A New Perspective on Interpreting the Accounting Information of Listed Companies: Research on the Asset Structure Difference and Earnings Value Based on a Sustainable Development Strategic Perspective

Abstract

The quality of accounting information of listed companies has always been of great concern for investors, creditors, government departments, and other stakeholders. As the core component of accounting information, earnings information is essential, which is hailed as a wind vane by every stakeholder. The sustainable development strategy of companies requires companies to reasonably allocate company resources and promote the long-term and rapid development of the company. Based on the sustainable development strategic perspective, this paper takes Chinese A-share non-financial listed companies from 2007 to 2020 as a sample to study the correlation between asset structure differences and earnings value relevance. The research logic of this paper is: the corporate strategy determines the resource allocation scheme of the company, the resource allocation scheme determines the asset structure, and the asset structure determines the generation mode of profits. Therefore, there are different profit compositions, and the differences in profit compositions lead to differences in the corporate earnings value relevance. The research results show that, overall, compared to the companies with a large proportion of investment assets, the companies with a large proportion of operating assets have a stronger earnings value relevance to their net profit. From the perspective of the main composition of profits, companies with a large proportion of operating assets have a stronger value relevance to their core profits. Companies with a large proportion of investment assets have a stronger value relevance to their investment income. Listed companies’ different asset structures is an essential factor affecting the corporate earnings value relevance. Additionally, listed companies’ different asset structure results from the corporate strategic development choice, which shows that the corporate strategy affects the relevance of the corporate earnings value. This provides a new perspective for accounting information users to interpret the corporate earnings information, helps accounting information users to understand and analyze corporate financial statements more accurately, and has a particular reference significance for investment decision making.

1. Introduction

Earnings refer to the achievement of a company in a specific accounting period, including operating profit, investment income, etc. [1]. Earnings are a critical factor for a company’s survival and future development. As a piece of essential accounting information, investors, debtors, and other stakeholders are very concerned about it [2]. Since Ball and Brown [3] creatively verified that earnings have an investment reference value, the earnings value relevance has always been an important topic in academic research, which is hailed as a wind vane by investors, creditors, and other stakeholders. Earnings value relevance mainly reflects the relationship between stock price and accounting information. If the earnings value relevance is strong, it shows that accounting information plays a decisive role in stock investment. Earnings value relevance is the test of the “decision usefulness” of earnings information, which explicitly refers to the judgment of whether earnings information is helpful to information users.

In essence, accounting income generation is the process of corporate resource allocation and utilization. The choice of corporate resource allocation is reflected in the different asset structures of the company. The asset structure of the parent company no longer cares about liquidity and illiquidity. Still, assets’ strategic attributes are operating assets and investment assets [4]. Therefore, by analyzing the relevant accounting information of companies, we can infer the choice of corporate strategy and the effect of strategy implementation. To further illustrate, earnings value relevance is fundamentally determined by the corporate strategy. However, the existing studies either analyze the value relevance of each accounting subject one by one according to the top-down order of the income statement [5,6] or divide the earnings into two parts: persistence and non-persistence [7,8,9]. Most of these studies focus on discussing corporate earnings itself, which is the analysis from numbers to numbers. They do not analyze the root causes of the difference in the relevance of the earnings value through the essence of accounting information.

According to Hitt [10], strategy refers to the combination of agreements and actions that a company intends to develop for a long time to obtain a leading position in the market under market competition. Companies formulate it by adapting to the changing environment, seeking core competitiveness, and obtaining a competitive advantage. The existing literature does not fully realize that corporate strategy is the internal basis affecting the earnings quality. Research into corporate strategy is also mainly carried out using relevant financial indicators, such as current ratio, interest coverage ratio, return on net assets, debt asset ratio, and so on [11,12,13]. However, the relevant research ignores that the strategy determines the indicators themselves. Zhang [14] constructed the strategic division method of corporate resource allocation based on the internal relationship between the elements of the financial statements and the corporate operation and management activities. According to the asset structure of the parent company, the company can be operation-oriented or investment-oriented, or both, i.e., by combining operation and investment orientation. Further research shows that the characteristics of asset structure can clearly explain the features of the corporate process and reflect the requirements of the corporate business development strategy. In other words, different corporate strategies must have corresponding asset structures. Therefore, by comparing the corporate asset structure with the corporate strategy statement, we can find the adaptability between the profit model and the corporate strategic choice. It allows external stakeholders to understand the corporate-specific profit source model and resource allocation consequences to a great extent. Therefore, the research logic of this paper is: the corporate strategy determines the resource allocation scheme of the company, the resource allocation scheme determines the asset structure, and the asset structure determines the generation mode of profits. Therefore, there are different profit compositions, and the differences in profit compositions lead to differences in the corporate earnings value relevance. The specific logic diagram is shown in Figure 1.

