Abstract
CHINA JOURNAL OF ACCOUNTING STUDIES 2020, VOL. 8, NO. 3, 410–434 https://doi.org/10.1080/21697213.2021.1881276 ARTICLE Reclassification of income statement items and weight adjustment of compensation performance indicators Jing Chen and Junxiong Fang School of Management, Fudan University, Shanghai, China ABSTRACT KEYWORDS Executive compensation The selection and weighting of performance indicators are of vital contract; new accounting importance for an effective compensation contract. We examine standards; investment the effect of the reclassification of income statement items, caused income by China’s new Accounting Standards for Business Enterprises (ASBE) in 2007 on the weight adjustment of compensation perfor- mance indicators. The results show that the sensitivity of executive pay and investment income increases significantly after ASBE moves investment income in the income statement from below- the-line of operating income to above-the-line, which indicates that the disclosure position of income statement items is directly related to the weight of compensation performance indicators. We also find that the earnings persistence of investment income increases significantly after ASBE, which implies that the reclassification of investment income conforms to business practice and also per- forms well. However, the increased sensitivity of executive pay and investment income may induce management’s opportunistic investment in financial assets. 1. Introduction Executive compensation contracts play a crucial role in modern firms to align the interests of shareholders to those of management and thus to solve the agency problem (Jensen & Meckling, 1976; Jensen & Murphy, 1990). However, information asymmetry makes it either impossible or excessively costly for shareholders to obtain the complete information of management’s efforts. Therefore, performance-based compensation contracts which link executive compensation to firm performance become a suboptimal solution (Holmstrom, 1979). In executive compensation contracts, performance indicators not only convey corporate business objectives (Angelis & Grinstein, 2015; Balsam et al., 2011), but also affect management’s behaviour (Huang et al., 2014; Marquardt & Wiedman, 2005; Young & Yang, 2011). Hence, the key to an ideal compensation contract lies in proper selection and weighting of performance indicators. Specifically, performance indicators with high information content should be included, and a corresponding weight should be given according to their sensitivity to management’s efforts (Bushman et al., 1996; Holmstrom & Milgrom, 1991). For example, as an aggregate item in the income statement, operating CONTACT Jing Chen 18110690016@fudan.edu.cn School of Management, Fudan University, Yangpu, Shanghai, 200433, China Paper accepted by Kangtao Ye. © 2021 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/ licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. CHINA JOURNAL OF ACCOUNTING STUDIES 411 income reflects corporate recurring production and operation outcome directly related to management’s efforts (Lu & Jiang, 2012). And that is why operating income is often included and given a high weight in executive compensation contracts (Holmstrom & Milgrom, 1991). However, with the continuous change of economic environment, the economic con- notation of performance indicators is constantly evolving. At the early stage of China’s capital market, corporate production and operation activities were relatively simple. Internal physical investment was the main operation model of firms. Only a small amount of external equity investment was made for short-term financing purposes. In this case, the resulting investment income could not reflect corporate recurring operation results and thus was excluded from operating income. However, with the development of a market economy, listed firms are becoming increasingly diversified. More and more listed firms make long-term equity investment to expand their business and obtain capital gain, therefore investment income is increasingly not distinct from core operating income in nature. Taking Huaqiaocheng-A (000069), a cultural tourism and real estate firm, as an example, it made a net profit of 580.88 million RMB in 2006, of which its major subsidiary, Beijing Overseas Chinese Town (29.28% stake), made 137.19 million RMB, accounting for about 20%. As a comparison, Huaqiaocheng-A directly acquired investment income of 297.05 million RMB from its joint-operated company Overseas Chinese Town Real Estate (40% stake) in 2006, which exceeded 50% of its total net profit of that year. However, under China’s old ASBE before 2007, the net profit from subsidiaries could be classified as consolidated operating income, while the investment income from joint-operated com- panies could not be classified as consolidated operating income, which was obviously contrary to the firm’s actual strategies. Furthermore, Huaqiaocheng-A could use the proportional consolidation method to incorporate its joint ventures, Shenzhen World Window and Jinxiu Zhonghua (49% stake for both), into its consolidated statements before 2007, but the ASBE No. 33-Consolidated Financial Statements (2006) specifies in its gui- dance that a proportional consolidation method is replaced by an equity method when listed firms incorporate joint ventures into their consolidated statements. In such condi- tions, the income statement could not reflect firm operation results properly if investment income were not reclassified accordingly. Based on such background, one of the important changes in China’s new ASBE in 2007 is reclassifying items in the income statement, and moving investment income from below the line of operating income to above the line. Operating income is the most important and stable profit for firms as it is obtained in recurring production and opera- tion activities. Only if operating income truly reflects management’s efforts is it viewed as an effective indicator in an executive compensation contract. The reason why investment income is not included in operating income under old ASBE is that the value driver behind investment income is quite different from that behind traditional operating income. Investment income is less persistent. As business practice changes, external equity investment has gradually become an important way for firms to expand, diversify and transform strategically. After the reclassification of income statement items, all of the equity investment income for business strategies except for short-term spreads is regarded as operating income, which not only conforms to the asset-liability view, but also embodies the substance-over-form principle. If the reclassification of income state- ment items really reflects the actual demand, we can predict that the rational board of 412 J. CHEN AND J. FANG directors will increase the weight of investment income in a compensation contract. Even if the board of directors fails to know the underlying cause of the reclassification of income statement items, the natural dependence on operating income may lead to the weight adjustment of performance indicators in the compensation contract. We construct a sample of listed firms in China from 2001 to 2017. We find that the sensitivity of executive compensation and investment income increases significantly after 2007, suggesting that firms adjust the weight of compensation performance indicators in response to the reclassification of income statement items in the new ASBE. Further tests show that this relationship is more pronounced in firms with higher accounting sensitivity and in private firms. If the new ASBE does meet the actual demand and affects the design of a compensation contract, management would exert more efforts in external invest- ment activities and improve the quality of investment income. In this way, the difference between the value relevance of investment income and that of traditional operating income would be reduced. Consistent with our prediction, the persistence of investment income is significantly improved. More specifically, the persistence of investment income is significantly lower than that of traditional operating income before 2007, but the difference between them becomes statistically insignificant after 2007. The above evi- dence indicates that the weight adjustment of compensation performance indicators is a proper response to the changes in value driver of profits. We conduct a further test to explore how the weight adjustment of compensation performance indicators affects management’s behaviour. As the sensitivity of executive compensation and investment income increases, management has to pay more attention to investment forms. By investment objectives, investment can be divided into long-term equity investment and financial asset investment, in which the former aims at controlling or significantly impacting