leisure consumption, it can be seen that: (1) from 1998 to 2001, the proportions of transportation and healthcare consumption increased, while the proportions of education, culture, entertainment, daily necessities and service consumption declined slightly; (2) in 2002, the proportions of education, culture and entertainment consumption increased significantly, while the proportion of other supplies and services consumed decreased significantly; and (3) after 2003, the consumption proportions of transportation and communication first increased rapidly and then slowly, while the proportions of education, culture and entertainment consumption declined slightly, while the consumption of medical care, daily necessities and services, and other supplies and services did not change very much. However, the proportion of healthcare consumption increased since 2014, while those of daily necessities and services and other supplies and services declined slightly.

Fourth, the marginal propensity to consume and income elasticity have gradually increased. From the mixed regression of urban residents’ leisure consumption to disposable income in 1998–2002, 2003–2007, 2008–2012 and 2013–2016, the marginal propensity to consume and income elasticity were obtained, as indicated in

Table 3. Marginal Consumption Tendency and Income Elasticity of Leisure Consumption.

	Items	1998~2002	2003~2007	2008~2012	2013~2016	Mean	Rank
Marginal Consumption Tendency	Daily necessities and services	0.128	0.232	0.314	0.323	0.249	4
	Transportation and communication	0.336	0.487	0.774	1.021	0.655	1
	Education, culture and entertainment	0.418	0.321	0.520	0.653	0.478	2
	Healthcare	0.217	0.244	0.222	0.627	0.328	3
	other supplies and service	0.099	0.135	0.178	0.060	0.118	5
Income Elasticity	Daily necessities and services	0.055	0.069	0.216	0.683	0.256	3
	Transportation and communication	0.164	0.068	0.267	1.163	0.416	1
	Education, culture and entertainment	0.103	0.040	0.180	0.695	0.254	4
	Healthcare	0.122	0.058	0.143	1.162	0.371	2
	other supplies and service	0.083	0.068	0.186	0.079	0.104	5

. Except for the category of consumption of other supplies and services, the marginal consumption propensity and income elasticity of the other four leisure consumptions reached their maximum during the period from 2013 to 2016. From the means and rankings of the marginal propensity to consumption values in each period, we can see that the transportation and communication, education culture and entertainment, and healthcare consumptions were ranked in the top three. While the income elasticity rankings of healthcare, daily necessities and services all advanced slightly, the income elasticity rankings of daily necessities, services and education, culture and entertainment were not very different. Based on the above analysis, it can be seen that with an increase in income level, the consumption expenditure proportions of transportation, communication, education, culture and entertainment, and healthcare increased significantly.

3.3. Regional Difference Analysis

To analyze regional differences, first, regional scale analysis was conducted. From 1998 to 2016, the leisure consumption of urban residents in China generally showed a catch-up trend, especially in the median provinces and cities. To more clearly examine the leisure consumption changes for urban residents, the four times nodes of 2001, 2006, 2011 and 2016 were selected according to the uniform distribution rule; with the addition of 1998, the comparison of five years of the interprovincial period of urban residents’ leisure consumption was possible. It can be seen that the leisure consumption ranges of urban residents in the five years are [975, 2863], [1426, 4284], [1737, 7610], [3172, 11,917] and [5286, 15,414]. The analysis found that: consumption data of the main cities Beijing, Shanghai, Zhejiang, Tianjin and Guangdong have always been in the dominant position. The Yangtze River Delta, the Pearl River Delta and Beijing-Tianjin-Hebei regions have relatively high levels of social and economic development and are still the most active areas for leisure consumption. More specifically, the leisure consumption scales of urban residents in Beijing and Shanghai have increased one after another, with rapid development momentum. In 2017, Beijing issued the “Implementation Opinions on Cultivating the Expansion of Service Consumption, Optimizing and Upgrading Commodity Consumption”, and Shanghai issued the “Shanghai Action Plan for Promoting New Consumption Development to Promote New Consumption (2016–2018)”, which brought further expansion of leisure consumer demand. Second, the rankings of the northeastern and northwestern provinces, such as Liaoning, Ningxia, and Xinjiang, have demonstrated a catch-up trend. In recent years, to implement the "Guiding Opinions of the State Council on Actively Giving Play to the Role of New Consumption and Accelerating the Cultivation of New Supply and New Motive Forces", many provinces and municipalities have successively released implementation plans, and leisure consumption has undergone rapid development. Third, the lower-ranking leisure consumption scales in the provinces and cities have remained basically stable, and the changes in their differences have not been obvious.

Second, spatial equilibrium analysis was conducted. The CVs for various types of leisure consumption were analyzed using STATA 12.0 software (Stata Corp LLC, College Station, TX, USA), and the results showed that the fluctuation range of the CVs was [0.269, 0.512] for daily necessities and services, [0.272, 0.425] for transportation and communication, [0.242, 0.354] for education, culture and entertainment, and [0.327, 0.430] for healthcare, other supplies and services, as demonstrated in

Table 4. Variation coefficient of various leisure consumption of urban residents from 1998 to 2016.

