Regional Investment Preferences and Corporate Cash Holdings: Evidence from China

Abstract

Based on the sample of Chinese A-share listed companies from 2006 to 2020, this paper studies the impact of regional investment preference on corporate cash holdings. We find that the stronger the investment preference in the region where the company is located, the higher the level of the company’s cash holdings. After a series of robustness tests, such as the instrumental variable method and propensity score matching method, the main regression results of this paper remain unchanged. The mechanism test shows that the regional investment preference can increase the preventive cash holding and strategic cash holding motivation of enterprises so as to improve the cash holding level of companies. Further research shows that strict internal control and external supervision can weaken the impact of regional investment preference on corporate cash holdings. This paper not only enriches the research on the factors affecting enterprise cash holding but also deepens the theoretical cognition of the impact of the regional environment on corporate governance.

1. Introduction

Compared with other assets, cash is more liquid and flexible. It plays a very important role in company management and is the “blood” for the survival and development of an enterprise. Many entrepreneurs believe that “cash is king”. A reasonable level of cash holdings is crucial for companies, which can not only reduce the company’s cost of capital but also avoid some risks that may occur in the future [1]. When the company is short of cash, it may lead to the break of the capital chain, making the company fall into financial difficulties, and may even lead to bankruptcy of the company [2]. When an enterprise holds too much cash, it will not only produce higher opportunity costs but also facilitate the abuse of funds by the management [3]. The level of cash holdings of a company may directly affect the company’s liquidity risk management and asset allocation, and may also directly affect the company’s investment capacity and financing costs, and ultimately affect the company’s profitability and market value [4,5].

At present, the research on the level of corporate cash holdings has been gradually improved, from the motivation of cash holdings, the influencing factors to the value effect of cash holdings, and then to its competitive effect, the research perspective ranges from the individual characteristics of enterprises and corporate governance to industry characteristics [6,7,8]. A large number of the classic literature have emerged, but there is little research on whether the regional environment of the company will affect the company’s cash holdings, especially whether the regional investment preference affects the level of cash holdings of the company.

Regional investment preference refers to the degree to which a region is keen on investment. Its calculation method is the ratio of the annual per capita investment amount of a region to the per capita income of the region in that year. It measures the proportion of income that people in a region are willing to invest. Upper echelons theory holds that the value orientation and behavioral preferences of corporate management are largely affected by the growth environment and cultural soil and will be reflected in corporate governance. As an informal system, the culture of the region where the enterprise is located will influence and consolidate people’s cognition, preferences, decision-making, and behavior [9]. Some scholars have found that religious belief can deepen people’s aversion to risk. When a company is more influenced by religion, its risk-taking level is lower [10]. Some scholars have also found that in areas highly influenced by Confucian culture, the proportion of female directors is lower [11]. As an important informal system, the regional investment preference will also affect the governance and operation of listed companies in that region. When investment preference in a region is high, will local companies choose to increase investment levels and hold less cash, or will they be conservative and invest less and hold more cash?

The contributions of this paper are mainly reflected in the following aspects: firstly, this paper supplements the relevant literature on the influence of informal institutions on micro-firm behavior. As a very important informal system, regional investment preference can have a profound impact on the behavior of enterprises in the region, which may affect the investment efficiency, risk-taking, and cash holdings of enterprises. However, few scholars have studied the relevant research of regional investment preference on corporate governance. Secondly, this paper expands the existing literature on corporate cash holdings. Most of the previous literature studies the key factors affecting cash holdings from the perspective of the company and the institution [12,13]. This paper breaks through the traditional theoretical framework of contracts and institutions and studies the impact of regional investment preferences on corporate cash holdings from the perspective of informal institutions. Its mechanism of action provides a new dimension of explanation for the level of cash holdings of enterprises. Finally, this paper provides a new method to measure the level of regional investment. Most scholars study the regional investment preference based on the per capita lottery transaction volume in each region, but the small lottery amount cannot truly reflect the regional investment preference [14,15]. This paper uses the ratio of the total per capita investment trading volume of the region (stocks, funds, bonds, and futures, etc.) to the per capita GDP, which can more objectively reflect the investment preference of the region.

2. Literature Review and Research Hypotheses

2.1. Literature Review

The cash held by the company is not only an important liquid asset of the company but also an important financial decision of the company. On the one hand, because the capital market is not perfect, companies with sufficient cash can avoid expensive external financing costs, capture emerging investment opportunities and prevent financial crises. On the other hand, the high cash holding level of enterprises will also make the company’s management seek private interests or major shareholders infringe on the interests of minority shareholders. Therefore, it has attracted the attention of scholars for a long time. Some researchers find that the level of cash holdings in countries with poor investor legal protection is twice that of countries with better protection [8]. Liu et al. argue that the controlling shareholders’ meetings hollow out through a large amount of cash, which harms the interests of small and medium investors [16]. Bates et al. document that the greater the uncertainty risk of corporate cash flow, the more inclined to hold cash [17]. Mikkelson et al. argue that persistent high cash holdings are an optimal choice for companies based on intrinsic characteristics, and their operating performance is significantly higher than that of companies with low cash holdings [18]. Fresard finds that when there is a risk in the market, holding sufficient cash flow can gain a competitive advantage in the market [19].

The culture of the region where the enterprise is located can be deeply imprinted in the thinking style and behavioral norms of social members, thus affecting their investment decisions [9]. Some researchers have found that individualism has a positive impact on corporate risk-taking, while uncertainty avoidance and harmonyism have a negative impact on risk-taking [12]. Chen et al. argue that countries with stronger individualism generally have fewer cash holdings, while countries with stronger uncertainty avoidance have higher levels of cash holdings [20]. Gh et al. find that the stronger the religious atmosphere in the region where the company is located, the lower its R&D expenditure and the higher its cash holdings [10]. Christensen et al. find that when companies are located in areas with strong gaming preferences, company management prefers to take speculative and high-risk projects, which are more prone to stock price crashes [14]. Kumar et al. document that regional gaming preferences are related to investors’ stock choices, employee compensation plans, and IPOs [21]. Ji et al. take the per capita stock trading volume as the proxy variable of gambling preference [15]. The study finds that the stronger the gambling preference, the higher the probability of a stock price crash, mainly because the company’s management will adopt speculative accounting practices. Chen et al. argue that companies that are greatly affected by gambling preference will invest more money in innovation and obtain more output [22]. Callen and Fang (2020) find that when the gambling preference of the region where the company is located is large, the company will pay higher fees to the auditor [23]. Ma et al. document that when the company’s regional gambling preference is large, it will increase the cost stickiness caused by managers’ agency or behavioral bias [24].

