Abstract
China Journal of aCC ounting StudieS , 2017 Vol . 5, no . 1, 1–27 http://dx.doi.org/10.1080/21697213.2017.1304539 Whose cost of equity capital reduces after IFRS convergence and why? Heterogeneity evidence from Chinese stock market* a b c Min Chen , Xiaohui Qu and Xuejiao Sun a b d epartment of a ccounting, School of f inance and economics, Jimei university, China; Center for a ccounting Studies, Xiamen university, China; department of a ccounting, School of Business, t ianjin university of f inance and economics, China ABSTRACT KEYWORDS ifrS convergence; cost of This paper extends the studies of IFRS convergence in China by equity capital; heterogeneity; analysing heterogeneity with respect to the reduction in the cost institutional settings; of equity capital, and investigating the potential causes of that management incentives heterogeneity using the sample of listed A-share companies of 2004– 2010. This paper finds that state-owned companies have experienced a greater cost reduction than non-state-owned companies. The further analyses find that institutional settings and management incentives are critical driving factors for state-owned companies benefiting more from IFRS convergence. These findings not only enrich the literature by providing direct evidence to reveal and explain the heterogeneity in the reduction in the cost of equity capital caused by IFRS convergence in China, but also contribute to the appraisal of the convergence. To implement IFRS-converged accounting standards in transitional economies such as China, it is important to strengthen the development of external marketisation and the governance of incentives for earnings management. 1. Introduction On 15 February 2006, the Ministry of Finance (MOF) of China issued the revised Accounting Standards for Business Enterprises (ASBE), which has been implemented in all China’s listed companies commencing from 1 January 2007, and also officially alleged having substantially converged with the IFRS (International Financial Reporting Standards). Because of the prom- inent involvement of China’s economy with the world, especially the increasing cross-border investment, the CAS (Chinese Accounting Standards) convergence with IFRS aroused wide attention from academics and practitioners. The study of IFRS convergence in China could contribute to the literature and be crucial in defusing the concerns about the applicability of IFRS convergence to emerging markets. After the preparation of consolidated n fi ancial statements of listed companies in EU coun - tries that mandatorily followed IFRS from the year of 2005, the consequences of IFRS CONTACT Xiaohui Qu xhqu@xmu.edu.cn, janechm@jmu.edu.cn Paper accepted by Zhijun lin. t his paper also uses CaS (2006) to represent aSBe which is the acronyms of Accounting Standards for Business Enterprises officially named by the Mof of China. © 2017 a ccounting Society of China 2 M. CHEN ET AL. adoption in the EU have been examined, including accounting information quality, market reaction and the contractual influence. The issuance of CAS (2006) has triggered studies on IFRS convergence in China, including the influence of IFRS convergence on the cost of equity capital. However, few studies extend to heterogeneity in economic consequences. Moreover, it is emphasised that institutional factors and management incentives should not be ignored in the study of IFRS adoption or convergence (Cang, Chu, & Lin, 2014; He, Wang, & Young, 2012; Jeanjean & Stolowy, 2008), while almost no explicit evidence of influence of institu- tional factors and management incentives on IFRS adoption or convergence has been documented. The proponents of IFRS assert that IFRS will ‘reduce the cost of capital and open new opportunities for diversification and improved investment returns’ ( Tweedie, 2006a). Li (2010) argues that the effects of a reduction in the cost of equity capital depend on the strength of the legal enforcement. The contributions of this paper are threefold. (1) This paper analyses the heterogeneity of the influence of IFRS convergence on the cost of equity capital in China. According to our empirical results, state-owned companies experience a greater reduction in the cost of equity capital than non-state-owned companies. (2) The influence of institutional settings and management incentives on the heterogeneity of IFRS convergence is evidenced, which not only extends the existing study empirically, but also theoretically emphasises the significant influence of external and internal driving factors for institutional implementation. (3) Above all, the conclusions contribute to the appraisal of the convergence, which has meaningful policy implications for transitional economies. The findings of this paper highlight the critical factors to implement IFRS-converged accounting standards in a transitional economy where there is a strong influence from ‘culture and historical roots’ (Radebaugh & Gray, 1997). It is important for regulators to promote an external marketisation process, and to govern earn- ings management incentives. The remainder of this paper is organised as follows: Section 2 reviews the prior studies and discusses the study motivation; Section 3 develops the research hypotheses and empir- ical models; Section 4 shows the research design; Section 5 presents the main empirical findings; Section 6 reports the results of robustness tests; and Section 7 concludes. 