Year	Daily Necessities and Services	Transportation and Communication	Education, Culture and Entertainment	Healthcare	Other Supplies and Service
1998	0.448	0.375	0.331	0.291	0.335
1999	0.454	0.412	0.333	0.353	0.362
2000	0.421	0.424	0.344	0.300	0.330
2001	0.395	0.391	0.346	0.314	0.355
2002	0.292	0.386	0.366	0.354	0.327
2003	0.344	0.409	0.381	0.317	0.331
2004	0.350	0.428	0.399	0.353	0.338
2005	0.325	0.505	0.398	0.336	0.368
2006	0.329	0.494	0.423	0.352	0.359
2007	0.295	0.512	0.399	0.311	0.344
2008	0.295	0.469	0.417	0.309	0.389
2009	0.310	0.454	0.425	0.271	0.405
2010	0.327	0.432	0.416	0.267	0.389
2011	0.297	0.389	0.413	0.244	0.403
2012	0.272	0.381	0.403	0.242	0.380
2013	0.247	0.390	0.314	0.258	0.430
2014	0.232	0.299	0.277	0.290	0.383
2015	0.234	0.294	0.281	0.286	0.347
2016	0.235	0.269	0.272	0.290	0.360
Max	0.454	0.512	0.425	0.354	0.430
Min	0.232	0.269	0.272	0.242	0.327
Mean	0.321	0.406	0.365	0.302	0.365

. As seen from the averages, the maximum consumption was 0.406 for transportation and communication, and the minimum consumption was 0.302 for healthcare. That is, the consumption expenditures of transportation and communication varied greatly throughout different regions, while the regional differences in healthcare were relatively small.

The long-term consumption CVs of education, culture and entertainment, transportation and communication, daily necessities and services fluctuated greatly and declined gradually, indicating that the differences in these consumer expenditures scales gradually converged. From the perspective of time stages, the regional difference in goods and services consumption was the largest in the period before 2002, transportation and communication consumption was significantly larger in the period from 2003 to 2010, and educational, cultural and entertainment consumption was relatively large in the period from 2013 to 2016.

Third, spatial correlation analysis was conducted. Using Geoda spatial econometric software, four typical years (2001, 2006, 2011 and 2016) were selected to measure and test the Moran’s I values of urban residents’ leisure consumption.

Table 5. Moran’s I index of urban residents’ leisure consumption in the main years.

Year	Daily Necessities and Services	Transportation and Communication	Education, Culture and Entertainment	Healthcare	Other Supplies and Service
2001	0.011	0.01	0.027	0.037	0.009
2006	0.057	0.032	0.069	0.043	0.012
2011	0.065	0.058	0.069	0.055	0.019
2016	0.107	0.122	0.133	0.122	0.093

shows that the Moran’s I values for all types of leisure consumptions were positive and showed volatility, but the index level was relatively low. That is, there was a certain correlation between the various types of leisure consumption in different areas, but the degree of association was relatively small.

From the comparison of the Moran’s I values, we found that the association degrees of the healthcare, education, culture and entertainment consumptions were relatively large in 2001. In 2006, the regional correlations among education, culture, entertainment, daily necessities and service consumptions were relatively large. In 2011, the regional association degrees of the various types of leisure consumption improved. The regional correlations of the education, culture, entertainment, healthcare, transportation and communication consumptions were relatively large in 2016.

4. Discussion

Different from previous studies which focused on the characteristics and influencing factors of leisure consumption [16,17,21,31,32], this study explored the laws and trends of leisure consumption of urban residents in China through the statistical analyses of large-scale and cross-industry residents’ leisure consumption. It found that the income elasticity of medical care, daily necessities and services has increased, but the income elasticity of daily necessities and services is not different from that of education, culture and entertainment. The reason for this is the characteristics of China’s urban household consumption structure. In recent years, residents’ transportation and communication consumption and residential consumption have increased rapidly. Residential consumption has increased the burden of families, causing them to decrease the share of education and culture consumption and maintain housing expenditures by reducing long-term human capital investment [47]. However, the elasticity of developing enjoyment-oriented education and culture, medical care and other consumption expenditures is greater than 1 with enormous potential, the proportion will increase with the increase in income, and the population structure of older ages and fewer children will also promote this trend. Therefore, against the background of the improvement of population quality, the initial formation of the middle class and the slowing of population growth, urban residents are paying increasing attention to development and enjoyment consumption. Leisure consumption such as for medical treatment, education, cultural tourism, household goods, transportation and communication, will become a new consumption driving force and industrial growth point. However, from the perspective of time trends, the impact of disposable income on leisure consumption expenditures is not stable [26]. Generally, from 1998 to 2016, the regional differences in urban residents’ various types of leisure consumption fluctuated. The level of economic and social development, the degree of leisure industry development, and the supply of leisure facilities and services in all regions played important roles in the leisure behaviors and choices of urban residents [6]. This study also found that the Yangtze River Delta, the Pearl River Delta and Beijing Tianjin Hebei are the most active regions in leisure consumption. The northeast and northwest provinces have the tendency to catch up, while the gap in leisure consumption among residents in other provinces and cities is not obvious. In addition, there is a certain correlation among various leisure consumption areas of urban residents, but the degree of correlation is relatively small, while the education, culture and entertainment consumptions show a certain degree of a mutual pulling effect. Residents’ leisure consumption has obvious spatial spillover effect [7]. Atmosphere construction, leisure industry policy guidance, and leisure infrastructure construction, all of which lead to healthy growth in the leisure consumption of urban residents and enables the effective key role of leisure consumption in regional economic growth and the improvement of residents’ living standards.