2.2. Research Hypothesis

The investment level of a region will affect the investment decisions of individuals and the financial decisions of Companies in the region. Some researchers have found that in areas with high gambling preferences, investors show a stronger tendency to hold lottery tickets and stocks, stock option plan for corporate employees is more popular, the first-day return after an initial public offering is higher, and the premium of lottery stocks is greater [25]. Chen et al. document that companies with strong gambling preferences tend to take on more risky projects. Some researchers have found that in regions with strong gambling preferences, the management of companies in the region is more likely to take financial reporting risks, thereby increasing the possibility of financial restatements [22]. Ji et al. argue that companies in regions with stronger gambling preferences have a higher probability of stock price crashes, mainly because company management will adopt aggressive corporate strategies and speculative accounting practices [15].

China’s capital market starts late and is dominated by retail investors, and the phenomenon of chasing up and down is more serious. Therefore, this immature investment sentiment will affect the investment efficiency and cash holdings of local listed companies. As Christensen et al. (2018) study, local investment preferences magnify the original risk tolerance of listed company management [14]. When the investment preference of a region is stronger, the management of listed companies in the region will increase investment, which will lead to a decrease in the company’s cash holdings [26]. Therefore, we propose the first hypothesis of this paper:

H1a. 

Regional investment preference is negatively related to corporate cash holdings.

This paper believes that there may be another possibility. When investment is prevalent in the region where the company is located, company executives will think that external risks are high, so they will be relatively conservative and have the idea of being vigilant in times of peace, thereby reducing foreign investment and holding more cash. The reason why the company holds more cash is that, on the one hand, it is preventive cash holding, and the company needs to hold more cash to avoid falling into financial difficulties. On the other hand, it is a strategic cash holding. When the company suddenly has a better strategic opportunity, the enterprise has more cash to maintain strategic flexibility so as to seize the opportunity and make the enterprise bigger and stronger.

Firstly, regional investment preferences will enhance the preventive cash-holding motivation of companies. Due to the imperfection of the capital market and external adverse shocks such as the financial crisis, enterprises often face capital chain breakage and financial crisis. As the “blood” for the survival of enterprises, cash is an important tool to ensure corporate liquidity and prevent financial risks [17]. Therefore, in order to avoid a financial crisis, companies will hold large amounts of cash on a precautionary basis. Studies have shown that the culture of a company’s region has a significant impact on the level of risk-taking and cash holdings [12]. When the investment preference in the region where the company is located is relatively large, the company’s management will be relatively conservative and be prepared for danger in times of safety. This sense of urgency will affect the risk awareness of corporate executives, which may lead companies to reduce risky investment projects and hold more cash on a precautionary basis to ensure financial security and avoid falling into a financial crisis.

Secondly, the regional investment preference will enhance the company’s strategic cash-holding motive. In the face of a complex and changeable market competition environment, slow and rigid strategic decisions will bring huge losses to enterprises, and maintaining strategic flexibility will help enterprises to quickly adjust strategic actions according to changes in external conditions [27]. Some scholars have found that enterprises holding sufficient cash and maintaining flexibility can deal with the risk exposure from product market competition and help them implement a more favorable market competition strategy [19]. At the same time, when the company holds a relatively high level of cash, it can signal to its rivals to expand production capacity, increase its sales network, or even initiate mergers and acquisitions, thereby forming a deterrent to vicious external competition or potential entrants so as to win and maintain a dominant position in market competition for a long time. When the investment preference is high in the region where the company is located, the impact of this environment will lead entrepreneurs to pay attention to the long-term survival and development of the company, especially since the market currently faced by Chinese companies is characterized by high growth and uncertainty. Holding more cash and maintaining strategic flexibility enable companies to seize market opportunities in a timely manner. Therefore, we propose another hypothesis for this paper:

H1b. 

Regional investment preference is positively correlated with corporate cash holdings.

3. Research Design

3.1. Sample Selection and Data Sources

This paper takes the data of China’s Shanghai and Shenzhen A-share listed companies from 2006 to 2020 as the sample source and screens the data as follows: excluding the samples of financial and insurance companies, ST and foreign-controlled companies, and excluding the samples of companies with some missing data, and finally, obtaining 30,700 sample observations. The core variable “regional investment preference” data is collected and sorted manually, and the financial data of other companies are from Csmar and Wind databases. In order to avoid the influence of extreme values, all continuous variables are subject to tailing treatment at the upper and lower 1% levels.

3.2. Model Setting and Variable Definition

To test the impact of regional investment preferences on corporate cash holdings, this paper constructs the following models:

Cash_i,t = α₀ + α₁Reinvest_i,t−1 + α_jControl_i,t−1 + Industry + Year + ε_i,t

(1)

3.2.1. Dependent Variable

Referring to the research of Opler et al. and Dittmar et al., the company’s cash holdings are measured in the following two ways: (1) the ratio of cash and cash equivalents to total assets (Cash 1); (2) the ratio of cash and cash equivalents to net assets (Cash 2), net assets equal to total assets minus cash and cash equivalents [2,8].

3.2.2. Independent Variable

This paper uses two approaches to measure regional investment preferences. The first method is to use the ratio of per capita total trading volume to per capita GDP in each province (Reinvest 1), where the total transaction value includes stocks, funds, bonds, and futures, etc. The second method is to use the ratio of per capita stock trading volume to per capita GDP in each province (Reinvest 2).

3.2.3. Other Variables

Referring to the existing literature, the control variables in this paper include corporate finance and governance variables such as enterprise size (Size), asset-liability ratio (Lev), operating cash flow (Cf), net working capital (Nwc), capital expenditure (Capex), sales growth rate (Growth), dividend payout ratio (Div), the shareholding ratio of the largest shareholder (Top 1), board size (Board), and the ratio of independent directors (Indratio), as well as annual and industry fixed effects.

Table 1. Variable table.