2. Literature review and motivation 2.1. Examination of endorsement approach Scholars have taken this opportunity of IFRS adoption in EU to study the impact from per- spectives such as information quality, market reaction and contractual outcome. Although Jeanjean and Stolowy (2008) find evidence of increased earnings management in France, and Atwood, Drake, Myers, and Myers (2011) find less persistent earnings, many more studies document improved accounting information quality after IFRS adoption, in terms of earnings quality, value relevance, disclosure quality and comparability. Reduced earnings management and more timely loss recognition demonstrate higher earnings quality (Barth, Landsman, & Lang, 2008; Hung & Subramanyam, 2007; Iatridis & Rouvolis, 2010). Prior researchers also have provided evidence that IFRS adoption has led to higher value relevance (Bartov, Goldberg, & Kim, 2005; Barth et al., 2008; Capkun, Cazavan-Jeny, Jeanjean, & Weissc, 2008; Iatridis & Rouvolis, 2010), increased disclosure quality (Daske & Gebhardt, 2006) and CHINA JOURNAL OF ACCOUNTING STUDIES 3 improved comparability (DeFond, Hu, Hung, & Li, 2011). Using a comprehensive research setting, Chen, Tang, Jiang, and Lin (2010) demonstrate that the majority of accounting quality indicators improved after IFRS adoption in the EU. More studies examine the capital market influences in terms of market reaction, cost of capital, market liquidity and market efficiency. Both positive and negative market reactions are documented (Armstrong, Barth, Jagolinzer, & Riedl, 2010; Christensen, Lee, & Walker, 2007; Jaafar & McLeay, 2007). Most studies tend to reach the conclusions of negative asso- ciation of IFRS adoption with the cost of equity capital (Daske, 2006; Daske, Hail, Leuz, & Verdi, 2008; Lee, Walker, & Christensen, 2010). Market liquidity tends to increase, embodied in lower bid-ask spreads, higher trading volume and smaller cross-border liquidity differences (Leuz, 2003; Leuz & Verrecchia, 2000; Platikanova, 2007). More efficient shifts in capital allo - cation across firms (Zhang, 2013) and decreases in IPO underpricing (Hong, Hung, & Lobo, 2014) are also evidenced. Many studies explore the effect of IFRS adoption from the contractual perspective. Christensen, Lee, and Walker (2009), Chen, Chin, Wang, and Yao (2015) and Ball, Li, and Shivakumar (2015) show the evidence of IFRS adoption effect on debt contracts, while Wu and Zhang (2009), Ozkan, Singer, and You (2012), Voulgaris, Stathopoulos, and Walker (2014) concentrate on the evidence of the effect of IFRS adoption on compensation contracts. 2.2. Examination of convergence approach The CAS (2006), consisting of one Basic Standard, 38 Specific Standards and Implementation Guidance, has been recognised by the IASB as achieving substantial convergence with IFRS (Tweedie, 2006b). The implementation of CAS (2006) has stimulated studies on the conver- gence of CAS with IFRS. Some scholars demonstrate that the choice of convergence strategy is adapted to Chinese characteristics. Ding and Su (2008) provide a descriptive analysis of the process of China’s move toward IFRS, revealing that IFRS can work properly in a regulated market. Peng, Tondkar, Smith, and Harless (2008), Peng and Smith (2010) conclude that a combination of staged implementation and direct import is practical and effective in the convergence of CAS with IFRS, and that the convergence of accounting standards in China has led to the convergence of accounting practices. Qu and Zhang (2010), applying fuzzy clustering analysis, provide evidence of the overall substantial convergence for most items of the standards and a few subtle differences. Other studies, meanwhile, examine the influence of convergence on accounting infor - mation quality in China. The principles-orientated CAS (2006), incorporating fair value, has a twofold effect on accounting information quality. On the one hand, increased value rele - vance of accounting information after convergence has been demonstrated in many studies (Chen & Qu, 2014; Li & Park, 2011; Luo, Xue, & Zhang, 2008). On the other hand, CAS (2006) creates many new opportunities for earnings management (Cang et al., 2014; He et al., 2012; Wang, Xue, & Chen, 2009). With a systematic test of IFRS convergence, Chen and Qu (2014) argue that IFRS