Based on the current research status, this paper takes Chinese A-share non-financial listed companies from 2007 to 2020 as a sample. From the perspective of corporate strategy, this paper empirically explores the impact of listed company asset structure on the earnings value relevance. The financial statements of listed companies are divided into parent company financial statements and consolidated financial statements. Based on the research purpose of this paper, the data of this paper mainly come from the parent company’s financial statements of listed companies. By analyzing the investment asset accounts such as “long-term equity investment” in the parent company’s financial statements, we can know the parent company’s investment status in subsidiaries. Thus, we can know the strategic choice of the company, that is, the operation-oriented company or investment-oriented company mentioned in Figure 1 (the relationship between listed companies, parent companies, and their subsidiaries is shown in Figure 2). Firstly, this paper tests whether the value relevance of net profit is different under different asset structures. Then, the profit is divided into three parts—the core profit, investment income, and net non-operating income and expenditure—and it is tested whether the earning value relevance is different under different asset structures. The research results show that, when other conditions remain unchanged, from the perspective of net profit itself, compared to the companies with a large proportion of investment assets, the companies with a large proportion of operating assets have a stronger earnings value relevance. From the perspective of the main composition of profits, the companies with a large proportion of operating assets have stronger value relevance to their core profits. Companies with a large proportion of investment assets have stronger value relevance to their investment income. The listed company’s asset structure results from the corporate strategic development choice, which shows that the corporate strategy affects the earnings value relevance. The expected contributions of this paper are: First, the existing literature mainly discusses the earnings value relevance itself, which is from number to number. This paper explores the essence behind the number: a corporate strategy. From the perspective of strategy, this paper empirically studies the impact of listed companies’ asset structure on earnings value relevance, which enriches the research on the influencing factors of earnings value relevance. Second, it provides a new perspective for users of accounting information to interpret corporate earnings. It is helpful for accounting information users to understand and analyze corporate financial statements more accurately and has particular reference significance for investment decision making.

2. Theoretical Analysis and Research Hypothesis

2.1. Asset Structure Differences and Earnings Value Relevance

As non-accounting information, the research on the value relevance between corporate strategic information and accounting information has attracted more and more attention [15]. There are many perspectives for classifying corporate strategies, such as classification by type, management level, or region, etc. The main emphasis of this paper is how to allocate companies’ existing resources. Corporate strategy is a decision-making process of the optimal allocation of resources. Different strategic positioning will produce different resource allocation structures. Therefore, according to the corporate resource allocation structure, we can judge the difference between its strategic positioning and strategic choice. It means that the corporate strategic difference depends on its resource allocation structure. Different resource allocation schemes will constitute different asset structures of companies [16,17]. Dichev [18] studied the relationship between corporate strategy and corporate accounting earnings management. They found that the corporate strategy has an important impact on the accounting earnings management. At the same time, there is a significant negative correlation between the degree of earnings management and the quality of accounting information; that is, the higher the degree of earnings management, the lower the quality of accounting information. Ye [19] concluded that investors pay attention to strategic information and accounting information, and the value relevance of the owner’s equity is high. Zhu [20] believed that the corporate business model is the critical factor affecting the usefulness of earnings information. After interviewing the CFOs of several companies, Dichev [18] believed that corporate strategic decision making primarily influences earnings quality indicators. Scholars have also found that strategic differences will impact earnings quality from earnings management behavior, earnings management level, earnings timeliness, and earnings sustainability [21,22]. Zhou [23] found that the earnings value relevance of differentiated strategic companies is lower than that of cost-leading strategic companies.

As a mechanism of transmitting information, accounting helps to reduce information obstacles in the market. Earnings value relevance is the prediction and interpretation of corporate earnings information on the securities market value, reflecting the decision-making usefulness of corporate earnings information. Ball and Brown [3] empirically studied the information content of corporate earnings for the first time, which opened a new situation in the study of the relevance of the earnings value. Subsequently, many works of literature use empirical methods to supplement and expand it. The research conclusion is consistent: corporate earnings have value relevance. Some studies have pointed out that the explanatory power of earnings gradually decreases. However, as the concentrated embodiment of operating results, earnings are still the primary basis for investors to understand the actual operation of companies and make decisions [24]. The research on the influencing factors of earnings value relevance involves many aspects, such as the characteristics of accounting information itself, business fundamentals, market information environment, etc. The research based on the factors of accounting information itself mainly examines the incremental impact of accounting standards on the usefulness of earnings decision making. Companies pay attention to the effect of the accounting reporting system on the resource allocation function of the capital market. Business fundamentals such as business conditions, related party transactions, corporate strategic choices, and technological innovation can affect earnings value relevance [25]. They help to deeply explore the key factors influencing the efficiency of the capital market. The existing literature mainly discusses the usefulness of earnings decision making from an earnings information generation and dissemination perspective. The accounting reporting systems and business fundamentals work together to report the information content of earnings. Rumors and other market information will affect the efficiency of earnings information dissemination. Finally, the market information leads to a change in earnings value relevance. As the primary basis for investors’ decision making [24], whether earnings are sustainable and predictable is related to investors’ expectations of the future investment value of companies. Then, it shows the change in earnings value relevance. It is necessary to further study how the difference in corporate asset structure affects the earnings value relevance from a strategy perspective.