Variable	Definition
Dependent variable
Cash 1	The ratio of cash and cash equivalents to total assets
Cash 2	The ratio of cash and cash equivalents to net assets
Independent variable
Reinvest 1	The ratio of per capita total trading volume to per capita GDP in each province
Reinvest 2	The ratio of per capita stock trading volume to per capita GDP in each province
Control variable
Size	The logarithm of the company’s total assets at the end of the previous year
Lev	The ratio of total liabilities to total assets at the end of the previous year
Cf	The remaining cash flow of the profits obtained by the enterprise from operating activities after deducting the costs related to long-term investment and investment in securities.
Nwc	The balance of the total current assets of the enterprise minus various current liabilities
Capex	Cash paid for acquisition and construction of fixed assets, intangible assets, and other long-term assets
Growth	Forecast annual sales growth rate
Div	Dividends distributed to shareholders as a percentage of the Company’s profits
Top1	Shareholding ratio of the largest shareholder
Board	The logarithm of the number of board members
Indratio	The ratio of independent directors

shows all variables and their definitions in this paper.

3.3. Descriptive Statistics and Correlation Analysis

Table 2. Descriptive statistics of main variables.

Variables	N	Mean	St. Dev.	Min	25%	Median	75%	Max
Cash 1	30,700	0.2003	0.1485	0.0138	0.0951	0.1574	0.2627	0.7151
Cash 2	30,700	0.3193	0.3980	0.0132	0.1049	0.1865	0.3560	2.4917
Reinvest 1	30,700	0.0942	0.0167	0.0001	0.0010	0.0027	0.1010	0.7551
Reinvest 2	30,700	0.0312	0.0043	0.0001	0.0005	0.0014	0.0350	0.2323
Size	30,700	22.0429	1.2852	19.6934	21.1065	21.8580	22.7741	26.0480
Lev	30,700	0.4310	0.2083	0.0508	0.2648	0.4265	0.5893	0.8965
Cf	30,700	0.0651	0.1049	−0.2291	0.0087	0.0575	0.1147	0.4479
Nwc	30,700	0.1968	0.2301	−0.4674	0.0483	0.1958	0.3428	0.7990
Capex	30,700	2.7356	3.2240	0.0158	0.8169	1.6904	3.3357	19.3012
Growth	30,700	0.1472	0.3307	−0.5916	−0.0242	0.1056	0.2613	1.7143
Div	30,700	0.2727	0.3443	−0.4223	0	0.2091	0.3903	1.9937
Top 1	30,700	35.0549	14.9948	8.7700	23.2300	33.0400	45.3200	74.6600
Board	30,700	2.1460	0.2019	1.6094	1.9459	2.1972	2.1972	2.7081
Indratio	30,700	0.3715	0.0521	0.3000	0.3333	0.3333	0.4286	0.5714

shows the descriptive statistical results of the main variables. Among them, the variable mean values of cash holding level Cash 1 and Cash 2 are 0.2003 and 0.3193, respectively, and the median values are 0.1574 and 0.1865, respectively, which indicates that there is an obvious phenomenon of “high cash holding” in China’s listed companies. The standard deviations of Cash 1 and Cash 2 are 0.1485 and 0.3980, respectively, indicating that there is a large gap in the level of cash holdings among different companies. The results in Table 1 also show that the mean values of regional investment preferences Reinvest 1 and Reinvest 2 are 0.0942 and 0.0312, respectively, and the standard deviations are 0.0167 and 0.0043, respectively. It can be seen that there are indeed large differences in investment preferences in different regions. The distribution of the remaining variables in the table is basically consistent with the results of existing studies, and all are within a reasonable range.

4. Analysis and Discussion of Empirical Results

4.1. Regional Investment Preference and Corporate Cash Holdings: The Main Regression Test

Table 3. Regression results of regional investment preference and corporate cash holdings.

	(1)	(2)	(3)	(4)
	Cash 1	Cash 1	Cash 2	Cash 2
Reinvest 1	0.4753 *** (6.48)		0.7599 *** (3.07)
Reinvest 2		1.7611 *** (6.29)		2.6687 *** (2.82)
Size	−0.0094 *** (−9.13)	−0.0094 *** (−9.13)	−0.0290 *** (−8.33)	−0.0289 *** (−8.32)
Lev	−0.2042 *** (−32.26)	−0.2037 *** (−32.15)	−0.7653 *** (−35.83)	−0.7647 *** (−35.77)
Cf	0.1034 *** (19.37)	0.1033 *** (19.36)	−0.9085 *** (−50.46)	−0.9087 *** (−50.47)
Nwc	−0.0354 *** (−8.76)	−0.0353 *** (−8.75)	−0.4677 *** (−34.33)	−0.4676 *** (−34.32)
Capex	−0.0000 (−0.76)	−0.0000 (−0.75)	−0.0003 (−1.47)	−0.0003 (−1.47)
Growth	0.0006 (0.70)	0.0006 (0.68)	0.0039 (1.29)	0.0039 (1.28)
Div	0.0003 (1.53)	0.0003 (1.53)	0.0008 (1.42)	0.0008 (1.42)
Top 1	0.0004 *** (6.11)	0.0004 *** (6.05)	0.0014 *** (5.99)	0.0014 *** (5.97)
Board	0.0053 (0.85)	0.0055 (0.89)	0.0005 (0.02)	0.0008 (0.04)
Indratio	0.0127 (0.58)	0.0145 (0.66)	0.0068 (0.09)	0.0096 (0.13)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
R2_adj	0.1072	0.1071	0.1554	0.1554
N	30,700	30,700	30,700	30,700

The numbers in parentheses are the t values of the variable; *** indicate significance at the 1% levels.

shows the multiple regression results of regional investment preference and corporate cash holdings according to model (1). Columns (1) and (2) are the regression results of using Cash 1 as the measurement index of enterprise cash holdings. The regression coefficients of Reinvest 1 and Reinvest 2 are 0.4753 and 1.7611, respectively, which are significant at the 1% level. The above results support the previous hypothesis H1b, indicating that the stronger the investment preference in the region where the company is located, the higher the company’s cash holding level. Columns (3) and (4) are the regression results where the explanatory variables are replaced by Cash 2, and the conclusions are consistent with this.