with a longer holding period and higher holding cost, while the latter aims to obtain short-term profits with a shorter holding period and lower holding cost. Therefore, self-interested management may invest more in financial assets to obtain higher pay. Consistent with the prediction, we find that financial asset investment increases significantly after 2007. We also find that financial asset investment has a limited effect on firm value, which may indicate that the management’s investment in financial assets is not for the purpose of improving firm value, but more for compensation manipulation. This article makes three major contributions. First, ASBE is of great significance in standardising corporate accounting and management practice and in improving the effectiveness of the capital market. Therefore, the economic consequence of changes in accounting standards has been a typical topic in accounting and corporate governance research. For example, Lou et al. (2010) find that the new ASBE decreases the explanatory power of accounting earnings to cash dividends. Zhang et al. (2013) find that fair value gain/loss has no significant explanatory power to executive compensation. Jia and Zhang (2016) find that the sensitivity of net profit and executive compensation decreases while the sensitivity of net asset and executive compensation increases significantly after the new ASBE change from an income-expense view to an asset-liability view. The above research mainly focuses on the economic consequences of changes in the information content of financial statements, while this article examines the impact of changes in the disclosure format of financial statements. We also complement the work of Luo et al. CHINA JOURNAL OF ACCOUNTING STUDIES 413 (2018), which examines how investors react to a reclassification of income statement items. Second, proper accounting information disclosure of listed firms is the basis on which the capital market can operate in an orderly and efficient manner. In recent years, many more scholars have paid attention to the impact of the corporate disclosure format. For example, Hirst and Hopkins (1998), Maines and McDaniel (2000), and Lee et al. (2006) focus on the different effect of income statement disclosure and equity statement disclosure. Riedl and Srinivasan (2010) and Chen and Schoderbek (2000) examine the different effects of statement disclosure and notes disclosure. And Luo et al. (2018) examine the effect of the reclassification of investment income. Unlike the above studies, which focus on how investors react to changes in disclosure format, we study the reaction of insiders, the board of directors, especially when making compensation contracts. Our results indicate that changes in the information disclosure format will alter corporate governance practice, which has important implications for regulators. Lastly, this paper finds that a weight adjustment of performance indicators leads to management’s opportunistic behaviour, which reminds shareholders and the board of directors to pay close attention to the economic essence of performance indicators and to be vigilant regarding management’s compensation manipulation. In addition, our research indicates that accounting standards will alter management’s decision-making, which has implications for policymakers when making and evaluating regulations. The remainder of the paper proceeds as follows. Section 2 reviews the literature and develops the hypotheses. Section 3 describes our sample and research design. Sections 4, 5 and 6 report the empirical results, and Section 7 concludes. 2. Literature review and research hypotheses 2.1. Accounting information and executive compensation contracts A firm is a nexus of contracts. Accounting information, as a comprehensive reflection of operating results and financial conditions, plays an important role in the formulation, implementation and supervision of contracts (Jensen & Meckling, 1976; Watts & Zimmerman, 1986). The contractual usefulness of accounting information has always been a classic topic in financial accounting, corporate governance and capital market research (Jensen & Murphy, 1990; Lu et al., 2008; Sun et al., 2006). Among all corporate contracts, the executive compensation contract is the main governance mechanism to align the goal of shareholders and management and reduce the principal–agent cost. However, the information asymmetry makes it either impossible or excessively costly for shareholders to observe management’s efforts. Therefore, corporate performance is used as a signal to convey the information about management’s efforts in executive compen- sation contracts (Du & Wang, 2007; Fang, 2009; Holmstrom, 1979; Murphy, 1985; Sloan, 1993). Corporate performance indicators with a high signal-to-noise ratio are often selected in compensation contracts so that they can convey more information with lower noise (Banker & Datar, 1989). Therefore, the ideal compensation contract has to strike a balance between relevance and reliability when choosing and weighting perfor- mance indicators. 414 J. CHEN AND J. FANG Accounting information is reliable, accurate and comparable, and is most widely used in executive compensation contracts (Angelis & Grinstein, 2015; Li et al., 2013). Since accounting information is relatively less noisy, whether or not to select an accounting performance indicator depends on how much information it delivers. Earlier studies find that net profit is an aggregate indicator with high information content (Ball & Brown, 1968; Beaver, 1968) and is most commonly used in executive compensation contracts (Li et al., 2013). However, later studies state that it is not enough to focus only on the aggregate earnings. Each component of earnings contains a different piece of informa- tion, and the sum of components provides incremental information than the aggregate earnings (Fairfield et al., 1996; Lipe, 1986; Ohlson & Penman, 1992; Strong & Walker, 1993). In practice, firms often use a variety of accounting indicators. Li et al. (2013) collected 228 compensation contracts in China from 2004 to 2010, and found that each compensation contract contains 4.5 performance indicators, among which the most commonly used accounting indicators are profit, operating income, ROE, etc. The statistics of firms in the UK and US also suggest that earnings indicators are most frequently used, such as EPS, net profit, operating income, etc. (Angelis & Grinstein, 2015; Conyon et al., 2000; Ittner et al., 1997). Furthermore, performance indicators are given weight according to their sensitivity to executive behaviour, which is usually measured by pay–performance sensitivity in empiri- cal research. The existing studies basically find that there is a significant positive correla- tion between executive compensation and corporate performance (Leone et al., 2006; Murphy, 1985). And the pay–performance sensitivity gradually increases as China’s execu- tive compensation contracts become more market-oriented. For example, Fang (2009) finds that there is a significant positive correlation between executive compensation and net profit, and that executive compensation is asymmetrically sticky. Wan (2014) further finds that the stickiness between executive compensation and operating income is weak, whereas the stickiness between executive compensation and non-operating income is strong. Zou et al. (2010) find that after fair value is introduced into the new ASBE, the pay of CFOs, other than that of CEOs and chairmen, is correlated with fair value gain/loss. To sum up, performance indicators are selected and given weight based on the extent that they reflect management’s efforts. In particular, operating income is one of the most important indicators in compensation contracts since it is generated from firms’ regular production and operation (Lu & Jiang, 2012; Murphy, 2001). On the contrary, the non- operating income is given less weight as it is greatly affected by an external environment and contingency factors. 