Investors or creditors, usually judging the operation of a company, will pay more attention to its earnings level. However, the information view of decision-making usefulness holds that the information content of accounting information should not be limited to the attention to corporate earnings. Through the available financial statement information, we should also be able to obtain more information to judge the future development prospect or profitability of the company. Such forecasting ability is obtained through the reaction of the stock market. As long as it is a practical or weakly practical capital market, it will respond to the profit information, mainly reflected in the stock price. Before the announcement of earnings information, the market will first have a prediction of the corporate stock return. If the announced earnings are positive, it is good news for the company, and the stock price may produce a positive abnormal return. If the stock market’s reaction is closer to the change in corporate earnings, it shows that the prediction ability of financial information is stronger. Therefore, the earnings response coefficient is introduced to measure this prediction ability. The stronger the prediction ability, the more information the earnings information published in the report has, and the more valuable it is for investors to make decisions. In other words, in the market, the more valuable the earnings information is, the stronger the earnings value relevance and the greater the earnings response coefficient. The earnings response coefficient refers to the reaction degree of the securities market to earnings information or the ability of corporate earnings to predict the future operation of the company. To link with the future development prospect and the investment value of the company, it needs to be further combined with the development strategy that can represent the future of the company. There are significant differences in the asset structure of different companies, and the essence behind it is the choice of corporate strategy.

If the company’s operating assets account for a large proportion, its strategic choice is mainly to develop the operation activities of the company. Such a company takes production and sales as its primary business and has more current accounts and fixed assets. It can be concluded that, for a company with a large proportion of operating assets, whose business is concentrated in its central business field, its profit and future development potential depend on the profitability of its major business, and its core competitiveness is formed through the development of its primary industry [14]. The main business is more prominent, which shows that the current business mode is successful. The company will generally continue to choose the current mode. In this way, the corporate profit generation has stability and sustainability. Therefore, the current earnings can predict future profits and cash flow and have higher value relevance. If the parent company’s investment assets account for a large proportion, its development center is not to generate and operate but to carry out investment business. For example, a parent company invests in other companies to form long-term equity investments. The income of the company comes from its investment income. Investment assets mainly include control assets and non-control assets. For controlled investments, the companies dominated by assets are often large-scale group companies. Such company is guided by the overall strategy of diversification or integration. It mainly focuses on subsidiaries’ operation and asset management and realizes expansion and strengthening quickly through external development. Such company essentially controls the subsidiaries through mergers, reorganization, and asset injection (a large number of long-term equity investments and other receivables are reflected in the statements of the parent company). The company itself has little or no participation in business activities. For non-controlling investment assets, profits are mainly obtained through short-term investments or gains and losses from changes in fair value (there will be many assets available for sale in the financial statements of the parent company). Therefore, it is the “deviation” or “transfer” that both controlled and non-controlled investment assets show that the company’s own main business is not prominent. To sum up, the source of profits of companies does not depend on their primary business but on finance and investment. Therefore, the variability and volatility of profits are significant, the unpredictability is vital, and the persistence of earnings is weak.

Based on the above inference, this paper puts forward Hypothesis 1:

H1: 

Companies with a large proportion of operating assets of the parent company have more relevance to the earnings value than companies with a large proportion of investment assets.

2.2. Asset Structure Differences and Value Relevance of Each Earnings Component

Through empirical research, Zhao [7] concluded that earnings components could better explain corporate stock returns than aggregate earnings. There are significant differences in the coefficients between the expected number of earnings items and stock returns. Although decomposing earnings provides less information, the relationship is substantial. There are many ways to decompose earnings. According to the persistence of earnings, Ramakrishnan and Thomas [26] decomposed earnings into permanent earnings, short-term earnings, and the earnings that are not related to price. They also constructed a research model on the relationship between the unexpected part of earnings items and price changes. Some scholars also studied the value correlation of decomposition earnings from operating profit, net profit, and comprehensive income. Agnes [27] chose the relative value correlation and the value-added relevance. A comparative study on the correlation between earnings and returns under each earnings indicator is carried out by year and industry goodness of fit. As earnings components, the value relevance of revenue and expenses is different. Swaminathan [28] and Jiang [29] split corporate earnings into income and expenses and then used the event study method to study value relevance. Many studies break down the earnings in more detail. Cheng and Gong [5] used the data of A-share listed companies in the Shanghai Stock Market in 2007 as the research sample. They regarded asset impairment, investment income, and other items as non-basic earnings items, and the net profit deducted from these items as essential earnings items. Other comprehensive income items are also considered to examine the relationship between earnings and specific objects, stock prices, and stock returns. Similar to Cheng and Gong [5], Chen [30] decomposed earnings into core and non-core earnings, studying the persistence, value correlation, and market pricing of earnings components of two parts of earnings. Based on a similar perspective, Zhang [31] decomposed net profit into cash flow from operating activities, controlled accrual profits, and non-controlled accrued profits, studying its effect on cumulative abnormal returns of stocks.