4.2. Robustness Test

Although the dependent variable lags behind the independent variable in the regression model of this paper, which has solved the interference of the endogenous problem to the research conclusion to some extent, in order to consider the robustness, this paper will use the following methods to further solve the endogenous problem of this paper.

4.2.1. Two-Stage Least Squares (2SLS)

There is no reverse causality between regional investment preferences and corporate cash holdings, but there may still be endogeneity problems caused by missing variables. In order to solve the possible endogeneity problems, in this paper, two-stage least squares (2SLS) is used to alleviate this endogeneity problem. The ideal instrumental variable should be related to regional investment preferences but not to corporate cash holdings. Therefore, in this paper, we select the number of all securities business halls (Num_hall) in each region as an instrumental variable. The more the number of securities business offices in each region, the more convenient the investment of residents in the region will be, and the greater the investment preference of the region will be. However, there is no evidence that the number of securities business offices will directly affect the level of cash holdings of enterprises, indicating that the selection of the instrumental variable is more reasonable.

Table 4. Regression results of regional investment preference and corporate cash holdings after 2SLS.

	(1)	(2)	(3)	(1)	(2)	(3)
	Panel A: First Stage Regression	Panel B: Second Stage Regression		Panel A: First Stage Regression	Panel B: Second Stage Regression
	Reinvest 1	Cash 1	Cash 2	Reinvest 2	Cash 1	Cash 2
Num_hall	0.0327 *** (18.67)			0.0100 *** (21.94)
Reinvest 1		2.8709 *** (4.09)	9.0496 *** (3.81)
Reinvest 2					9.3562 *** (4.11)	29.4926 *** (3.83)
Size	0.0009 *** (10.84)	−0.0114 *** (−9.51)	−0.0359 *** (−8.86)	0.0002 *** (11.32)	−0.0111 *** (−9.58)	−0.0350 *** (−8.93)
Lev	−0.0040 *** (−8.25)	−0.1932 *** (−26.87)	−0.7273 *** (−29.95)	−0.0013 *** (−10.68)	−0.1920 *** (−26.35)	−0.7235 *** (−29.39)
Cf	−0.0004 (−0.94)	0.1039 *** (19.15)	−0.9066 *** (−49.47)	−0.0001 (−0.57)	0.1034 *** (19.16)	−0.9083 *** (−49.83)
Nwc	0.0008 ** (2.53)	−0.0378 *** (−9.08)	−0.4761 *** (−33.85)	0.0002 ** (2.02)	−0.0371 *** (−9.01)	−0.4738 *** (−34.08)
Capex	−0.0000 (−0.18)	−0.0000 (−0.78)	−0.0003 (−1.48)	−0.0000 (−0.50)	−0.0000 (−0.73)	−0.0003 (−1.44)
Growth	−0.0001 * (−1.81)	0.0010 (1.03)	0.0050 (1.61)	−0.0000 (−1.25)	0.0008 (0.88)	0.0046 (1.47)
Div	−0.0000 (−0.52)	0.0003 (1.55)	0.0009 (1.45)	−0.0000 (−0.71)	0.0003 (1.58)	0.0009 (1.47)
Top1	0.0003 *** (4.99)	0.0004 *** (4.89)	0.0012 *** (4.75)	0.0000 *** (6.61)	0.0004 *** (4.74)	0.0012 *** (4.61)
Board	−0.0014 *** (−2.96)	0.0097 (1.50)	0.0158 (0.72)	−0.0005 *** (−4.05)	0.0104 (1.60)	0.0180 (0.82)
Indratio	0.0024 (1.42)	0.0076 (0.34)	−0.0110 (−0.14)	−0.0003 (−0.77)	0.0178 (0.80)	0.0212 (0.28)
Industry	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes
R2_adj	0.1859	0.0774	0.1257	0.2419	0.0870	0.1344
N	30,700	30,700	30,700	30,700	30,700	30,700
Shea’s Partial-R2	0.0315			0.0341
Cragg-Donald Wald F-statistic	324.2156 (p = 0.0000)			421.5247 (p = 0.0000)
Sargan test		0.2354 (p = 0.3125)	0.3047 (p = 0.3245)		0.3401 (p = 0.3608)	0.4212 (p = 0.3514)

The numbers in parentheses are the t values of the variable; *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

shows the results obtained by two-stage least squares (2SLS) regression. Columns 1–3 are the regression results using Reinvest 1 as a measure of regional investment preference. In the first stage, the regression coefficient between Reinvest 1 and the instrumental variable Num_hall is 0.0327, which is significant at the 1% level, indicating that the more the number of securities business offices in the region, the stronger the investment preference in the region. In the second stage regression, the regression coefficients of Cash 1, Cash 2, and Reinvest 1 are all significantly positive and significant at the 1% level, indicating that after controlling for endogeneity, regional investment preference will still significantly increase the level of corporate cash holdings. In addition, the F-statistic is 324.2156, which is significantly larger than the critical value of 10. Therefore, the weak instrumental variable hypothesis is rejected. At the same time, according to the p value after the Sargan test, there is no over-identification problem of tool variables. Columns 4–6 are the regression results using Reinvest 2 as a measure of regional investment preference, and the same is the conclusion.

4.2.2. Propensity Score Matching (PSM)

In this paper, there may be systematic differences in regions with higher or lower investment preferences. In order to more directly compare similar companies with different investment preferences, we use the propensity score matching method to re-examine the results obtained above. We define the control group as those companies in the top quartile with the strongest regional investment preference each year and the control group as those companies in the bottom quartile with the weakest regional investment preference each year. After matching according to the processing indicator variables, there is no significant difference in the core control variables between the control group and the processing group. The regression results after matching are shown in

Table 5. Regression results of regional investment preference and corporate cash holdings after PSM.