2.2. Accounting standards changes and accounting information function Valuation and contracting, which are two basic functions of accounting information, require different information quality. These two functions are revolutionarily impacted by China’s new ASBE in 2007, which converges to IFRS (International Financial Reporting Standards). In terms of valuation function, Zhu et al. (2009) find that the introduction of fair value in the new ASBE does not improve the value relevance of accounting earnings. Zhang et al. (2013) also find that fair value gain/loss has no explanatory power to stock returns. Lu and Zhang (2009) find that after the new ASBE, the difference between the net profit of consolidated statements and parent company statements can provide additional CHINA JOURNAL OF ACCOUNTING STUDIES 415 information beyond that of consolidated statements only. Bu and Ye (2009) find that the value relevance of asset impairment improves after the new ASBE. In terms of the contractual function, on the one hand, the new ASBE reduces the space of earnings management and improves the reliability of accounting information. On the other hand, its introduction of fair value may weaken the contractual usefulness of accounting information (Chen, 2014). Yuan et al. (2013) find that when the new ASBE shifts focus from contractual function to valuation function, accounting earnings are less reliable and thus poorly explain the obtaining of bank loans. Chen (2014) finds that since the new ASBE makes accounting information less reliable, accounting performance indicators are partly replaced by market performance indicators in compensation con- tracts. On the contrary, Luo and Pang (2014) find that the new ASBE improves the quality of accounting information and increases the sensitivity of executive pay and accounting performance. Jia and Zhang (2016) find that with the new ASBE shifts from the income- expense view to the asset-liability view, the sensitivity of net profit and compensation decreases significantly, while the sensitivity of net assets and compensation increases significantly. Zhang et al. (2013) find that the adjusted fair value gain/loss has no explanatory power on executive compensation. In addition, the new ASBE also revises the way financial statements report, such as the repositioning of investment income, minority shareholders’ equity and minority share- holders’ gain/loss. Previous studies demonstrated that the repositioning of accounting items may affect their valuation function. For example, Zhang and Zhang (2008) find that the repositioning of minority shareholders’ equity and minority shareholders’ gain/loss makes consolidated financial statements more informational. Bartov and Mohanram (2014) find that gain/loss from early debt extinguishments is taken into account by investors after it is moved from below to above the line. Luo et al. (2018) find that investment income is used for earnings management after it is moved from below to above the line, but investors can’t see through it. In spite of considerable evidence about how format changes in financial statements affect their valuation function, the impact of format changes in financial statements on their contractual function remains to be studied. 2.3. Hypothesis development In 2006, the ASBE No. 30 – Presentation of Financial Statements issued by the Ministry of Finance of People’s Republic of China revised the presentation format of the income statement. First, the new ASBE no longer distinguishes between the primary operating income and non-primary operating income, but combines them as ‘operating income’. The underlying reason is that as firms are constantly expanding and increasingly diversi- fying, the boundary between primary operating income and non-primary operating income is gradually blurring. Second, the fair value gain/loss is added, and the investment income is moved from below the line of operating income to above it. Since external equity investment is becoming a regular way for firms to expand, part of capital income is recognised as operating income in line with the asset-liability view. Finally, the earnings per share is added to help investors and other information users to evaluate firms’ profitability and growth potential. This article focuses on whether insiders, the board of directors, adjust the weight of compensation performance indicators when income 416 J. CHEN AND J. FANG statement items are reclassified. As fair value gain/loss is not available in the pre- regulation period, we consider investment income which is both available before and after the new ASBE. After the reclassification of income statement items, the board of directors may adjust the weight of compensation performance indicators actively or passively. On the one hand, the board of directors is not only familiar with business practice, but also has a keen insight into the policy intention. The board of directors may realise that investment income is becoming an important part of operating income and conveys more incre- mental information about management’s efforts. Therefore, it will increase the weight of investment income in the compensation contract. On the other hand, even if the board of directors is not aware of the motivation of reclassifying investment income, it may also increase the weight of investment income due to the functional fixation to operating income. We expect that the weight given to income investment is significantly increased after the new ASBE. The hypothesis is proposed as follows: H1. Ceteris paribus, the sensitivity of executive compensation and investment income significantly increases after the implementation of the new ASBE in 2007. 3. Research design 3.1. Model specification Following prior literature (Fang, 2009; Leone et al., 2006), this article uses the change model to test the sensitivity of ΔPAY and ΔPERFORMANCE. The specific model is as follows: PAY CH ¼ αþ β � SALE þ β � SALE þ β � LEV þ β � BM þ β � BH þ β t t t t t t 1 2 3 4 5 6 � PRIVATE þ β � AGE þ β � MINDEX þ β � RETURN þ β t t t t 7 8 9 10 � CORE CH þ β � INVESTINCOME CH þ β � OTHERINCOME CH t t t 11 12 (1) þ β � POST 2007 þ β � RETURN � POST 2007 þ β � CORE CH t t t t 13 14 15 � POST 2007 þ β � INVESTINCOME CH � POST 2007 þ β t t t 16 17 � OTHERINCOMR CH � POST 2007 þ γþ ηþ ε t t where PAY is executive compensation, defined as the natural logarithm of the mean value of ‘total compensation of top three executives’ disclosed in annual reports. PERFORMANCE refers to corporate performance and is divided into three components that are core income (CORE), investment income (INVESTINCOME) and other income (OTHERINCOME), respectively. INVESTINCOME was presented below operating income before 2007, but above operating income after 2007. CORE refers to operating income prior to 2007 and operating income excluding investment income and fair value gain/loss after 2007. OTHERINCOME is derived from total profit less core income and investment income. All of the performance indicators are standardised by total assets. In particular, the above variables are suffixed with ‘_CH’ in the change model, indicating the difference between the current value and the previous value. In addition, the model includes the capital market performance (RETURN), which is the annual stock return. CHINA JOURNAL OF ACCOUNTING STUDIES 417 Table 1. Variable definition. Variable Definition PAY The logarithm of mean value of total compensation of top three executives CORE Current core income divided by total assets INVESTINCOME Current investment income divided by total assets OTHERINCOME Current other income divided by total assets POST2007 A dummy variable that equals to 1 in 2007 and beyond, otherwise 0 SIZE The logarithm of total assets SALE The logarithm of operating income SALE The square of logarithm of operating income LEV Current liabilities divided by total assets ROA Net profit divided by total assets OCF Operating cash flow divided by total assets BM Book value divided by market value BH A dummy variable equals to 1 if a firm issue A and B share simultaneously or A and H share simultaneously, otherwise 0 ATURN Total asset turnover PRIVATE A dummy variable that equals to 1 for private firms, otherwise 0 OWNERSHIP Largest shareholder ownership AGE The logarithm of firm age MINDEX Regional marketisation index – Wang et al. (2017) RETURN Annual stock return FIN_ASSET Financial assets investment divided by total assets GDP The logarithm of GDP M2 Broad money supply We control financial characteristics, equity characteristics, market environment and other firm characteristics. Definitions of control variables are presented in Table 1. We also control for the year fixed effect (γ) and the industry fixed effect (η). POST2007 is a dummy variable which equals 1 in the post-2007 period, otherwise 0. Our primary variable of interest is the interactive term of INVESTINCOME CH and POST 2007 . We expect its coeffi - t t cient to be significantly positive, which indicates that the sensitivity of executive com- pensation and investment income is significantly increased after 2007. 3.2. Data source and descriptive statistics We collected listed firms in A-share, B-share and the growth-enterprise market from 2001 to 2017 whose stock return and financial data are obtained from CSMAR (China Stock Market & Accounting Research) databases. After excluding financial firms and observa- tions with missing values, our final sample consists of 26,705 firm-year observations. All continuous variables are winsorised at 1% and 99%. Table 2 provides descriptive statistics of variables. The mean (median) value of PAY is 0.125 (0.068) with the standard deviation 0.409, which indicates a highly-dispersed and right-skewed distribution of executive compensation. The mean value of POST2007 is 0.761, indicating that there are more observations in post-2007 period. The mean (med- ian) value of RETURN is 0.057 (−0.031), and the standard deviation is 0.503, which indicates that stock return is highly heterogeneous and right-skewed. In addition, the skewness of CORE, INVESTINCOME and OTHERINCOME is relatively low since they are change variables. Table 3 tabulates the correlation matrix of the variables. CORE, INVESTINCOME and OTHERINCOME are significantly positively correlated. The coefficients on all performance indicators and PAY are significantly positive, indicating that there is significant sensitivity of executive compensation and each component of corporate performance. 