At present, there are two main methods of earnings decomposition in the research on value correlation of earnings structure: One is according to the income statement. Scholars divided earnings into revenue and expenses, operating and net profit, core, and non-core earnings, etc. Another is to divide earnings into accruals, cash flows, etc. Most of the literature breaks down profits directly by accounting items in the income statement and the persistence of earnings composition. These studies are very important for a comprehensive understanding of earnings value relevance. However, there is a lack of necessary refinement and exploration of the business essence, business characteristics, development model, and the strategy behind different profit compositions.

According to the Chinese “The Corporate Accounting Standard”, earnings are the operating results of a particular period. Earnings include operating profit, investment income, and net non-operating income and expenditure. As the core profit comes from regular corporate activities, if the company’s core profit is more significant, it means that the company has a more robust ability to obtain earnings from its main business and more substantial core competitiveness. Investment income (the sum of profit and loss from changes in fair value and investment income) mainly emphasizes the appreciation of investment assets. On the one hand, it reflects the income from changes in the fair value of investment assets (especially trading financial assets, available for sale and held to maturity investments) in the holding process. On the other hand, it reflects the dividend income of the subsidiary company of external control investment (mainly refers to long-term equity investment and other receivables) to the parent company. Most of the net non-operating income and expenditure consists of some occasional non-recurring net profit and loss items, such as gains and losses on the disposal of non-current assets and subsidy income. Although, the net non-operating income and expenditure are also part of the profit and loss of the company. However, non-operating income and non-operating expenditure have nothing to do with business activities. Usually, they have the characteristics of “contingency” or “one-time”. Due to the great randomness, it is difficult to treat the value relevance of non-operating income and expenditure differently in different asset structures.

Therefore, this paper only considers whether the value correlation of core profit and investment income is significantly different in different asset structures. For companies with a large proportion of operating assets, the company’s operating activities dominate. Its focus is on the main business, so the core profit reflecting the profitability of its main business has a more critical value meaning. For companies with a large proportion of investment assets, their focus is on external investment. The investigation of its value should pay more attention to the profit and loss of changes in fair value and the dividends generated by external control investment. Therefore, investment income has a more important value meaning.

Although, from the perspective of statements, no matter the company’s asset structure, the investment income accounts for a small proportion of the total profit in the consolidated statements. However, compared with the companies with a large proportion of operating assets, the companies with a large proportion of investment assets have a larger investment ratio, and their marginal contribution of investment income to earnings is much larger. Therefore, investment income has stronger value relevance.

No matter the kind of asset structure of the company, the value relevance of core profits is strong. It shows that operating assets are the assets that bring profits to the company. However, investment income mainly exists in companies with a large proportion of investment assets. Therefore, compared with the companies with a large proportion of operating assets, the companies with a large proportion of investment assets have a stronger value relevance of investment income.

Based on the above inference, this paper puts forward Hypothesis 2:

H2a: 

Compared with the company with a large proportion of investment assets of the parent company, the company with a large proportion of operating assets of the parent company has a stronger value relevance to its core profits.

H2b: 

Compared with the company with a large proportion of operating assets of the parent company, the company with a large proportion of investment assets of the parent company has a stronger correlation with the value of its investment income.

3. Research Design

3.1. Sample Selection and Data Source

Since the new accounting standards came out in 2007, there have been significant changes in the income view of the income statement. To maintain the consistency of relevant profit calculation, this paper selected A-share listed companies from 2007 to 2020 as the initial sample. Additionally, the following treatment was carried out: (1) excluding financial and insurance listed companies, (2) eliminate delisting, St, *ST, and other companies with special financial conditions, (3) eliminate companies with incomplete data, and (4) eliminate extreme values (the continuous variables are treated with ±1% winsorize). Finally, 20,234 observations were obtained. The sample data come from the CSMAR database, and the data processing and analysis were conducted using the STATA software.

3.2. Models and Variables

The models used in the existing research literature on corporate earnings value relevance mainly include the price model and the income model. The price model takes the stock price as the dependent variable to study the correlation between the stock price and the book value of net assets and accounting income. The income model takes stock return as the dependent variable to explore the relationship between stock return and accounting return and its changes. It can be seen from the derivation of the price model and the income model that both models are derived from the standard price model of discounted future expected cash flow. Additionally, they are based on the assumption that the current accounting surplus contains the expected future cash flow [32,33]. From the market perspective, the current corporate earnings can be divided into two parts. One part is the unexpected part of the market: the unexpected surplus. The other part is the part that the market has expected in the early stage (such as the previous year): the expected surplus. The expected earnings are not related to stock returns in the income model. The selection error of the earnings model will lead to the deviation of the response coefficient. Under the price model, the current stock price reflects the cumulative income information, and there is no problem with the dependent variable error. Therefore, the surplus response coefficient obtained by the price model will be an unbiased estimate of the actual response coefficient. Therefore, this paper intends to select the price model to study the value relevance of corporate earnings. The specific models to be tested are as follows:

$$P_{i,t}=\alpha +{\beta }_{1}AS+{\beta }_{2}Ep{s}_{i,t}+{\beta }_{3}A{S}^{*}Ep{s}_{i,t}+{\displaystyle \sum I}d+{\displaystyle \sum Y}ear+{\mu }_{i,t}$$

(1)

In model (1), P_i,t represents the stock price of the company i at the end of year t. Eps_i,t is the earnings per share of the company i in year t. AS represents the asset structure of the parent company. The specific measurement method is shown in

Table 1. Variable definitions.