	(1)	(2)	(1)	(2)
	Cash 1	Cash 2	Cash 1	Cash 2
Reinvest 1	0.5162 *** (4.87)	0.8671 *** (5.38)
Reinvest 2			2.1001 *** (5.22)	3.3715 *** (5.50)
Size	−0.0101 *** (−5.17)	−0.0225 *** (−7.59)	−0.0103 *** (−5.30)	−0.0226 *** (−7.65)
Lev	−0.1734 *** (−13.50)	−0.3950 *** (−20.19)	−0.1691 *** (−13.25)	−0.3825 *** (−19.68)
Cf	0.2511 *** (14.71)	0.8916 *** (34.29)	0.2551 *** (15.14)	0.9047 *** (35.25)
Nwc	−0.0093 (−1.00)	0.0765 *** (5.41)	−0.0074 (−0.84)	0.0906 *** (6.79)
Capex	0.0005 (1.12)	0.0009 (1.37)	0.0007 (1.41)	0.0013 * (1.84)
Growth	0.0018 (1.18)	0.0033 (1.44)	0.0017 (1.17)	0.0031 (1.38)
Div	0.0006 (0.38)	0.0020 (0.90)	0.0006 (0.42)	0.0022 (0.94)
Top1	0.0003 * (1.94)	0.0002 (1.13)	0.0002 * (1.73)	0.0002 (0.82)
Board	0.0072 (0.60)	0.0105 (0.58)	0.0071 (0.60)	0.0081 (0.45)
Indratio	−0.0043 (−0.10)	0.0489 (0.77)	−0.0064 (−0.15)	0.0423 (0.67)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
R2_adj	0.0771	0.2129	0.0774	0.2157
N	5413	5413	5413	5413

The numbers in parentheses are the t values of the variable; * and *** indicate significance at the 10% and 1% levels.

. The results show that the coefficient between the regional investment preference and the company’s cash holding level is significantly positive at the 1% level, indicating that the conclusion of this paper is still valid after the propensity score matching method.

4.2.3. Eliminate the Interference of Regional Economic Development Level

The results of this paper may also exist in the following situations; that is, regions with strong regional investment preferences are more advanced in economic and financial development, resulting in easier financing for enterprises and more cash holdings. Therefore, it may not be the regional investment preferences that play a role in it. In order to eliminate the interference of this factor, this paper removes the samples of listed companies in the three regions with the strongest and weakest regional investment preferences each year.

Table 6. Regression results of regional investment preference and corporate cash holdings after excluding regional economic development levels.

	(1)	(2)	(3)	(4)
	Cash 1	Cash 1	Cash 2	Cash 2
Reinvest 1	0.4670 *** (5.28)		0.7830 *** (2.71)
Reinvest 2		1.6092 *** (4.72)		2.5790 *** (2.32)
Size	−0.0104 *** (−7.66)	−0.0104 *** (−7.65)	−0.0412 *** (−9.31)	−0.0412 *** (−9.31)
Lev	−0.1885 *** (−23.65)	−0.1880 *** (−23.57)	−0.7894 *** (−30.34)	−0.7888 *** (−30.29)
Cf	0.0777 *** (12.73)	0.0776 *** (12.72)	−1.1211 *** (−56.27)	−1.1211 *** (−56.27)
Nwc	−0.0405 *** (−7.89)	−0.0404 *** (−7.88)	−0.6520 *** (−38.91)	−0.6519 *** (−38.90)
Capex	−0.0001 (−1.03)	−0.0001 (−1.03)	−0.0004 ** (−1.98)	−0.0004 ** (−1.98)
Growth	−0.0004 (−0.34)	−0.0004 (−0.36)	0.0024 (0.66)	0.0023 (0.65)
Div	0.0005 (1.13)	0.0005 (1.13)	0.0020 (1.44)	0.0020 (1.45)
Top1	0.0005 *** (5.33)	0.0005 *** (5.32)	0.0016 *** (5.42)	0.0016 *** (5.41)
Board	0.0092 (1.15)	0.0095 (1.18)	0.0227 (0.87)	0.0231 (0.88)
Indratio	−0.0111 (−0.39)	−0.0090 (−0.31)	0.0266 (0.28)	0.0302 (0.32)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
R2_adj	0.0789	0.0787	0.2104	0.2104
N	22,551	22,551	22,551	22,551

The numbers in parentheses are the t values of the variable; ** and *** indicate significance at the 5% and 1% levels.

shows the results after the regression. The regression coefficients of the independent variables are all significantly positive at the 1% level, indicating that after excluding the regional economic development factors, the regional investment preference and the company’s cash holdings are still significantly positively correlated. The robustness of the previous conclusions is verified.

4.2.4. Eliminate the Interference of Local Residents Investing in Local Listed Companies

The results of this paper may also exist in the following situations, investors in the region are relatively familiar with the local listed companies, so they invest more funds in the local listed companies, resulting in an increase in the company’s cash holdings. To rule out this factor, we use Exreinvest to measure regional investment levels. Exreinvest is the ratio of per capita regional investment minus per capita regional investment in stocks and corporate bonds to per capita GDP.

Table 7. Regression results of regional investment preference and corporate cash holdings after excluding the interference of local residents investing in local listed companies.

	(1)	(2)
	Cash 1	Cash 2
Exreinvest	0.5836 *** (5.17)	0.9915 *** (2.69)
Size	−0.0104 *** (−7.65)	−0.0412 *** (−9.31)
Lev	−0.1888 *** (−23.69)	−0.7899 *** (−30.36)
Cf	0.0777 *** (12.73)	−1.1210 *** (−56.27)
Nwc	−0.0405 *** (−7.88)	−0.6519 *** (−38.91)
Capex	−0.0001 (−1.03)	−0.0004 ** (−1.98)
Growth	−0.0004 (−0.34)	0.0024 (0.66)
Div	0.0005 (1.13)	0.0020 (1.44)
Top1	0.0005 *** (5.35)	0.0016 *** (5.43)
Board	0.0090 (1.12)	0.0224 (0.86)
Indratio	−0.0118 (−0.41)	0.0256 (0.27)
Industry	Yes	Yes
Year	Yes	Yes
R2_adj	0.0789	0.2104
N	30,700	30,700

The numbers in parentheses are the t values of the variable; **, and *** indicate significance at the 5% and 1% levels, respectively.

shows the regression results of this method. The results show that the regression coefficient between regional investment preference and company holding level is significantly positive at the 1% level, which verifies the robustness of the previous conclusions.

5. Mechanism Analysis

According to the theoretical derivation of the previous hypothesis H1b, we believe that regional investment preferences can significantly enhance the preventive cash holding and strategic cash holding motivation of enterprises, thereby improving the level of corporate cash holdings. Next, we focus on examining these two action pathways to reveal the mechanism behind them.