418 J. CHEN AND J. FANG Table 2. Descriptive statistics. Variable N Mean SD P25 P50 P75 PAY_CH 26,705 0.125 0.409 −0.026 0.068 0.253 POST2007 26,705 0.761 0.426 1 1 1 SALE 26,705 21.080 1.531 20.140 21.030 21.970 LEV 26,705 0.074 0.102 0 0.026 0.112 BM 26,705 0.439 0.473 0.212 0.354 0.559 BH 26,705 0.089 0.285 0 0 0 PRIVATE 26,705 0.474 0.499 0 0 1 AGE 26,705 2.208 0.593 1.792 2.303 2.708 MINDEX 26,705 0.710 0.294 0.556 0.778 1 RETURN 26,705 0.057 0.503 −0.217 −0.031 0.221 CORE_CH 26,705 0.006 0.059 −0.014 0.004 0.021 INVESTINCOME_CH 26,705 0.001 0.019 −0.001 0 0.003 OTHERINCOME_CH 26,705 0.002 0.034 −0.003 0.001 0.006 4. Main empirical results 4.1. Baseline empirical results Table 4 reports the baseline regression results. Column 1 shows that the coefficients on RETURN and CORE_CH are significantly positive at the 1% level, while the coefficients on INVESTINCOME_CH and OTHERINCOME_CH are not significant. In columns 2 and 3 we present the results of subsamples of pre- and post-2007 periods respectively. It turns out that the coefficient on INVESTINCOME_CH is negative and not significant before 2007, and significantly positive at the 1% level after 2007. In column 4, we add the interaction variables between POST2007 and each performance indicator to differ between pre- and post-2007 periods. We find that while the coefficients on CORE_CH *POST2007 , OTHERINCOME_CH *POST2007 and RETURN *POST2007 are insignificant or t t t t t significantly negative, the coefficient on INVESTINCOME_CH *POST2007 is significantly t t positive. In column 5, we use a consistent sample that only contains firms listed before 2000, and we find that the sensitivity of executive compensation and market perfor- mance, core income and other income has no difference between the pre- and post- 2007 periods, while the sensitivity of compensation and investment income significantly improves, which is consistent with our prediction. Other income is treated as a comparison since it is below the line in both the pre- and post-2007 periods. It is noted that there is no significant correlation between OTHERINCOME_CH and PAY_CH in t t columns 1 through 3, and the coefficient on OTHERINCOME_CH*POST2007 is not statisti- cally significant in columns 4 and 5, which further excludes the possible explanation that our findings are the result of changes in the economic environment. Among control variables, the coefficient on SALE is significantly negative, and the coefficient on SALE is significantly positive, which indicates that there is an inverted U-shaped relationship between firm size and executive compensation. The coefficient on PRIVATE is significantly positive, especially after 2007, which reveals that executive compensation in private firms is growing faster than that in state-owned firms given the policy of limiting executive pay in state-owned firms. The coefficient on BM is significantly negative, which means that firm growth is significantly positively related to executive compensation growth. The coefficient on AGE is significantly positive, indicating that the older the firm is, the faster executive compensation grows. CHINA JOURNAL OF ACCOUNTING STUDIES 419 Table 3. Correlation matrix of variables. PAY_CH POST2007 SALE LEV BM BH PRIVATE PAY_CH 1 POST2007 −0.076* 1 SALE −0.006 0.234* 1 LEV −0.005 0.039* 0.220* 1 BM −0.030* −0.119* 0.249* 0.162* 1 BH −0.004 −0.043* 0.1628* 0.079* 0.272* 1 PRIVATE −0.015* 0.176* −0.2497* −0.178* −0.178* −0.161* 1 AGE −0.005 0.227* 0.1747* 0.167* 0.044* 0.159* −0.239* MINDEX −0.014* 0.044* 0.084* −0.110* 0.017* 0.136* 0.124* RETURN 0.066* 0.148* 0.008 −0.009 −0.166* −0.014* 0.043* CORE CH 0.121* 0.005 0.025* −0.036* −0.070* 0.003 0.029* INVESTINCOME_CH 0.020* 0.026* −0.004 −0.009 −0.021* 0.004 0.007 OTHERINCOME_CH 0.015* 0.039* −0.013* −0.015* −0.026* 0.006 0.009 AGE MINDEX RETURN CORE_CH INVESTINCOME_CH OTHERINCOME_CH AGE 1 MINDEX −0.066* 1 RETURN −0.002 0.004 1 CORE_CH 0.022* 0.004 0.159* 1 INVESTINCOME_CH 0.019* 0.014* 0.053* 0.131* 1 OTHERINCOME_CH 0.007 0.008 0.061* 0.112* 0.538* 1 420 J. CHEN AND J. FANG Table 4. Regression of income statement item reclassification on pay-performance-sensitivity. (1) (2) (3) (4) (5) Variable PAY_CH PAY_CH PAY_CH PAY_CH PAY_CH t t t t t Pre-2007 Full sample period Post-2007 period Full sample Consistent sample SALE 0.160*** 0.150* 0.073** 0.156*** 0.138*** (6.251) (1.921) (2.481) (6.090) (4.024) SALE −0.004*** −0.003 −0.002** −0.003*** −0.003*** (−5.926) (−1.511) (−2.295) (−5.776) (−3.781) LEV −0.025 −0.082 0.020 −0.021 −0.021 (−1.103) (−1.307) (0.823) (−0.954) (−0.680) BM −0.018*** −0.001 −0.029*** −0.018*** −0.010* (−3.134) (−0.151) (−3.780) (−3.116) (−1.814) BH −0.011* −0.009 −0.018*** −0.010* −0.011 (−1.780) (−0.546) (−2.811) (−1.704) (−1.580) PRIVATE 0.022*** −0.016 0.034*** 0.021*** 0.019*** (5.476) (−1.355) (7.559) (5.325) (3.089) AGE 0.020*** 0.023* 0.023*** 0.020*** 0.019 (6.502) (1.842) (6.980) (6.308) (1.002) MINDEX −0.022*** −0.017 −0.027*** −0.022*** −0.030*** (−3.822) (−1.000) (−3.910) (−3.738) (−3.486) RETURN 0.025*** 0.037*** 0.015** 0.040*** 0.037** (3.837) (3.229) (2.021) (3.467) (2.574) CORE_CH 0.775*** 0.642*** 0.884*** 0.663*** 0.603*** (12.830) (4.708) (12.819) (4.874) (3.960) INVESTINCOME_CH 0.219 −0.510 0.602*** −0.565 −0.625 (1.099) (−1.120) (2.659) (−1.242) (−1.270) OTHERINCOME_CH −0.094 −0.211 −0.164 −0.221 −0.258 (−0.766) (−0.744) (−1.202) (−0.783) (−0.856) RETURN *POST2007 −0.023* −0.025 t t (−1.743) (−1.269) CORE_CH *POST2007 0.225 0.248 t t (1.496) (1.397) INVESTINCOME_CH *POST2007 1.156** 1.309** t t (2.288) (2.291) OTHERINCOME_CH *POST2007 0.067 −0.171 t t (0.217) (−0.507) CONSTANT −1.499*** −1.574* −0.800** −1.453*** −1.266*** (−5.418) (−1.941) (−2.515) (−5.254) (−3.435) YEAR YES YES YES YES YES INDUSTRY YES YES YES YES YES adj. R 0.042 0.037 0.042 0.042 0.037 Obs 26,705 63,80 20,325 26,705 13,703 *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively. Robust t-statistics are in parentheses. 4.2. Robustness tests 4.2.1. Firm fixed effect We introduce the firm fixed effect to model (1) to get rid of the impact of firm hetero- geneity. The results are displayed in Table 5 are consistent with Table 4 4.2.2. Partition by investment income It is concerning that the above findings are caused by the growing proportion of invest- ment income after the basic completion of Non-tradable Shares Reform in 2007, but not the new ASBE as we stated. In that way, our results can only be found in firms with a high proportion of investment income. We partition our sample into two groups via the median value of INVESTINCOME in 2001–2006. As tabulated in Table 6, the coefficients on INVESTINCOME_CH*POST2007 are both significantly positive in two groups, while the CHINA JOURNAL OF ACCOUNTING STUDIES 421 Table 5. Regression of firm fixed effect model. (1) (2) (3) (4) (5) Variable PAY_CH PAY_CH PAY_CH PAY_CH PAY_CH t t t t t Pre-2007 Full sample period Post-2007 period Full sample Consistent sample RETURN 0.021*** 0.033** 0.010 0.038*** 0.035** (3.101) (2.466) (1.236) (3.176) (2.280) CORE_CH 0.762*** 0.519*** 0.848*** 0.660*** 0.603*** (12.213) (3.445) (12.084) (4.650) (3.791) INVESTINCOME_CH 0.235 −0.325 0.637*** −0.521 −0.595 (1.159) (−0.664) (2.760) (−1.099) (−1.159) OTHERINCOME_CH −0.105 −0.135 −0.214 −0.194 −0.240 (−0.837) (−0.457) (−1.540) (−0.664) (−0.770) RETURN *POST2007 −0.026* −0.027 t t (−1.884) (−1.363) CORE_CH *POST2007 0.208 0.245 t t (1.324) (1.337) INVESTINCOME_CH *POST2007 1.117** 1.279** t t (2.122) (2.154) OTHERINCOME_CH *POST2007 0.020 −0.198 t t (0.064) (−0.564) CONSTANT −2.795*** −7.295*** 0.266 −2.701*** −2.912*** (−4.341) (−2.959) (0.280) (−4.185) (−3.692) FIRM YES YES YES YES YES YEAR YES YES YES YES YES adj. R 0.037 0.030 0.036 0.038 0.038 Obs 26,705 63,80 20,325 26,705 13,703 *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. Table 6. Subsample regression partitioned by investment income. (1) (2) Variable PAY_CH PAY_CH t t Low investment income High investment income RETURN 0.048*** 0.009 (2.688) (0.617) CORE_CH 0.817*** 0.540*** (4.179) (2.785) INVESTINCOME_CH −1.085 −0.500 (−1.141) (−0.908) OTHERINCOME_CH 0.065 −0.463 (0.142) (−1.331) RETURN *POST2007 −0.026 0.005 t t (−1.138) (0.258) CORE_CH *POST2007 0.149 0.235 t t (0.657) (1.067) INVESTINCOME_CH *POST2007 1.853* 1.098* t t (1.778) (1.740) OTHERINCOME_CH *POST2007 −0.357 0.330 t t (−0.729) (0.836) CONSTANT −1.623*** −0.904** (−3.556) (−2.063) YEAR YES YES INDUSTRY YES YES adj. R 0.035 0.020 Obs 9816 10,124 *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. 