Type	Symbol	Name	Definition
Dependent variable	P	Stock price	Closing price of the company’s shares at the end of the period
Independent variable	AS	Asset structure of parent company	The sample is grouped by the size of the parent company’s operating assets/(parent company’s operating assets + parent company’s investment assets) by year. Companies below the minimum 1/3 quantile and above the maximum 1/3 quantile are, respectively, defined as companies with a large proportion of investment assets of the parent company and companies with a large proportion of operating assets of the parent company. Companies with equal emphasis on both assets are in the middle 1/3 quantile (for the convenience of research, this sample was deleted). For companies with a large proportion of operating assets of the parent company, the value is 1. For companies with a large proportion of investment assets of the parent company, the value is 0.
	Eps	Earnings per share	Net profit attributable to owners of the parent company/total equity
	Core	Core profit per share	Core profit/total equity. Core profit = operating revenue - operating costs - business taxes and surcharges - selling expenses - administrative expenses - financial expenses (applicable before 2018). Core profit = operating revenue - operating costs - business taxes and surcharges - sales expenses - management expenses - R&D expenses - interest expenses (applicable after 2018).
	Inv	Investment income per share	(Fair value gains and losses + investment income)/total equity
	Oth	Net non-operating income and expenditure per share	(Non-operating income–non-operating expenses)/total equity
Control variable	Naps	Net assets per share	Net assets at the end of the period/total equity
	Size	Company size	Natural logarithm of net assets at the end of the period

. AS* Eps_i,t is the interactive term between the net profit and the asset structure of the parent company. In addition, it also controls the year and individual fixed effect (Id). The individual fixed effect is an influential factor that does not change with time at the individual level. Each individual has unique characteristics. Its purpose is to control the unique characteristics that do not change with time at the individual level, which can be used to overcome the problem of missing variables. The firm fixed effect in this study is to compare each company with itself when the explanatory variables change to eliminate the heterogeneity between companies. From the hypothesis of this paper, we need to focus on the test of the interaction term coefficient between AS and Eps_i,t. This paper predicts that β₃ is positive when operating assets dominate the parent company’s asset structure.

To test Hypothesis 2, we investigated the significant differences between core profit and investment income under different asset structures. Combined with the existing research [34,35], the models to be tested are as follows:

$$\begin{array}{l}{P}_{i,t}=\alpha +{\beta }_{1}AS+{\beta }_{2}Cor{e}_{i,t}+{\beta }_{3}A{S}^{*}Cor{e}_{i,t}+{\beta }_{4}In{v}_{i,t}+{\beta }_{5}A{S}^{*}In{v}_{i,t}+{\beta }_{6}Ot{h}_{i,t}\\ +{\displaystyle \sum I}d+{\displaystyle \sum Y}ear+{\mu }_{i,t}\end{array}$$

(2)

In model (2), Core_i,t is the company’s core profit per share, Inv_i,t is the investment income per share, and Oth_i,t is the net non-operating income and expenditure per share. AS represents the asset structure of the parent company. AS * Core_i,t is the interaction item between the core profit and the parent company’s asset structure, and AS * Inv_i,t is the interaction term between investment income and the asset structure of the parent company. The coefficients of these two interaction terms are the key to this paper. When operating assets dominate the parent company’s asset structure, the correlation between its core profits is stronger, while the correlation between investment income is weaker. β₃ is expected to be positive, and β₅ is negative. See Table 1 for specific variable definitions and calculations.

4. Empirical Analysis

4.1. Descriptive Statistics

The descriptive statistics in

Table 2. Descriptive statistics.