convergence overall enhances accounting information quality in China. Several recent studies focus on the direct economic consequences of IFRS convergence in China. Wang and Ye (2011) and Gao and Fu (2012) find evidence of a reduced cost of equity capital after the convergence. Hou, Jin, and Wang (2014) find a positive role of convergence in the accounting-based performance sensitivity of executive compensation. On the contrary, 4 M. CHEN ET AL. Du, Lei, and Zhu (2009) find out that only two of the 11 events related to the promulgation of CAS (2006) have significantly positive market reactions. Regarding foreign investment, DeFond, Gao, Li, and Xia (2014) and Sun (2011) have the opposite empirical results. Liu (2015) demonstrate increased efficiency of resource allocation. Hou, Jin, Wang, and Zhang (2016) find that mandatory adoption of IFRS has led to deteriorating real investment efficiency in China. 2.3. Motivation Many studies of IFRS adoption or convergence argue that institutional factors and manage- ment incentives are crucial to implementing accounting standards. Jeanjean and Stolowy (2008) suggest that the authorities should devote their eo ff rts to harmonising management incentives and institutional factors rather than harmonising accounting standards. Both He et al. (2012) and Cang et al. (2014) emphasise the importance of compatibility between institutional factors and accounting standards. There is, however, seldom direct empirical evidence of the influence of institutional factors and management incentives on IFRS convergence. The proponents of IFRS assert that IFRS will ‘reduce the cost of capital’ (Tweedie, 2006a). The prior studies on reduced cost of capital after convergence in China neither conduct heterogeneity analysis nor explore the driving factors. Therefore, this paper seeks to inves- tigate the impacts of institutional factors and management incentives on the relationship between IFRS convergence on the cost of equity capital. The investigation contributes to appraising the careful process of developing a substantive convergence strategy in China (World Bank, 2009). 3. Hypothesis development and empirical models 3.1. Basic analysis New institutional economics argues that the value of institutions exists in the reduction of transaction costs (Schultz, 1968). As an institutional arrangement, accounting standards are expected to reduce transaction costs by improving the quality of accounting information. Many studies provide evidence of the improved quality of accounting information after IFRS adoption (Barth et al., 2008; Daske & Gebhardt, 2006; DeFond et al., 2011; Hung & Subramanyam, 2007; Iatridis & Rouvolis, 2010). Prior researchers have provided evidence that IFRS convergence leads to higher value relevance in China (Chen & Qu, 2014; Li & Park, 2011; Luo et al., 2008). However, increased earnings management is also documented (Cang et al., 2014; He et al., 2012; Wang et al., 2009). To assess the accounting information quality after convergence with the comprehensive viewpoint, Chen and Qu (2014) establish an analytical system of accounting information quality characteristics based on the Conceptual Framework (IASB & FASB, 2010). On the one hand, CAS (2006) induces more discretionary accruals and less conservatism; on the other Based on the Conceptual f ramework (iaSB & faSB, 2010), the fundamental characteristics of useful financial information are relevance and faithful presentation; while improving characteristics are comparability, timeliness, variability and under- standability. t he Conceptual f ramework (iaSB & faSB, 2010) does not include conservatism as an aspect of faithful rep - resentation; instead, it stresses neutrality rather than conservatism. CHINA JOURNAL OF ACCOUNTING STUDIES 5 hand, simultaneously, CAS (2006) brings about improved value relevance, lowered earnings smoothness, and increased comparability and neutrality. Thus, it is concluded that IFRS convergence overall improves the quality of accounting information in China (Chen & Qu, 2014). Information quality affects a firm’s cost of capital (Easley & O’Hara, 2004). More extensive financial disclosures and higher quality reporting are inversely related to firms’ (implied) cost of equity capital (Botosan, 1997; Botosan & Plumlee, 2002; Francis, Olsson, & Schipper, 2004; Hail, 2002; Hail & Leuz, 2006). Lower cost of equity capital has been observed in Europe (Daske, 2006; Daske et al., 2008; Lee et al., 2010; Li, 2010) and in China (Gao & Fu, 2012; Wang & Ye, 2011). Tweedie (2006a) also expresses the viewpoint that IFRS convergence will ‘reduce the cost of capital’. Thus, the first hypothesis to be tested is as follows: H1: The cost of equity capital reduces after the IFRS convergence of accounting standards in China. The following basic model is for the basic test: R = + POST + CONTROLS + (1) e,it 0 1 it j it where