5.1. Preventive Motivation

When companies are in a highly uncertain environment, companies usually have a stronger preventive cash-holding motivation to deal with the current crisis. When the investment is prevalent in the region where the company is located, the company’s executives will think that the external risk is high, resulting in the idea of thinking about danger in times of peace. Then they will be relatively conservative. In order to solve various uncertain events that the company may face, the company will choose to hold more cash. Based on this, this paper adopts the economic policy uncertainty index constructed by Baker et al. to measure the uncertainty environment faced by enterprises and divides the samples into a high-uncertainty group and a low-uncertainty group according to the median value [28]. The results of the regression are shown in

Table 8. Regression results of regional investment preference and corporate cash holdings when economic policy is uncertain.

	(1)	(2)	(3)	(4)
	Cash 1	Cash 1	Cash 2	Cash 2
	Low Uncertainty	High Uncertainty	Low Uncertainty	High Uncertainty
Reinvest 1	0.4305 *** (3.83)	0.5305 (1.36)	0.4477 *** (4.22)	0.4564 (1.53)
Size	−0.0388 *** (−10.41)	−0.0090 * (−1.69)	−0.0102 *** (−5.43)	−0.0082 *** (−8.53)
Lev	−0.3090 *** (−12.69)	−1.0224 *** (−34.60)	−0.1674 *** (−13.26)	−0.2244 *** (−41.38)
Cf	1.0487 *** (40.31)	−1.9671 *** (−97.74)	0.1902 *** (15.44)	0.0726 *** (17.24)
Nwc	0.0858 *** (5.82)	−0.9080 *** (−41.10)	0.0109 (1.51)	−0.0780 *** (−19.93)
Capex	−0.0001 (−0.32)	−0.0004 ** (−2.01)	−0.0002 (−0.78)	−0.0000 (−0.68)
Growth	0.0053 * (1.71)	0.0031 (0.59)	0.0007 (0.29)	0.0007 (1.00)
Div	0.0002 (0.31)	0.0042 * (1.68)	0.0002 (0.73)	0.0006 ** (2.03)
Top1	0.0007 *** (2.65)	0.0013 *** (3.69)	0.0005 *** (3.84)	0.0003 *** (4.86)
Board	0.0225 (0.92)	0.0335 (1.20)	0.0007 (0.06)	0.0121 ** (2.29)
Indratio	0.0963 (1.13)	0.0321 (0.32)	−0.0030 (−0.07)	0.0336 * (1.74)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
R2_adj	0.1460	0.5774	0.0757	0.2227
N	14,205	16,495	14,205	16,495

The numbers in parentheses are the t values of the variable; *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

. When the uncertainty faced by the company is high, the relationship between regional investment preference and the company’s cash holdings is not significant, while when the uncertainty faced by the company is low, the relationship between regional investment preference and company cash holdings is significantly positive at the level of 1%, indicating that regional investment preferences have enhanced the company’s sense of urgency and it will hold more cash to defend against possible future risks.

5.2. Strategic Current Motivation

High-growth companies often have more investment opportunities and growth space, so they will improve their strategic cash holding level in order to seize market opportunities and win strategic advantages. When the investment preference is strong in the region where the company is located, the strategic cash-holding motivation of the company’s executives will be stronger, and they will hold more cash. When the company is in a high-growth industry with more investment opportunities, the company’s cash-holding behavior will be more obvious. Based on this, this paper divides the sample into a low-growth industry group and a high-growth industry group according to the median of the Tobin Q value of each industry in each year. The results of the regression are shown in

Table 9. Regression results of regional investment preference and corporate cash holdings when companies are in different growth industries.

	(1)	(2)	(3)	(4)
	Cash 1	Cash 1	Cash 2	Cash 2
	High Growth	Low Growth	High Growth	Low Growth
Reinvest 1	0.5931 *** (5.36)	0.6731 (0.98)	0.4883 *** (4.75)	0.5075 (1.32)
Size	−0.0148 *** (−8.73)	−0.0071 *** (−6.98)	−0.0325 *** (−6.06)	−0.0415 *** (−7.99)
Lev	−0.1560 *** (−14.66)	−0.1448 *** (−22.19)	−0.4961 *** (−14.73)	−0.1941 *** (−5.83)
Cf	0.3088 *** (18.22)	0.4095 *** (35.62)	0.8096 *** (15.09)	1.3820 *** (23.56)
Nwc	0.0142 * (1.67)	−0.0282 *** (−5.19)	−0.0831 *** (−3.08)	−0.0530 * (−1.91)
Capex	−0.0006 (−1.26)	0.0024 *** (4.88)	−0.0033 ** (−2.38)	0.0041 * (1.65)
Growth	0.0042 (0.85)	0.0042 (1.27)	−0.0028 (−0.18)	0.0540 *** (3.23)
Div	0.0197 *** (3.89)	0.0158 *** (5.35)	0.0468 *** (2.93)	0.0120 (0.80)
Top1	0.0004 *** (3.27)	0.0001 ** (2.10)	0.0005 (1.51)	0.0003 (0.70)
Board	−0.0003 (−0.03)	0.0070 (1.14)	−0.1003 *** (−3.21)	0.0610 * (1.94)
Indratio	−0.0162 (−0.45)	0.0520 ** (2.30)	−0.1719 (−1.51)	0.2924 ** (2.53)
Industry	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes
R2_adj	0.0994	0.2588	0.0816	0.0916
N	18,254	12,446	18,254	12,446

The numbers in parentheses are the t values of the variable; *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

. When companies are in low-growth industries, the relationship between regional investment preference and corporate cash holdings is not significant, while when enterprises are in high-growth industries, the relationship between regional investment preference and corporate cash holdings is significantly positive at the level of 1%, indicating that regional investment preferences have enhanced the company’s strategic cash holding motivation to hold more cash.

6. Further Research

6.1. Internal Control

Good internal control can not only restrain managers’ deviant operations and investment behaviors but also keep the company’s cash holdings within a reasonable range [29,30]. Effective internal supervision is particularly important in regions with investment preferences. Without effective internal supervision, the management is more likely to over-invest or under-invest, leaving the company with cash that is too high or too low. This paper uses the number of board members and the ratio of CEO salaries to measure the quality of internal control [31].