Variable	Unit	Observations	Average	Standard Deviation	1/4 Quantile	Median	3/4 Quantile
P	CNY	20,234	17.81636	0.1205437	7.31	12.29	21.41
AS	/	20,234	0.499654	0.0035151	0	0	1
Eps	CNY	20,234	0.4756073	0.0034797	0.14	0.33	0.6329
Core	CNY	20,234	0.5080793	0.0041346	0.1321002	0.3537272	0.6979188
Inv	CNY	20,234	0.0681851	0.00109	0	0.0123846	0.0624958
Oth	CNY	20,234	0.036961	0.0005488	−0.0003689	0.011074	0.0449691
Naps	CNY	20,234	4.866121	0.021508	2.770302	4.135953	6.111473
Size	CNY	20,234	21.47025	0.0082638	20.63665	21.34687	22.13074

Note: “Observations” is the number of samples for each variable. Except “AS”, the value unit of each variable is Chinese Yuan. The value of “AS” is 0 or 1, with no unit.

report the mean, standard deviation, 1/4 quantile, median, and 3/4 quantile of each variable. The median value of the parent company’s asset structure (AS) is 0, indicating that the sample consists of more companies with a large proportion of the parent company’s investment assets. From the average of core profit (Core), investment income (INV), and net non-operating income and expenditure (Oth), the value of core profit is the largest. It shows that, whatever the company’s asset structure and strategy, the company depends on its primary business for survival. The standard deviation between the stock price (P) and the net assets per share (Naps) is significant at the end of the period. It shows that their fluctuation range is extensive, which is in line with the characteristics of the A-share market.

4.2. Correlation Analysis

The Pearson correlation coefficients are listed in

Table 3. Pearson correlation analysis.

	P	AS	Eps	Core	Inv	Oth	Naps	Size
P	1
AS	0.195 ***	1
Eps	0.632 ***	0.132 ***	1
Core	0.558 ***	0.129 ***	0.882 ***	1
Inv	0.012 *	−0.208 ***	0.230 ***	−0.004	1
Oth	0.097 ***	−0.001	0.088 ***	−0.037 ***	−0.012	1
Naps	0.598 ***	0.149 ***	0.715 ***	0.643 ***	0.176 ***	0.068 ***	1
Size	−0.097 ***	−0.234 ***	0.226 ***	0.258 ***	0.240 ***	0.066 ***	0.248 ***	1

Note: *, ***, respectively, significance levels of 5%, and 0.1%.

. The results show that: net profit (EPS), core profit (Core), investment income (INV), and net non-operating income and expenditure (Oth) are significantly positively correlated with stock price (P). Companies with a large proportion of operating assets of the parent company usually have a larger net profit, larger core profit, higher equity face value, lower investment income, and higher stock price. There is a high correlation coefficient between individual variables, but the overall VIF value of each variable and the model is less than 10, so there is no multicollinearity problem.

4.3. Regression Analysis

Table 4. Regression results.

Variable	Model (1)	Model (2)
AS	−1.400 *** (−4.40)	−0.441 (−1.35)
Eps	13.64 *** (46.77)
AS* Eps	8.249 *** (21.76)
Core		9.860 *** (40.11)
AS* Core		8.166 *** (24.76)
Inv		7.418 *** (10.70)
AS*Inv		−4.124 *** (−3.14)
Oth		16.91 *** (16.02)
_cons	18.30 *** (48.24)	17.89 *** (45.41)
Id	Control	Control
Year	Control	Control
Obs	20,234	20,234
R−sq	0.483	0.453

Note: ***, significance level of 0.1%, and t statistics in parentheses.

reports the regression results of the two models. In model (1), there is a significant positive correlation between net profit and equity value, which is consistent with previous studies. The coefficient of AS * Eps_i,t is significantly positive (β₃ = 8.249, t = 21.76). It shows that compared with the company with a large proportion of investment assets of the parent company, the company with a large proportion of operating assets of the parent company has a stronger value correlation with its net profit, which is consistent with Hypothesis 1. In model (2), the value correlation of core profit, investment income, and non-operating income is significantly positive, which is consistent with previous studies. Further analysis found that, compared with the company with a large proportion of investment assets of the parent company, the value relevance of the core profits of the company with a large proportion of operating assets of the parent company is stronger (β₃ = 8.166, t = 24.76), and the value relevance of investment income is weaker (β₅ = −4.124, t = −3.14). The relevance of core profit is stronger in companies with a large proportion of operating assets of the parent company, and the relevance of investment income is stronger in companies with a large proportion of investment assets of the parent company, which is consistent with Hypothesis 2.

4.4. Robustness Check

4.4.1. The Alternative Variable of the Dependent Variable

According to the regulations of the CSRC (China Securities Regulatory Commission), the annual report disclosure of the Chinese securities market is completed from January to April of each fiscal year. Therefore, this paper selected the closing price of the last trading day of four months after each fiscal year as the alternative variable of the stock price in the model. The test results of model (1) are shown in column (1) of

Table 5. Robustness test results of model (1).

	(1)	(2)	(3)	(4)	(5)
AS	−1.012 ** (−3.29)	−1.426 *** (−3.34)	−1.991 *** (−4.71)	−1.549 *** (−5.12)	−1.853 *** (−5.71)
Eps	12.35 *** (43.83)	13.23 *** (37.52)	8.875 *** (21.83)	8.731 *** (29.07)	14.03 *** (47.33)
AS*Eps	6.526 *** (17.81)	10.06 *** (21.37)	12.98 *** (23.76)	6.994 *** (19.34)	8.126 *** (21.44)
Naps				1.764 *** (42.29)
Size					−1.128 *** (−7.03)
_cons	13.03 *** (35.53)	18.37 *** (38.76)	18.62 *** (46.14)	13.85 *** (36.86)	41.49 *** (12.49)
Id	Control	Control	Control	Control	Control
Year	Control	Control	Control	Control	Control
Obs	20,234	15,176	30,361	20,234	20,234
R−sq	0.411	0.484	0.478	0.533	0.485

Note: **, ***, respectively, significance levels of 1%, and 0.1%, and t statistics in parentheses.