422 J. CHEN AND J. FANG coefficients on OTHERINCOME_CH*POST2007 are both insignificant. Hence, it is the new ASBE that improves the sensitivity of executive compensation and investment income. 4.2.3. Dynamic effect test Following Bertrand and Mullainathan (2003), we conduct a dynamic effect test to elim- inate the possible interference of the time trend. Specifically, we replace the dummy variable POST2007 with six dummy variables: POST1 equals one for the year of 2007 when new ASBE is put into effect, POST2 equals one for the year of 2008, POST3 equals one for the year of 2009 and beyond, and BEFORE1, BEFORE2 and BEFORE3 equal one for the years of 2006, 2005 and 2004, respectively. The results are displayed in Table 7. The coefficients on interactive terms between INVESTINCOME and BEFORE1, BEFORE2, BEFORE3 are not statistically significant, while the coefficients on interactive terms between INVESTINCOME and POST2 and POST3 is significantly positive, which further excludes the influence of the time trend. 4.2.4. Cross-sectional test To further reveal the conditions in which the above relationship is more pronounced, we conduct several cross-sectional tests. First, we examine how firm accounting sensitivity affects the above relationship. If the reclassification of income statement items in new ASBE leads to increased sensitivity of executive compensation and investment income, then firms that are more sensitive to changes in accounting standards are expected to react more. For this reason, we calculate firm accounting sensitivity following Yuan et al. (2013), Jia and Zhang (2016). Specifically, in the annual report of 2007, which is the first year of the new ASBE implementation, listed firms are required to disclose net profit of 2006 adjusted in accordance with the new ASBE. The impact of the new ASBE on firm accounting practice can be derived from the Table 7. Regression of dynamic effect. (1) Variable PAY_CH INVESTINCOME_CH *BEFORE1 0.621 t t (0.792) INVESTINCOME_CH *BEFORE2 −0.872 t t (−1.038) INVESTINCOME_CH *BEFORE3 0.656 t t (0.729) INVESTINCOME_CH *POST1 0.815 t t (1.141) INVESTINCOME_CH *POST2 2.055*** t t (2.644) INVESTINCOME_CH *POST3 1.061** t t (1.965) CONSTANT −1.348*** (−4.367) YEAR YES INDUSTRY YES adj. R 0.044 Obs 26,705 *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. CHINA JOURNAL OF ACCOUNTING STUDIES 423 change value between the retroactively adjusted net profit of 2006 and the original net profit of 2006 in line with the old ASBE. The calculation is presented as formula (2), where NI is the net profit calculated according to the new ASBE, and NI is the net profit new old calculated according to the old ASBE. The larger the value of SENSITIVITY, the greater the accounting sensitivity of the firm. We then partition the samples by the median value of accounting sensitivity. Columns 1 and 2 of Table 8 show that the coefficients on INVESTINCOME_CH*POST2007 are only significantly positive in the subsample with high accounting sensitivity, and the difference of coefficients between groups is significant. NI NI new old SENSITIVITY ¼ j j (2) NI old Next, China’s state-owned firms and private firms differ greatly in the business objectives and agency conflicts. They may be influenced by changes in accounting standards in different way. From the traditional point of view, since state-owned firms bear non- economic goals, the correlation between management compensation and firm perfor- mance is relatively low. In addition, with the absence of owners, compensation incentive plans are difficult to conduct efficiently. On the other hand, with the executive compensa- tion reform of state-owned firms, the executive compensation system of state-owned firms has been gradually improved (Xin & Tan, 2009). Besides, excessive executive com- pensation of state-owned firms has drawn wider public concern and stronger social supervision (Fang, 2009), which may prompt state-owned firms to react more to the Table 8. Subsample regression partitioned by accounting sensitivity. (1) (2) (3) (4) Variable PAY_CH PAY_CH PAY_CH PAY_CH t t t t Low accounting High accounting State-owned Private sensitivity sensitivity firms firms RETURN 0.029* 0.028 0.020 0.035 (1.720) (1.506) (1.326) (1.571) CORE_CH 0.359 0.715*** 0.786*** 0.321 (1.416) (3.691) (4.084) (1.305) INVESTINCOME_CH 0.410 −1.665*** −0.277 −1.803** (0.475) (−2.652) (−0.496) (−1.974) OTHERINCOME_CH −0.107 0.226 −0.602 0.804* (−0.182) (0.573) (−1.335) (1.719) RETURN *POST2007 −0.010 −0.019 −0.002 −0.025 t t (−0.461) (−0.848) (−0.130) (−0.966) CORE_CH *POST2007 0.414 0.040 0.108 0.327 t t (1.522) (0.190) (0.510) (1.247) INVESTINCOME_CH −0.463 2.405*** 0.630 2.344** *POST2007 (−0.490) (3.547) (0.979) (2.440) OTHERINCOME_CH 0.375 −0.468 0.675 −0.925* *POST2007 (0.594) (−1.059) (1.324) (−1.830) CONSTANT −1.429*** −1.149*** −1.113** −0.792 (−3.319) (−2.786) (−2.510) (−1.361) YEAR YES YES YES YES INDUSTRY YES YES YES YES adj. R 0.017 0.028 0.027 0.020 Obs 10,280 10,905 13,040 8145 Difference 2.868** 1.714 (6.11) (1.90) *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. 424 J. CHEN AND J. FANG new ASBE. Therefore, we do not make specific prediction on the results. As columns 3 and 4 of Table 8 show, the coefficient on INVESTINCOME_CH* POST2007 is only significantly positive in the subsample of private firms, but the difference of coefficients between groups is not significant. 5. Additional tests: is the weight adjustment of compensation performance indicators effective? The above results show that the sensitivity of executive compensation and investment income increases significantly after the implementation of the new ASBE in 2007. The next question is: is the weight adjustment of compensation performance indicators effective? If it is effective, then investment income is supposed to be more persistent after 2007. Based on this, we test the relationship between investment income and future accounting performance and also future operating cash flow. Table 9 presents the relationship between all performance indicators and firm future performance. In the full sample displayed in column 1, RETURN , CORE and INVESTINCOME have a significant positive effect on firm future t t t performance. Next, we present the results of subsamples of the pre- and post- 2007 periods in columns 2 and 3, respectively. We find that the coefficients on RETURN and CORE are both significant in columns 2 and 3, while the coefficient on t t INVESTINCOME is not significant before 2007 and is significantly positive after 2007. We further introduce interactive terms between all performance indicators and the POST2007 dummy in column 4. The coefficient on CORE *POST2007 is significantly negative and the t t coefficient on INVESTINCOME *POST2007 is significantly positive. The above results indi- t t cate that core income is less powerful in explaining future firm performance, whereas investment income is more powerful in explaining future firm performance after 2007. Table 10 shows the impact of all performance indicator on firm future cash flow. In the full sample displayed in column 1, the coefficients on RETURN and CORE are significantly t t positive, while the coefficients on INVESTINCOME and OTHERINCOME are significantly t t negative. Subsample results presented in columns 2 and 3 show that, although the coefficients on INVESTINCOME are both negative, the one after 2007 is, to some extent, larger. To further reveal the effect of the new ASBE, we introduce interactive terms between all performance indicators and the POST2007 dummy in column 4. It is found that the coefficient on INVESTINCOME *POST2007 is significantly positive, indicating that t t investment income is more powerful in explaining future cash flow. In summary, the earning persistence of investment earnings improves significantly after 2007, indicating that the weight adjustment of compensation performance indica- tors is in line with the economic essence and market demand. 6. Additional tests: is management opportunistic? 