Board members have the right to vote. More board members will help to better curb the over-investment or under-investment caused by regional investment preference and make the company’s cash holdings within a reasonable range. In this paper, the sample is divided into a high board member group and a low board member number group according to the median number of board members of the company in each year. The regression results are shown in Panel A in

Table 10. The impact of internal control on regional investment preference and corporate cash holdings.

	Panel A: Number of Boards				Panel B: CEO Salary Ratio
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	Cash 1	Cash 1	Cash 2	Cash 2	Cash 1	Cash 1	Cash 2	Cash 2
	Low	High	Low	High	High	Low	High	Low
Reinvest 1	0.6675 ** (2.45)	0.7297 (1.37)	0.5143 *** (4.93)	0.7069 (1.04)	0.6043 *** (2.89)	0.5211 (1.29)	0.6317 *** (2.97)	0.5286 (1.33)
Size	−0.0469 *** (−12.24)	−0.0179 ** (−2.14)	−0.0470 *** (−12.27)	−0.0064 *** (−5.20)	−0.0239 *** (−8.53)	−0.0452 *** (−7.46)	−0.0260 *** (−9.38)	−0.0445 *** (−7.28)
Lev	−0.2761 *** (−11.49)	−0.5697 *** (−10.43)	−0.2746 *** (−11.42)	−0.1890 *** (−23.52)	−0.2201 *** (−12.11)	−0.8574 *** (−24.26)	−0.2223 *** (−12.31)	−0.8476 *** (−23.91)
Cf	1.1833 *** (28.94)	0.7578 *** (8.86)	1.1838 *** (28.95)	0.3402 *** (27.02)	1.6448 *** (61.80)	−1.3022 *** (−55.15)	1.6303 *** (61.95)	−1.3008 *** (−54.81)
Nwc	−0.0399 ** (−2.01)	−0.1038 ** (−2.40)	−0.0399 ** (−2.01)	0.0085 (1.33)	0.2186 *** (17.86)	−0.7262 *** (−33.29)	0.2151 *** (17.69)	−0.7200 *** (−32.92)
Capex	0.0007 (0.56)	−0.0031 (−1.23)	0.0007 (0.58)	0.0000 (0.12)	0.0004 (1.05)	−0.0005 ** (−2.05)	0.0004 (0.94)	−0.0005 ** (−1.99)
Growth	0.0419 *** (3.46)	−0.0144 (−0.60)	0.0417 *** (3.44)	0.0018 (0.51)	−0.0000 (−0.01)	0.0025 (0.56)	0.0012 (0.34)	0.0020 (0.47)
Div	0.0267 ** (2.39)	0.0510 ** (2.01)	0.0265 ** (2.37)	0.0254 *** (6.81)	0.0024 ** (2.02)	0.0008 (1.14)	0.0024 ** (2.02)	0.0008 (1.10)
Top1	0.0004 * (1.65)	0.0006 (1.10)	0.0004 (1.63)	0.0004 *** (4.44)	0.0006 *** (2.89)	0.0015 *** (3.76)	0.0008 *** (4.15)	0.0012 *** (3.01)
Board	−0.0170 (−0.52)	−0.1429 ** (−2.26)	−0.0159 (−0.49)	−0.0009 (−0.09)	−0.0337 ** (−1.99)	0.0319 (0.86)	−0.0181 (−1.08)	0.0126 (0.34)
Indratio	0.3290 *** (3.07)	−0.2103 (−1.16)	0.3327 *** (3.10)	0.0144 (0.54)	0.0366 (0.61)	−0.0774 (−0.60)	0.0383 (0.65)	−0.0893 (−0.69)
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
R2_adj	0.1118	0.0551	0.1119	0.2666	0.3076	0.2534	0.3098	0.2514
N	15,745	14,955	15,745	14,955	15,651	15,039	15,651	15,039

The numbers in parentheses are the t values of the variable; *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

. Columns (1) and (2) are the regression results when the dependent variable is Cash 1. Among them, column (1) is the regression result between regional investment preference and corporate cash holdings when the board members are lower than the median, and the regression coefficient is significantly positive at the level of 1%. Column (2) is the regression result between regional investment preference and corporate cash holdings when the board members are higher than the median, and the regression coefficient is not significant. It shows that good internal control weakens the correlation between regional investment preference and company cash. Columns (3) and (4) are the regression results with the dependent variable Cash 2, and the results are the same.

The greater the influence of the company CEO on the company, the worse the company’s internal control. Therefore, this paper uses the ratio of CEO compensation to management’s total compensation (Pay_slice) to measure the quality of internal control. The smaller the ratio, the better the internal control. This paper divides the sample into a high CEO pay group and a low CEO pay group according to the median of the company’s CEO pay ratio each year. The regression results are shown in Panel B in Table 10. Columns (1) and (2) are the regression results when the dependent variable is Cash 1. Among them, column (1) is the regression result of regional investment preference and corporate cash holdings when the CEO compensation ratio is higher than the median, and the regression coefficient is significantly positive at the 1% level. Column (2) is the regression result between regional investment preference and corporate cash holdings when the CEO salary ratio is lower than the median, and the regression coefficient is not significant. It shows that good internal control weakens the correlation between regional investment preference and corporate cash. Columns (3) and (4) are the regression results with the dependent variable Cash 2, and the results are the same.

6.2. External Supervision

External supervision can not only limit the abnormal behavior of the company’s management but also keep the company’s cash holdings within a reasonable range [32]. External supervision mechanisms are even more important for companies headquartered in regions with strong investment preferences. Without effective external supervision, managers may under-invest or over-invest, making the company’s cash holding level too high or too low. This paper measures the quality of external oversight using the volume and institutional ownership tracked by analysts.

Analysts play the role of information intermediaries between enterprises and investors, reducing the degree of information asymmetry between them. As a result, analysts can act as supervisors to curb misconduct by company management. This paper divides the sample into a high analyst number group and a low analyst number group according to the median number of analysts who track a company each year. The regression results are shown in Panel A in

Table 11. The effect of external control on regional investment preference and corporate cash holdings.