, in which the coefficient of AS* Eps_i,t is significantly positive (β₃ = 6.526, t = 17.81). The test results of model (2) are shown in column (1) of

Table 6. Robustness test results of model (2).

	(1)	(2)	(3)	(4)	(5)
AS	−0.344 (−1.09)	0.130 (0.30)	−0.788 (−1.81)	−0.835 ** (−2.72)	−0.950 ** (−2.85)
Core	9.014 *** (38.18)	9.654 *** (31.58)	5.030 *** (14.65)	5.800 *** (23.61)	10.24 *** (40.84)
AS * Core	6.703 *** (21.16)	9.331 *** (22.35)	12.95 *** (27.46)	6.559 *** (21.08)	8.099 *** (24.59)
Inv	6.052 *** (9.09)	7.556 *** (9.43)	10.82 *** (9.52)	3.028 *** (4.61)	7.664 *** (11.06)
AS * Inv	−2.068 (−1.64)	−6.346 *** (−3.62)	−5.816 ** (−3.20)	−3.153 * (−2.56)	−4.065 ** (−3.10)
Oth	11.97 *** (11.81)	15.81 *** (12.63)	15.79 *** (19.42)	10.30 *** (10.31)	16.63 *** (15.78)
Naps				1.982 *** (47.79)
Size					−1.233 *** (−7.40)
_cons	12.86 *** (33.96)	17.78 *** (36.00)	17.85 *** (43.07)	13.09 *** (34.16)	43.23 *** (12.54)
Id	Control	Control	Control	Control	Control
Year	Control	Control	Control	Control	Control
Obs	20,234	15,176	30,361	20,234	20,234
R−sq	0.384	0.449	0.449	0.518	0.454

Note: *, **, ***, respectively, significance levels of 5%, 1%, and 0.1%, and t statistics in parentheses.

, in which the coefficient of AS* Core_i,t is significantly positive (β₃ = 6.703, t = 21.16). The coefficient of AS* Inv_i,t is negative and negatively correlated. It shows that the substitution of explained variables does not affect the results of this paper.

4.4.2. Regroup Samples

The sample was divided into four equal parts according to the differences in the asset structure of the parent company. Companies below the minimum quartile and above the maximum quartile are defined as companies with a large proportion of operating assets of the parent company and companies with a large proportion of investment assets of the parent company, respectively. The middle half of the samples are companies with equal emphasis on two assets (for the convenience of research, this paper also deleted this sample). For companies with a large proportion of operating assets of the parent company, the value is 1. For companies with a large proportion of investment assets of the parent company, the value is 0. For thr retest, the test results of model (1) are shown in column (2) of Table 5, in which the coefficients of AS* Eps_i,t are significantly positive (β₃ = 10.06, t = 21.37). The test results of model (2) are shown in column (2) of Table 6, in which the coefficient of AS* Core_i,t is significantly positive (β₃ = 9.331, t = 22.35). The coefficient of AS * Inv_i,t is significantly negative (β₅ = −6.346, t = −3.62). It shows that the reclassification of the parent company’s asset structure did not affect the results of this paper.

4.4.3. Change the Variable Type of Asset Structure Difference

The parent company’s asset structure difference variable transforms from a virtual variable to a continuous variable. We can further find out whether the profit, core profit, and investment income are directly affected by the ratio of operating assets to investment assets. The test results of model (1) are shown in column (3) of Table 5, in which the coefficient of AS* Eps_i,t is significantly positive (β₃ = 12.98, t = 23.76). The test results of model (2) are shown in column (3) of Table 6, in which the coefficient of AS* Core_i,t is significantly positive (β₃ = 12.95, t = 27.46). The coefficient of AS* Inv_i,t is significantly negative (β₅ = −5.816, t = −3.20). These results further show that, the higher the proportion of operating assets in the total assets, the more significant the positive correlation between the net profit and core profit, and the more significant the negative correlation with investment income. It further shows that the difference in asset structure can affect the value relevance of net profit, core profit, and investment income.

4.4.4. Control Net Assets Per Share

Net assets per share (Naps) is one of the main factors affecting the relevance of the earnings value. Hayn [36] found that because there are liquidation options in loss-making companies, the earnings information hardly works after putting the net asset variable into the regression model. Black [37] found that the relative and incremental effects of earnings and net assets per share are determined by the current earnings level when explaining stock value. To explain that the omission of this control variable does not cause the regression result and to test the stability of the above results, this paper added the net assets per share to regression above the earnings capitalization model. The test results of model (1) are shown in column (4) of Table 5, in which the coefficient of AS* Eps_i,t is significantly positive (β₃ = 6.994, t = 19.34). The test results of model (2) are shown in column (4) of Table 6, in which the coefficient of AS* Core_i,t is significantly positive (β₃ = 6.559, t = 21.08). The coefficient of AS* Inv_i,t is significantly negative (β₅ = −3.153, t = −2.56). The results are consistent with the main regression.