6.1. Compensation incentives and corporate investment Does the increased sensitivity of executive compensation and investment income lead to management’s opportunistic behaviour, that is, to raising compensation by increasing investment income? Investment income comes mainly from financial assets investment CHINA JOURNAL OF ACCOUNTING STUDIES 425 Table 9. Regression of investment income and future performance. (1) (2) (3) (4) Variable ROA ROA ROA ROA t+1 t+1 t+1 t+1 Pre-2007 Full sample period Post-2007 period Full sample SIZE −0.001 −0.002 −0.000 −0.001* (−1.465) (−1.573) (−0.405) (−1.778) LEV −0.006* −0.001 −0.005 −0.005 (−1.770) (−0.109) (−1.243) (−1.389) ATURN 0.006*** 0.007*** 0.005*** 0.006*** (5.067) (2.954) (3.967) (4.628) OCF 0.089*** 0.096*** 0.082*** 0.087*** (11.788) (6.651) (9.293) (11.536) BH 0.001 −0.004 0.003 0.001 (0.352) (−1.212) (1.506) (0.477) PRIVATE 0.000 −0.007*** 0.003*** 0.000 (0.316) (−3.423) (2.997) (0.336) OWNERSHIP 0.014*** 0.020*** 0.010*** 0.014*** (4.596) (3.624) (2.976) (4.690) MINDEX 0.009*** 0.013*** 0.007*** 0.009*** (5.428) (4.517) (3.476) (5.570) RETURN 0.013*** 0.011*** 0.014*** 0.012*** (11.788) (6.677) (10.187) (6.907) CORE 0.390*** 0.424*** 0.378*** 0.446*** (37.145) (21.968) (28.231) (26.953) INVESTINCOME 0.267*** −0.020 0.375*** −0.045 (6.076) (−0.228) (6.751) (−0.531) OTHERINCOME 0.055 −0.020 −0.009 −0.018 (1.553) (−0.233) (−0.217) (−0.214) RETURN *POST2007 0.002 t t (1.038) CORE *POST2007 −0.068*** t t (−3.491) INVESTINCOME *POST2007 0.417*** t t (4.156) OTHERINCOME *POST2007 0.020 t t (0.210) CONSTANT −0.003 0.008 0.012 0.000 (−0.290) (0.379) (0.981) (0.029) YEAR YES YES YES YES INDUSTRY YES YES YES YES adj. R 0.283 0.288 0.284 0.285 Obs 20,790 6234 14,556 20,790 *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. and long-term equity investment. Financial assets investment is an indirect investment to obtain value-added income, which is relatively easier to implement. And long-term equity investment is considered as a direct investment to obtain production and operation profits, of which the implementation cost is relatively higher. Our first step is to examine whether management prefers to invest in financial assets to increase investment income. Specifically, financial assets investment is derived from the sum of trading financial assets, available-for-sale financial assets and held-to-maturity financial assets, and long-term equity investment equals the long-term equity investment in the parent company bal- ance sheet. The above two variables are both normalised by firm total assets. Panel A of Table 11 show how the implementation of the new ASBE affects firm financial assets investment. It is shown that the coefficients on INVESTINCOME are significantly positive, which means that the current investment income will promote 426 J. CHEN AND J. FANG Table 10. Regression of investment income and future cash flow. (1) (2) (3) (4) Variable OCF OCF OCF OCF t+1 t+1 t+1 t+1 Pre-2007 Full sample period Post-2007 period Full sample SIZE 0.002*** 0.003** 0.002*** 0.002** (2.629) (2.122) (2.828) (2.387) LEV −0.006* 0.014** −0.012*** −0.005 (−1.712) (2.341) (−3.310) (−1.507) ATURN 0.005*** 0.001 0.007*** 0.005*** (3.188) (0.388) (3.608) (3.118) OCF 0.240*** 0.159*** 0.266*** 0.237*** (20.967) (7.417) (20.594) (20.684) BH 0.004* 0.007** 0.003 0.004* (1.786) (2.061) (1.166) (1.832) PRIVATE 0.001 −0.001 0.001 0.000 (0.576) (−0.381) (0.373) (0.229) OWNERSHIP 0.008* 0.015** 0.005 0.007* (1.911) (2.426) (1.087) (1.886) MINDEX 0.008*** 0.007** 0.008*** 0.008*** (3.435) (2.089) (3.151) (3.603) RETURN 0.005*** 0.006*** 0.006*** 0.005*** (4.427) (3.108) (3.627) (3.066) CORE 0.136*** 0.166*** 0.136*** 0.116*** (10.838) (7.635) (8.913) (6.654) INVESTINCOME −0.251*** −0.545*** −0.178*** −0.472*** (−5.951) (−7.470) (−3.310) (−6.693) OTHERINCOME −0.079** −0.225*** −0.088** −0.240*** (−2.419) (−3.225) (−2.182) (−3.456) RETURN *POST2007 0.000 t t (0.206) CORE *POST2007 0.043** t t (2.227) INVESTINCOME *POST2007 0.292*** t t (3.371) OTHERINCOME *POST2007 0.161** t t (2.040) CONSTANT −0.010 −0.049* −0.022 −0.005 (−0.756) (−1.811) (−1.477) (−0.345) YEAR YES YES YES YES INDUSTRY YES YES YES YES adj. R 0.182 0.137 0.206 0.183 Obs 20,790 6234 14,556 20,790 *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. the financial asset investment in the next year. The most interesting variable is POST_2007 in column 4, which is significantly positive, indicating that management invests more in financial asset after 2007. However, the coefficient on the interactive term INVESTINCOME *POST_2007 is not significant, indicating that the impact of investment income on future financial asset investment does not change significantly. Columns 1 and 2 also show that core income is negatively related to future financial asset investment, indicating that the higher the core income, the less the firm will invest in financial assets in the future. However, this relationship disappears after 2007, as shown in column 3. Panel B shows that investment income and future long-term equity investment are significantly positively correlated, indicating that investment income will encourage management to carry out long-term equity investment in the future. However, column 4 shows that POST_2007 is significantly negative, indicating that long-term equity CHINA JOURNAL OF ACCOUNTING STUDIES 427 Table 11. Regression of new accounting standards implementation on corporate investment. Panel A Whether to increase financial assets investment (1) (2) (3) (4) Variable Full sample Pre-2007 Post-2007 period Full sample period FIN_ASSET FIN_ASSET FIN_ASSET FIN_ASSET t+1 t+1 t+1 t+1 SIZE 1.936*** 1.755*** 1.918*** 2.163*** (20.916) (9.784) (18.646) (24.636) LEV −1.495*** −2.556*** −1.218** −2.322*** (−3.152) (−3.354) (−2.205) (−4.931) ATURN 0.174 −0.182 0.352 0.041 (0.845) (−0.504) (1.561) (0.200) CASHHOLDING 1.431** 8.639*** −0.397 0.752 (2.016) (5.958) (−0.511) (1.056) MB −0.013 −0.021 −0.010 0.078*** (−0.693) (−0.571) (−0.471) (4.313) GDP_CH - - - −55.731*** - - - (−13.917) M2 - - - −6.510*** - - - (−9.594) R54M 0.907*** 1.014*** 1.934*** 1.556*** (5.973) (6.241) (12.001) (25.033) BH 0.539 −0.386 1.018** 0.436 (1.453) (−0.703) (2.522) (1.182) PRIVATE −0.822*** −0.984*** −0.753*** −0.726*** (−3.864) (−2.831) (−3.205) (−3.429) OWNERSHI −3.121*** −3.786*** −3.174*** −3.546*** (−5.136) (−3.727) (−4.772) (−5.904) MINDEX 1.482*** 0.957* 1.740*** 1.573*** (4.215) (1.790) (4.494) (4.472) CORE −2.356* −6.514*** −1.213 −2.895 (−1.902) (−2.967) (−0.834) (−1.539) INVESTINCOME 39.951*** 43.235*** 39.093*** 44.587*** (8.892) (5.052) (7.569) (5.056) OTHERINCOME 1.816 −3.814 2.692 −0.638 (0.636) (−0.610) (0.789) (−0.099) POST2007 2.949*** (12.608) CORE *POST2007 −0.989 t t (−0.495) INVESTINCOME *POST2007 −3.190 t t (−0.332) OTHERINCOME *POST2007 1.504 t t (0.206) CONSTANT −31.173*** −26.740*** −29.192*** −31.380*** (−15.500) (−6.958) (−12.923) (−15.230) YEAR YES YES YES NO INDUSTRY YES YES YES YES adj. R 0.278 0.157 0.246 0.255 Obs 25,478 6304 19,174 25,478 Panel B Whether to increase long-term equities investment (1) (2) (3) (4) Variable Full sample Pre-2007 Post-2007 period Full sample period LONG_INVEST LONG_INVEST LONG_INVEST LONG_INVEST t+1 t+1 t+1 t+1 SIZE 0.993*** 1.036*** 0.982*** 0.972*** (35.684) (23.001) (32.280) (37.038) LEV −0.935*** −0.732*** −1.008*** −0.875*** (−6.858) (−3.817) (−6.336) (−6.555) ATURN −0.156*** −0.124 −0.177*** −0.147** (Continued) 428 J. CHEN AND J. FANG Table 11. (Continued). (−2.582) (−1.416) (−2.633) (−2.437) CASHHOLDING −0.755*** −0.192 −0.883*** −0.741*** (−3.690) (−0.583) (−3.818) (−3.648) MB 0.019*** 0.000 0.025*** 0.014** (3.237) (0.017) (3.656) (2.542) GDP_CH - - - 2.854*** - - - (2.789) M2_R - - - 0.595*** - - - (3.635) R54M −0.165*** −0.201*** 0.008 −0.154*** (−5.452) (−6.236) (0.346) (−10.011) BH 0.181* −0.081 0.287*** 0.195** (1.927) (−0.681) (2.696) (2.079) PRIVATE −0.090 −0.132 −0.086 −0.109* (−1.516) (−1.531) (−1.295) (−1.855) OWNERSHI −0.333* −0.741*** −0.164 −0.308* (−1.854) (−3.119) (−0.808) (−1.728) MINDEX 0.298*** 0.358*** 0.257** 0.291*** (3.127) (2.741) (2.410) (3.058) CORE −1.732*** −2.265*** −1.435*** −2.431*** (−4.932) (−4.136) (−3.434) (−4.724) INVESTINCOME 23.836*** 24.290*** 25.039*** 24.167*** (19.310) (14.369) (15.457) (14.382) OTHERINCOME −1.471* 2.665 −2.353** 2.822* (−1.778) (1.614) (−2.148) (1.738) POST2007 −0.698*** (−13.052) CORE *POST2007 1.089** t t (2.005) INVESTINCOME *POST2007 0.804 t t (0.395) OTHERINCOME *POST2007 −5.178*** t t (−2.631) CONSTANT −2.662*** −3.529*** −3.289*** −2.490*** (−4.544) (−3.760) (−4.795) (−4.161) YEAR YES YES YES NO INDUSTRY YES YES YES YES adj. R 0.382 0.335 0.397 0.381 Obs 20,697 5808 14,889 20,697 *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. investment declines significantly after 2007. In addition, the coefficients on CORE are significantly negative before and after 2007, indicating that the higher the core income, the less long-term equity investment will be carried out considering its high implementa- tion cost. 6.2. Financial asset investment and firm value The above results show that management is more willing to invest in financial assets than long-term equity investment after 2007. Is it because financial asset investment can better promote firm value, or just because it is easier to implement? The relationship between financial asset investment and firm value is further studied in the next step. Panel A of Table 12 tabulates the impact of financial asset investment on firm value. In the full sample regression presented in column 1, the coefficient