	Panel A: Number of Analysts				Panel B: Institutional Investor Shareholding Ratio
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	Cash 1	Cash 1	Cash 2	Cash 2	Cash 1	Cash 1	Cash 2	Cash 2
	Low	High	Low	High	Low	High	Low	High
Reinvest 1	0.4439 *** (5.54)	0.5928 (1.64)	0.5015 *** (4.66)	0.5563 (1.58)	0.6463 *** (5.42)	0.1876 (0.42)	0.2556 ***(3.36)	0.9346 (0.54)
Size	−0.0195 *** (−4.63)	−0.0299 *** (−6.30)	−0.0205 *** (−4.66)	−0.0289 *** (−6.15)	−0.0335 *** (−12.27)	−0.0328 *** (−4.81)	−0.0333 *** (−12.19)	−0.0329 *** (−4.82)
Lev	−0.3700 *** (−16.21)	−0.8548 *** (−27.94)	−0.3960 *** (−16.56)	−0.8157 *** (−26.90)	−0.3086 *** (−17.19)	−0.9618 *** (−26.32)	−0.3082 *** (−17.15)	−0.9613 *** (−26.28)
Cf	0.7910 *** (23.63)	−1.1006 *** (−49.57)	0.9953 *** (28.92)	−1.1475 *** (−52.49)	0.7484 *** (38.86)	−1.6056 *** (−60.74)	0.7483 *** (38.84)	−1.6057 *** (−60.74)
Nwc	0.0467 *** (2.64)	−0.5477 *** (−30.27)	0.0582 *** (3.18)	−0.5397 *** (−30.22)	−0.0182 (−1.53)	−0.5634 *** (−24.40)	−0.0180 (−1.50)	−0.5634 *** (−24.40)
Capex	0.0013 * (1.81)	−0.0004 (−1.51)	0.0009 (1.49)	−0.0004 (−1.50)	−0.0001 (−0.63)	−0.0005 (−1.57)	−0.0001 (−0.60)	−0.0005 (−1.57)
Growth	0.0031 (1.34)	0.0066 (1.22)	0.0035 (1.37)	0.0058 (1.07)	0.0014 (0.42)	0.0053 (1.20)	0.0012 (0.37)	0.0053 (1.20)
Div	0.0011 (1.30)	0.0006 (0.81)	0.0011 (1.18)	0.0005 (0.74)	0.0000 (0.06)	0.0030 ** (2.06)	0.0000 (0.06)	0.0030 ** (2.06)
Top1	0.0008 *** (3.42)	0.0016 *** (4.37)	0.0006 ** (2.48)	0.0017 *** (4.81)	0.0001 (0.67)	0.0027 *** (6.57)	0.0001 (0.66)	0.0027 *** (6.56)
Board	0.0115 (0.53)	−0.0073 (−0.24)	0.0296 (1.32)	−0.0211 (−0.70)	0.0292 * (1.70)	0.0134 (0.37)	0.0296 * (1.72)	0.0137 (0.37)
Indratio	0.0347 (0.45)	0.0283 (0.26)	0.1468 * (1.82)	−0.0526 (−0.50)	0.0717 (1.18)	−0.0598 (−0.46)	0.0769 (1.26)	−0.0592 (−0.46)
Industry	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
R2_adj	0.1688	0.1912	0.1810	0.2020	0.1837	0.2824	0.1832	0.2824
N	14,240	15,910	14,240	15,910	15,658	15,042	15,658	15,042

The numbers in parentheses are the t values of the variable; *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively.

. Columns (1) and (2) are the regression results when the dependent variable is Cash 1. Among them, column (1) is the regression result of regional investment preference and corporate cash holdings when the number of analysts tracking is lower than the median, and the regression coefficient is significantly positive at the 1% level. Column (2) is the regression result between regional investment preference and enterprise cash holdings when the number of analysts is higher than the median, and the regression coefficient is not significant. It shows that good external supervision weakens the correlation between regional investment preference and company cash. Columns (3) and (4) are the regression results with the dependent variable Cash 2, and the results are the same.

Some researchers have found that the institutional investors of the company will manage or supervise the company through the threat of exit. Therefore, institutional investors can act as a supervisor to restrain the company’s misconduct. This paper divides the sample into a high institutional investor holding group and a low institutional investor holding group according to the median of the company’s institutional investor holding ratio each year. The regression results are shown in Panel B in Table 11. Columns (1) and (2) are the regression results when the dependent variable is Cash 1. Among them, column (1) is the regression result of regional investment preference and corporate cash holdings when the shareholding ratio of institutional investors is lower than the median. The regression coefficient is significantly positive at the 1% level. Column (2) is the regression result of regional investment preference and corporate cash holdings when the shareholding ratio of institutional investors is higher than the median, and the regression coefficient is not significant. It shows that good external supervision weakens the correlation between regional investment preference and company cash. Columns (3) and (4) are the regression results with the dependent variable Cash 2, and the results are the same.

7. Conclusions

Based on a sample of Chinese A-share listed companies from 2006 to 2020, this paper explores the impact of regional investment preferences on corporate cash holdings. The study finds that a strong regional investment preference will improve the level of cash holdings of enterprises because the regional investment preference can increase the preventive cash holding and strategic cash holding motivation of enterprises. After a series of robustness tests, such as the instrumental variable method and propensity score matching method, the main regression results of this paper remain unchanged. Further research shows that strict internal control and external supervision can weaken the impact of regional investment preference on the company’s cash holdings. This research is of great significance to the cash holdings of enterprises. Most of the existing research is based on the company level and institutional perspectives to study the key factors affecting cash holdings. This paper breaks through the traditional theoretical framework of contracts and institutions and studies from the perspective of informal institutions. This paper analyzes the effect and mechanism of regional investment preference on corporate cash holdings and provides a new explanatory dimension for the level of corporate cash holdings.

This study confirms that the investment preference of the region where the company is located can affect the cash holding level of the company, but this study still has some limitations. First of all, the regional investment preference used in this paper is mainly the ratio of financial assets such as stock funds and per capita income in the region. Although the measurement method in this paper can better measure investment preference than the indicator of per capita lottery transaction volume used by previous scholars, in China, most residents’ income is mainly concentrated in real estate, and financial assets are inferior to real estate investment. Secondly, due to the large regional differences among provinces in China, companies in different provinces may be affected differently by the investment environment of the province, so the conclusions may be insufficient.