4.4.5. Control Company Size

The company size effect also affects the relevance of earnings information. Cai [38], Li [39], and Wei [40] all believed that the amount of unexpected information transmitted to the market in the actual earnings announcement should be inversely proportional to the size of the company. Large-sized companies will transmit more non-accounting information to the market than small-sized companies. Therefore, the earnings relevance of large-sized companies is weak. To explain that the omission of this control variable does not cause the regression result and to test the stability of the above results, after adding the company size into the above earnings capitalization model, this paper performed a regression again. The test results of model (1) are shown in column (5) of Table 5, in which the coefficient of AS* Eps_i,t is significantly positive (β₃ = 8.126, t = 21.44). The test results of model (2) are shown in column (5) of Table 6, in which the coefficient of AS * Core_i,t is significantly positive (β₃ = 8.099, t = 24.59). The coefficient of AS * Inv_i,t is significantly negative (β₅ = −4.065, t = −3.10). The results are consistent with the main regression.

5. Conclusions and Contribution

5.1. Conclusions

From the corporate strategy perspective, this paper took Chinese Shanghai and Shenzhen A-share non-financial listed companies from 2007 to 2020 as a sample, empirically testing the correlation difference in earnings value under different asset structures. The research conclusions are as follows:

(1)

Listed companies’ different asset structure affects the correlation of the earnings value. Overall, compared with the companies with a large proportion of investment assets, the companies with a large proportion of operating assets have a stronger earnings value relevance.

(2)

From the perspective of the main composition of profits, the value relevance of core profits is the strongest among the companies with a large proportion of operating assets, and the value relevance of investment income is the strongest among the companies with a large proportion of investment assets.

(3)

The listed company’s asset structure results from the corporate strategic development choice, which further shows that the corporate strategy affects the earnings value relevance.

5.2. Contribution

The main contributions of this paper are as follows:

(1)

The research based on the perspective of corporate strategy has found new evidence for the influencing factors of earnings quality and further expands the literature on earnings quality. The previous research is basically “talking about numbers based on numbers”, which does not consider that the essence of corporate profits is the result of corporate strategic choice.

(2)

It was found that the connotation behind earnings information is corporate strategy, and corporate strategy is the root of the differences affecting earnings quality. The research logic of this paper is: the corporate strategy determines the resource allocation scheme of the company, the resource allocation scheme determines the asset structure, and the asset structure determines the generation mode of profits. Therefore, there are different profit compositions, and the differences in profit compositions lead to differences in the corporate earnings value relevance.

(3)

It provides a new perspective for investors to more accurately interpret the earnings information of companies. We should understand the significance of current earnings and their components to the future development prospect and value of companies in combination with corporate strategy. It has a specific reference value for external investors and financial analysts to optimize investment choices because it can better explain the logic behind the corporate operation based on the strategic perspective, which will help the users of accounting information to understand and analyze the corporate financial statements more accurately and make reasonable investment decisions to promote the healthy development of the capital market. Not only do the general earnings have value relevance, but each specific item of earnings may also have value relevance. Investors pay attention to the surplus and spend more and more on the components. Investors also pay extra attention to each item of earnings. That is, the value relevance of each part of earnings is different.

(4)

It also has some enlightenment for the standard-setting department. Since the specific strategy of the company is an important factor affecting the characteristics of accounting information, the standard-setting department can encourage the company to disclose more and more comprehensive strategic information to help investors understand and analyze the corporate financial statements more accurately, conduct valuation analysis of the company more effectively, and make the correct decisions.

(5)

The research of this paper reflects the importance and necessity of the Treasury to vigorously promote the construction of a management accounting system [41]. Promoting the structure of a management accounting system and scientifically planning the development strategy of management accounting have become the new requirements of Chinese social and economic development for the guidance of accounting practice in the new era. To promote the construction of a management accounting system, the Treasury issued the “basic guidelines for management accounting” (CK (2016) No. 10) [42] in June 2016. Take strategic orientation as the first application principle of management accounting and emphasize that the application of management accounting is guided by strategic planning. To further promote the practice of management accounting, in September 2017, the Treasury issued 22 guidelines for the application of management accounting (CK (2017) No. 24) [43]. Among them, “management accounting application guide No. 100: strategic management” defines the concept, principles, application environment, application procedures, and strategic management tools to improve the Chinese management accounting system. Strategic management will play a more critical role in corporate strategic decision making, evaluation, and control as an essential management accounting function. The research in this paper is helpful to encourage the company to strengthen strategic management and the application of management accounting.