on FIN_ASSET is sig- nificantly positive. Columns 2 and 3 present the subsample results of the pre- and post- CHINA JOURNAL OF ACCOUNTING STUDIES 429 Table 12. Regression of financial assets investment on firm value. Panel A financial assets investment and future performance (1) (2) (3) (4) Variable Full sample Pre-2007 Post-2007 period Full sample period ROA ROA ROA ROA t+1 t+1 t+1 t+1 SIZE −0.001 −0.002* −0.000 −0.001 (−1.228) (−1.782) (−0.217) (−1.205) LEV −0.008** −0.006 −0.006 −0.008** (−2.127) (−0.812) (−1.470) (−2.132) ATURN 0.006*** 0.006** 0.005*** 0.006*** (4.701) (2.451) (4.024) (4.706) OCF 0.087*** 0.093*** 0.083*** 0.087*** (11.383) (5.179) (9.705) (11.384) BH 0.001 −0.003 0.003 0.001 (0.627) (−0.885) (1.617) (0.623) PRIVATE 0.001 −0.008*** 0.004*** 0.001 (1.123) (−2.892) (3.400) (1.127) OWNERSHIP 0.015*** 0.027*** 0.010*** 0.015*** (4.713) (3.891) (3.086) (4.714) MINDEX 0.008*** 0.014*** 0.006*** 0.008*** (4.488) (3.691) (3.010) (4.490) RETURN 0.013*** 0.011*** 0.014*** 0.013*** (11.415) (6.151) (10.299) (11.417) CORE 0.386*** 0.402*** 0.381*** 0.386*** (34.018) (16.281) (28.610) (33.980) INVESTINCOME 0.280*** −0.058 0.390*** 0.281*** (5.918) (−0.568) (6.764) (5.919) OTHERINCOME 0.039 −0.039 −0.023 0.039 (1.012) (−0.380) (−0.511) (1.007) FIN_ASSET 0.034*** 0.024 0.033*** 0.024 (3.304) (0.709) (2.946) (0.701) FIN_ASSET *POST2007 0.012 t t (0.342) CONSTANT 0.017 0.060** −0.008 0.017 (1.541) (2.142) (−0.726) (1.520) YEAR YES YES YES YES INDUSTRY YES YES YES YES adj. R 0.284 0.284 0.290 0.284 Obs 18,871 4550 14,321 18,871 Panel B financial assets investment and future stock return (1) (2) (3) (4) Variable Full sample Pre-2007 Post-2007 period Full sample period RETURN RETURN RETURN RETURN t+1 t+1 t+1 t+1 SIZE −0.018*** 0.002 −0.021*** −0.019*** (−5.923) (0.187) (−7.197) (−6.059) LEV 0.099*** 0.153*** 0.092*** 0.102*** (5.128) (2.694) (5.094) (5.220) ATURN 0.003 −0.012 0.008 0.002 (0.376) (−0.485) (1.159) (0.315) OCF 0.201*** 0.336** 0.146*** 0.200*** (3.899) (2.244) (3.092) (3.896) BH −0.034*** −0.129*** −0.003 −0.034*** (−3.376) (−4.360) (−0.377) (−3.351) PRIVATE 0.021*** −0.017 0.028*** 0.021*** (3.103) (−0.771) (4.364) (3.075) OWNERSHIP 0.091*** 0.013 0.106*** 0.091*** (4.473) (0.216) (5.350) (4.454) MINDEX 0.006 −0.030 0.015 0.006 (Continued) 430 J. CHEN AND J. FANG Table 12. (Continued). (0.560) (−0.908) (1.533) (0.548) RETURN −0.112*** −0.152*** −0.091*** −0.112*** (−15.853) (−11.611) (−11.901) (−15.934) CORE −0.234*** −0.454** −0.184*** −0.231*** (−3.742) (−2.498) (−3.065) (−3.690) INVESTINCOME −0.374 −0.556 0.001 −0.383 (−1.591) (−0.968) (0.003) (−1.623) OTHERINCOME 0.060 0.431 −0.207 0.067 (0.304) (0.537) (−1.117) (0.341) FIN_ASSET 0.253*** 0.981*** 0.149** 0.844*** (3.155) (2.743) (2.152) (2.589) FIN_ASSET *POST2007 −0.688** t t (−2.090) CONSTANT 0.230*** 0.112 0.421*** 0.240*** (3.405) (0.444) (6.992) (3.541) YEAR YES YES YES YES INDUSTRY YES YES YES YES adj. R 0.103 0.095 0.125 0.104 Obs 18,871 4550 14,321 18,871 *, ** and *** denote significance at the 10%, 5% and 1% levels, respectively. Robust t-statistics are in parentheses. 2007 period. The coefficient on FIN_ASSET is not significant before 2007 and significantly positive after 2007. However, the value of the coefficient is 0.000, which is not significant. The coefficient on FIN_ASSET*POST2007 in column 4 is not statistically significant, which shows that the impact of financial asset investment on future performance does not improve after the new ASBE. Panel B of Table 12 presents the impact of financial asset investment on firm stock return. In columns 1 through 3, the coefficients on FIN_ASSET are all significantly positive, indicating that financial asset investment can significantly promote firm stock return. But the coefficient on FIN_ASSET*POST2007 in column 4 is significantly negative, which shows that the promotion effect of financial asset investment on stock return declines after 2007. Hence, the increase of financial asset investment may result from management’s com- pensation manipulation. 7. Conclusion The compensation contract is of great significance in improving corporate governance and alleviating agency problems. This article focuses on the impact of reclassification of income statement items caused by changes in accounting standards on the weight of compensation performance indicators. It is found that after the new ASBE moves invest- ment income from below the line to above it, the sensitivity between executive compen- sation and investment income increases significantly. Further tests show that the persistence of investment income indeed improves significantly after 2007, which reveals that the reclassification of investment income conforms to the change in business practice. Additional tests examine the real economic consequences of reclassification of investment income. It is found that, after 2007, management increases financial asset investment while firm value is less correlated with financial asset investment, which shows that the weight adjustment of performance indicators twists management behaviour to some extent. CHINA JOURNAL OF ACCOUNTING STUDIES 431 This article examines the economic consequences of the new ASBE, aiming at inter- preting the intention and examining the consequences of reclassifying income statement items in the new ASBE. The empirical results have important implications for supervisors to formulate and evaluate changes in accounting standards. At the same time, this article adds to the literature regrading the economic consequences of a change in information disclosure format, increasing the understanding of capital market efficiency. It also helps investors and the board of directors to understand the opportunistic behaviour of management driven by an executive compensation incentive. Finally, with the rise of behavioural finance, research in accounting information has gradually switched from disclosure quality to the pricing and governance functions. This study plays an important role in understanding the irrationality of capital markets. Disclosure statement No potential conflict of interest was reported by the authors. References Angelis, D.D., & Grinstein, Y. (2015). Performance terms in CEO compensation contracts. Review of Finance, 19(2), 619–651. https://doi.org/10.1093/rof/rfu014 Ball, R., & Brown, P. (1968). An empirical evaluation of accounting income numbers. Journal of Accounting Research, 6(2), 159–178. https://doi.org/10.2307/2490232 Balsam, S., Fernando, G., & Tripathy, A. (2011). The impact of firm strategy on performance measures used in executive compensation. Journal of Business Research, 64(2), 187–193. https://doi.org/10. 1016/j.jbusres.2010.01.006 Banker, R.D., & Datar, S.M. (1989). Sensitivity, precision, and linear aggregation of signals for performance evaluation. Journal of Accounting Research, 27(1), 21–39. https://doi.org/10.2307/ Bartov, E., & Mohanram, P.S. (2014). Does income statement placement matter to investors? The case of gains/losses from early debt extinguishment. The Accounting Review, 89(6), 2021 -2055. https:// doi.org/10.2308/accr-50839 Beaver, W.H. (1968). The information content of annual earnings announcements. Journal of Accounting Research, 6(Supplement), 67–92. https://doi.org/10.2307/2490070 Bertrand, M., & Mullainathan, S. (2003). Enjoying the quiet life? Corporate governance and manage- rial preferences. Journal of Political Economy, 111(5), 1043–1075. https://doi.org/10.1086/376950 Bu, D., & Ye, J. (2009). Economical consequences of asset impairment: Comparison of the old and new standards. China Accounting Review, 7(3), 315–328. (in Chinese). https://kns.cnki.net/kcms/ detail/detail.aspx?dbcode=CCJD&dbname=CCJDLAST1&filename=ZKJP200903009 Bushman, R., Indyejikian, R., & Indyejikian, R. (1996). CEO compensation: The role of individual performance evaluation. Journal of Accounting and Economics, 21(2), 161–194. https://doi.org/ 10.1016/0165-4101(95)00416-5 Chen, C. (2014). Research on international convergence of accounting standards and effectiveness of executive compensation contract——experience and data from China’s listed companies. Journal of Shanxi University of Finance and Economics, 36(5), 91–100. (in Chinese). http://doi. org/10.13781/j.cnki.1007-9556.2014.05.011 Chen, K., & Schoderbek, M. (2000). The 1993 tax rate increase and deferred tax adjustments: A test of functional fixation. Journal of Accounting Research, 38(1), 23–44. https://doi.org/10.2307/2672921 Conyon, M.J., Peck, S.I., Read, L.E., & Sadler, G.V. (2000). The structure of executive compensation contracts: UK evidence. Long Range Planning, 33(4), 478–503. https://doi.org/10.1016/S0024- 6301(00)00054-6 432 J. CHEN AND J. FANG Du, X., & Wang, L. (2007). Empirical research on correlation between compensation schemes of management and the change of performance of public listed companies. Accounting Research, (1), 58–65. (in