R is the cost of equity capital of company i in t period, POST is a dummy variable e,it it indicating pre- or post-convergence period, and CONTROLS denotes the sets of control var- iables in empirical tests (set out in Table 1), which will be discussed later in this paper. Based on the research results of Ruland, Shon, and Zhou (2007), this paper controls for the differ - ences across firms such as firm size and industry type. 3.2. Heterogeneity analysis Next, this paper analyses heterogeneity in the reduction in the cost of equity capital across firms. It is expected that certain listed companies could benefit more from the IFRS conver - gence than the others. Some listed companies in China are cross-border financing, not only issuing A-shares, but also B-shares, H-shares or N-shares. Cross-border financing companies were required to prepare financial reports in accordance with IFRS even before the IFRS convergence. IFRS convergence might not significantly influence cross-border financing companies. Compared with cross-border financing companies, those companies issuing only A-shares to the domes - tic market were required to provide more incremental information for investors after con- vergence. It is expected that there is significant difference in the reduction in the cost of equity capital between cross-border financing companies and domestic financing compa- nies. Therefore, this paper develops hypothesis H2a as follows: H2a: In terms of the reduction in the cost of equity capital, domestic financing companies benefit more from IFRS convergence than cross-border financing companies. all firms, listed in the Shanghai Stock exchange (ShSe) and Shenzhen Stock exchange (SZSe) in China, issue a-shares denom - inated in Chinese Yuan that are predominantly traded by domestic investors. Some listed firms are permitted to issue B-shares denominated in uS dollars in ShSe and in hK dollars in SZSe that are predominantly traded by international investors. Some firms are also cross-listed in hong Kong exchanges (hKeX) issuing h-shares denominated in hK dollars, or in new York Stock exchange (n YSe) issuing n-shares denominated in uS dollars. 6 M. CHEN ET AL. Table 1. Summary of variables. Variables Description R t he implied cost of equity capital of company i on 30 april period t e,it FEPS & FEPS t he average of the forecasted ePSs for the periods t and t+1 by analysts on 30 april of that period it it+1 P t he closing price on 30 april of period t it g t he long-term (asymptotic) growth rate in expected earnings, calculated as the difference between lt the average annual gni and average annual CPi in China during 1985 to 2009, ∑ ∑ 2009 2009 GNI CPI yr=1985 yr yr=1985 yr g = − lt 25 25 g t he short-term growth rate in expected earnings of company i at the expectation date, st,it FEPS −FEPS it+1 it g ≡ st,it FEPS it POST a dummy variable, if a firm-year observation falls in 2007 or later, Post=1, and 0 otherwise it TYPE a proxy of the particular category of company i, standing for OnlyA , STATE% , STATE_Dum or it–1 it–1 it–1 it–1 BIG4 in the heterogeneity analysis respectively it–1 OnlyA a dummy variable, if a firm-year observation is non-cross-border financing, only issuing a-share, it–1 OnlyA = 1, and 0 otherwise it–1 STATE% t he percentage of state-held shares of company i at year end of period t–1 it–1 STATE_Dum a dummy variable, if company i is state-owned at year end of period t–1, State_Dum = 1, and 0 it–1 it–1 otherwise BIG4 a dummy variable, if company i is audited by a Big 4 accounting firm for period t–1, BIG4 = 1, and it–1 it–1 0 otherwise MI t he yearly marketization index of the province where company i is headquartered for period t–1, it–1 from f an et al. (2011) MI_Dum a dummy variable, if MI is higher than the average MI of the same period, MI_Dum = 1, and 0 it–1 it–1 it–1 otherwise DA t he absolute value of discretionary accruals of company i scaled by the total assets for period t–1, it–1 measured using the amended Jones model (amended by matching industries) DA_Dum a dummy variable, if DA is higher than the average DA of the same period, DA_Dum = 1, and 0 it–1 it–1 it–1 otherwise MRV t he monthly standard deviation of daily market returns in april of period t it BM t he book-to-market ratio of equity for company i at year end of period t–1 it–1 LEV t he financial leverage of company i, calculated as total debts divided by total assets at the year-end it–1 of period t–1 EPS t he earnings per share of company i for period t–1 it–1 SIZE t he natural logarithm of total market value in thousands of rMB at year end of period t–1 it–1 EDR a dummy variable, if company i has completed e quity division reform on april of period t, EDR = 1, it it 0 otherwise IND d ummy variables, based on the 13 industries classified by the CSrC (China Securities regulatory it Commission) notes: t he sources of the data: t he data