5. Conclusions and Implications

5.1. Conclusions

This study explored the laws and trends of leisure consumption of urban residents in China through the statistical analysis of large-scale and cross-industry residents’ leisure consumption. The conclusions are as follows:

First, the scale of leisure consumption continues to expand, but the growth rate has gradually declined. The consumption of education, culture and entertainment, transportation and communication, and medical care is growing rapidly, and the growth trend is still obvious in the future. Second, the leisure consumption scale of the Yangtze River Delta, the Pearl River Delta and the Beijing Tianjin Hebei region has comparative advantages. The consumption level of transportation and communication among regions varies greatly, while the difference of medical and healthcare among regions is small; However, regional differences in education, culture and entertainment consumption are rapidly narrowing. Third, the regional correlation of various types of leisure consumption is not obvious, while education, culture and entertainment consumption show a certain degree of mutual pull effect.

5.2. Practical Implications

To further expand the leisure consumption needs of residents, this study proposes the following recommendations.

First, the government should constantly improve the leisure consumption environment of residents, cultivate the concepts and habits of residents’ leisure consumption, and create a market atmosphere for consumers “who want to consume and dare to consume” to realize the continuous expansion of the leisure consumption demand.

Second, the relevant departments begin to perform specialized statistical work on leisure consumption and micro-surveys of leisure consumption. The potential increase in leisure consumption still necessitates a return to the basic starting point of urban residents’ leisure consumption behavior, requiring careful determination of the needs and motives of residents’ leisure consumption and studying the changes and trends of residents’ leisure consumption behaviors.

Third, the government should do a better job of leading the residents’ cultural entertainment, education and training and healthcare consumption. This guidance should be combined with the actual development of the regions and the current state of their culture since cultural leisure is seen as a key task in industrial planning, marketing and publicity, and policy support. In addition to increasing investment in education and expanding the enrollment scale to stimulate education consumption, it is necessary to strengthen the regulations and guidance related to various types of social education and training. Additionally, the government and industries should combine their big data statistical analyses and adopt effective and targeted measures to effectively stimulate and meet the consumption demand for healthcare.

Fourth, the state should vigorously develop the urban leisure industry and effectively promote the upgrading of industrial structures. Under the premise of ensuring good momentum for national economic development and the continuous increase in residents’ income, fiscal, monetary, financial and industrial policies should provide strong support for the development of leisure-related industries, thus achieving a positive interaction among traditional leisure industries, emerging leisure industries and emerging leisure businesses.

6. Limitation and Suggestion for Further Research

The data used in this study are statistical data related to the leisure consumption of urban residents at the provincial level in China. The use of data has certain limitations, and to some extent, the data cannot fully show the whole picture of residents’ leisure consumption. How can we clearly define people’s leisure consumption behavior? How can we reasonably determine the boundary between people’s leisure consumption expenditure and other industries? How can we generate reasonable statistics about people’s leisure consumption expenditure? How can we conduct more scientific qualitative and quantitative research on people’s leisure consumption behaviors and their laws? These problems must be solved in the future research on leisure consumption by urban residents in China.

In terms of the research methods, this study mainly uses marginal propensity to consume and income elasticity to analyze the change of leisure consumption structure of urban residents in China and uses coefficient of variation and global Moran index to investigate the spatial balance and interdependence of leisure consumption of urban residents. Although it is appropriate, ESRI’s hotspot analysis, Mann–Kendall’s test and other spatial analysis methods should be considered for scientific and in-depth research in subsequent research.

In addition, leisure consumption is an obvious indicator of people’s leisure lives. However, through leisure consumption, it is more meaningful to explore changes in people’s leisure lifestyles, which could be an important direction for future research. The data used in this article were Chinese provincial-level statistics on leisure consumption by urban residents. To some extent, they do not fully show the overall picture of residents’ leisure consumption. In addition to the direct impact of residents’ income supply factors, such as industrial structures, will also have an important impact on leisure consumption, requiring special research. China is a typical urban–rural dual society. There is a large difference between the urban and rural residents. This study only included analysis of leisure consumption of urban residents. The leisure consumption of rural residents in China is also an important issue worthy of study in the future.