Chinese). https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CJFD&dbname= CJFD2007&filename=KJYJ200701010 Fairfield, P.M., Sweeney, R.J., & Yohn, T.L. (1996). Accounting classification and the predictive content of earnings. The Accounting Review, 71(3), 337–355. https://www.jstor.org/stable/248292 Fang, J. (2009). Is top management compensation of Chinese public companies sticky? Economics Research Journal, 44(3), 110–124. (in Chinese). https://kns.cnki.net/kcms/detail/detail.aspx? dbcode=CJFD&dbname=CJFD2009&filename=JJYJ200903012 Hirst, D.E., & Hopkins, P.E. (1998). Comprehensive income reporting and analysts’ valuation judg- ments. Journal of Accounting Research, 36(supplement), 47–75. https://doi.org/10.2307/2491306 Holmstrom, B. (1979). Moral hazard and observability. The Bell Journal of Economics, 10(1), 74–91. https://doi.org/10.2307/3003320 Holmstrom, B., & Milgrom, P.R. (1991). Multi-task principal-agent analyses: Incentive contracts, asset ownership and job design. The Journal of Law, Economics, and Organization, 7(special issue), 524–552. https://doi.org/10.1093/jleo/7.special_issue.24 Huang, R., Marquardt, C., & Zhang, B. (2014). Why do managers avoid eps dilution? Evidence from debt–equity choice. Review of Accounting Studies, 19(2), 877–912. https://doi.org/10.1007/s11142- 013-9266-3 Ittner, C.D., Larcker, D.F., & Rajan, M.V. (1997). The choice of performance measures in annual bonus contracts. The Accounting Review, 72(2), 231–255. https://www.jstor.org/stable/248554 Jensen, M., & Meckling, W. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. https://doi.org/10.1016/ 0304-405X(76)90026-X Jensen, M., & Murphy, K. (1990). Performance compensation and top management incentives. Journal of Political Economy, 98(2), 225–264. https://doi.org/10.1086/261677 Jia, X., & Zhang, X. (2016). Accounting concept, compensation contract, and management decision- making behavior——the empirical evidence of unintended consequences of accounting stan- dards reforms. Accounting Research, (10), 18–25+96. (in Chinese). https://kns.cnki.net/kcms/ detail/detail.aspx?dbcode=CJFD&dbname=CJFDLAST2017&filename=KJYJ201610003 Lee, Y., Petroni, K., & Shen, M. (2006). Cherry picking, disclosure quality, and comprehensive income reporting choices: The case of property-liability insurers. Contemporary Accounting Research, 23 (3), 655–692. https://doi.org/10.1506/5QB8-PBQY-Y86L-DRYL Leone, A., Wu, J., & Zimmerman, J. (2006). Asymmetric sensitivity of CEO cash compensation to stock returns. Journal of Accounting and Economics, 42(1–2), 0–192. https://doi.org/10.1016/j.jacceco. 2006.04.001 Li, Y., Lou, F., Wang, J., & Yuan, H. (2013). A survey of executive compensation contracts on China’s listed companies. China Journal of Accounting Research, 6(3), 211–231. https://doi.org/10.1016/j. cjar.2013.06.001 Lipe, R.C. (1986). The information contained in the components of earnings. Journal of Accounting Research, 24(supplement), 37–64. https://doi.org/10.2307/2490728 Lou, F., Li, Y., & Yuan, H. (2010). Research on the effect of new accounting standards on the relationship between cash dividends and accounting earnings. Management World, (1), 122–- 132. (in Chinese). https://doi.org/10.19744/j.cnki.11-1235/f.2010.01.014 Lu, Y., & Jiang, Y. (2012). On the institutional change, earnings sustainability and market pricing behavior. Accounting Research, (1), 58–67. (in Chinese). https://kns.cnki.net/kcms/detail/detail. aspx?dbcode=CJFD&dbname=CJFD2012&filename=KJYJ201201010 Lu, Z., & Zhang, H. (2009). Accounting standards reform and the decision usefulness of subsidiary earnings information——empirical evidence from Chinese capital market. Accounting Research, (5), 20–28. (in Chinese). https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CJFD&dbname= CJFD2009&filename=KJYJ200905005 Lu, Z., Zhu, J., & Sun, B. (2008). Earnings management, accounting information and bank debt contract. Management World, (3), 152–158. (in Chinese). https://doi.org/10.19744/j.cnki.11-1235/ f.2008.03.017 CHINA JOURNAL OF ACCOUNTING STUDIES 433 Luo, J., & Pang, X. (2014). Has international accounting standards convergence in China improved the corporate executive sensitivity of performance-based salary empirical data from China’s A-share market. Contemporary Finance & Economics, (1), 117–128. (in Chinese). https://doi.org/ 10.13676/j.cnki.cn36-1030/f.2014.01.014 Luo, M., Shao, S., & Zhang, F. (2018). Does financial reporting above or below the operating income matter to firms and investors? The case of investment income. Review of Accounting Studies, 23(4), 1–37. https://doi.org/10.1007/s11142-018-9455-1 Maines, L.A., & McDaniel, L.S. (2000). Effects of comprehensive–income characteristics on nonpro- fessional investors’ judgments: The role of financial–statement presentation format. The Accounting Review, 75(2), 179–207. https://doi.org/10.2308/accr.2000.75.2.179 Marquardt, C., & Wiedman, C. (2005). Earnings management through transaction structuring: Contingent convertible debt and diluted earnings per share. Journal of Accounting Research, 43 (2), 205–243. https://doi.org/10.1111/j.1475-679x.2005.00168.x Murphy, K.J. (1985). Corporate performance and managerial remuneration: An empirical analysis. Journal of Accounting and Economics, 7(1–3), 0–42. https://doi.org/10.1016/0165-4101(85)90026- Murphy, K.J. (2001). Performance standards in incentive contracts. Journal of Accounting and Economics, 30(3), 245–278. https://doi.org/10.1016/S0165-4101(01)00013-1 Ohlson, J., & Penman, S. (1992). Disaggregated accounting data as explanatory variables for returns. Journal of Accounting, Auditing & Finance, 7(4), 553–573. https://doi.org/10.1177/ 0148558X9200700407 Riedl, E.J., & Srinivasan, S. (2010). Signaling firm performance through financial statement presenta- tion: An analysis using special items. Contemporary Accounting Research, 27(1), 289–332. https:// doi.org/10.1111/j.1911–3846.2010.01009.x Sloan, R.G. (1993). Accounting earnings and top executive compensation. Journal of Accounting and Economics, 16(1–3), 55–100. https://doi.org/10.1016/0165-4101(93)90005-Z Strong, N., & Walker, M. (1993). The explanatory power of earnings for stock returns. The Accounting Review, 68(2), 385–399. https://www.jstor.org/stable/248407 Sun, Z., Li, Z., & Wang, J. (2006). The nature of ownership, accounting information and debt contract: Empirical evidence from Chinese listed companies. Management World, (10), 100–107. (in Chinese). https://doi.org/10.19744/j.cnki.11-1235/f.2006.10.012 Wan, L. (2014). Financial crisis, performance controllability and salary stickiness. Macroeconomics, (5), 55–63. (in Chinese). https://doi.org/10.16304/j.cnki.11-3952/f.2014.05.011 Wang, X., Fan, G., & Yu, J. (2017). Marketization index of China’s provinces: NERI report 2016. Social Sciences Academic Press: Beijing, China. Watts, R.L., & Zimmerman, J.L. (1986). Positive accounting theory. Prentice Hall, Inc. Xin, Q., & Tan, W. (2009). Market-oriented reform, firm performance and executive compensation in Chinese state-owned enterprise. Economics Research Journal, 11(1), 90–121. (in Chinese). https:// kns.cnki.net/kcms/detail/detail.aspx?dbcode=CJFD&dbname=CJFD2009&filename= JJYJ200911009 Young, S., & Yang, J. (2011). Stock repurchases and executive compensation contract design: The role of earnings per share performance conditions. The Accounting Review, 86(2), 703–733. https:// doi.org/10.2308/accr.00000024 Yuan, H., Li, Y., Lou, F., & Zhang, Y. (2013). New Chinese accounting standards and the usefulness of accounting information in debt contracts. China Accounting and Finance Review, 15(1), 80–135. (in Chinese). https://doi.org/10.7603/s40570-013-0003-x Zhang, J., Xin, Q., & Tong, Y. (2013). The nature and consequence of gains and losses from changes in fair value——new finding from the stock return and executive compensation. Accounting Research, (8), 17–23+96. (in Chinese). https://kns.cnki.net/kcms/detail/detail.aspx?dbcode= CJFD&dbname=CJFD2013&filename=KJYJ201308003 Zhang, R., & Zhang, H. (2008). Economic consequences of the change on consolidated financial statement theory in the new accounting standards. Accounting Research, (12), 39–46. (in Chinese). https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CJFD&dbname=CJFD2008&filename= KJYJ200812006 434 J. CHEN AND J. FANG Zhu, K., Zhao, X., & Sun, H. (2009). Accounting standards reform, information accuracy and value relevance: Based on the empirical evidence of China’s accounting standards reform. Management World, (4), 47–54. (in Chinese). http://doi.org/10.19744/j.cnki.11-1235/f.2009.04.006 Zou, H., Xin, Q., & Zhang, J. (2010). Fair value measurement and executive compensation contract. Economic Science, 32(5), 102–110. (in Chinese). http://doi.org/10.19523/j.jjkx.2010.05.011
Journal
China Journal of Accounting Studies
– Taylor & Francis
Published: Jul 2, 2020
Keywords: Executive compensation contract; new accounting standards; investment income