of gni and CPi are collected from the website of national Bureau of Statistics. t he data of Mi are cited from f an et al. (2011). t he other data are downloaded via batch mode from the financial databases of Wind and CSM ar. It is argued that equity capital structure influences the preparers’ incentives (Ball, Robin, & Wu, 2003; Soderstrom & Sun, 2007). A special feature of the Chinese capital market is that the majority of listed companies are state-owned due to the particular process of historical development. Compared with state-owned companies, the management of non-state- owned companies is faced with higher operating pressure and is more sensitive to account- ing-based compensation incentives. Therefore, they tend to have a stronger motivation of earnings management. Bo and Wu (2009) find that the level of positive earnings manage - ment in state-owned companies is significantly lower than non-state-owned companies. Yu (2009) shows that state-owned companies benefit more from IFRS convergence, in terms of value relevance. With lower earnings management and higher value relevance, it is antici- pated that the earnings quality of state-owned companies is higher. Thus it is expected that the state-owned companies should enjoy a lower cost of equity capital. Consequently, this paper hypothesises H2b as follows: CHINA JOURNAL OF ACCOUNTING STUDIES 7 H2b: In terms of the reduction in the cost of equity capital, state-owned companies benefit more from IFRS convergence than non-state-owned companies. The Big 4 accounting firms are more familiar with IFRS than non-Big 4 ones. For example, they have tutorials on line, and Deloitte and Touche owns the website of IAS-plus focusing on the IFRS issues. Therefore, it is expected that the audit quality of the Big 4 for IFRS-based financial statements is higher than that of non-Big 4. Moreover, western literatures almost have the consistent conclusion that companies audited by the Big 4 accounting firms are expected to have higher earnings quality (Berker, DeFond, & Subramanyam, 1998; DeAnglo, 1981; DeFond & Jiambalvo, 1993). Xin and Wang (2010) argue that choice of the Big 4 could be the reason for higher earnings quality of cross-listing companies than that of domestic listing companies in China. Accordingly, hypothesis H2c is: H2c: In terms of the reduction in the cost of equity capital, Big 4 audited companies benefit more from IFRS convergence than non-Big 4 audited companies. The following model is for the heterogeneity tests: R = + POST + TYPE + POST ∗ TYPE + CONTROLS + (2) e,it 0 1 it 2 it−1 3 it it−1 j it where R , POST and CONTROLS are defined as previously, and TYPE stands for proxies of e,it it it–1 the particular categories of companies. To test H2a, OnlyA substitutes for TYPE , indicating it–1 it–1 whether company i is cross-border financing or not during the t–1 period. To examine H2b, STATE% substitutes for TYPE of ‘β TYPE ’, and STATE_Dum substitutes for TYPE of it–1 it–1 2 it–1 it–1 it–1 the interaction term, respectively, indicating the percentage of state-held shares of company i and whether company i is state-owned or not during the period t–1. To test H2c, BIG4 it–1 substitutes for TYPE , indicating whether company i is audited by a Big 4 accounting firm it–1 or not during period t–1. In order to allow for a two-by-two difference-in-difference analysis of IFRS convergence effects on the cost of equity capital, the dummy indicator such as OnlyA , STATE_Dum or BIG4 is used in the interaction term. Except for the interaction it–1 it–1 it–1 term, this paper uses continuous variables if possible, since continuous variables contain more information than dummy variables; for example, using STATE% instead of it–1 STATE_Dum . All variables are defined in Table 1. it–1 3.3. Driving factor analysis Furthermore, this paper attempts to explore the drivers of heterogeneity from the perspec- tives of institutional factors and management incentives. Prior studies suggest that compa- nies in weak institutional settings are more likely to abuse the discretion of accounting judgement to manipulate earnings (Burgstahler, Hail & Leuz, 2006). To some extent, the quality of financial information prepared under IFRS depends on the institutional factors and market forces of a country (Shelton, Jackson, & Robinson, 2011). Therefore, the hypoth- esis to be tested is as follows: H3a: Institutional factors are significant for the beneficiary companies to benefit from the IFRS convergence, and the higher the degree of marketisation, the greater the reduction in the cost of equity capital. The following model is specified for the empirical test of institutional factors: 8 M. CHEN ET AL. R = + POST + TYPE + MI e,it 0 1 it 2 it−1 3 it−1 (3) + POST ∗ TYPE + POST ∗ TYPE ∗ MI_Dum 4 it it−1 5 it it−1 it−1 + CONTROLS + j it where R , POST , CONTROLS and TYPE are defined as previously, MI is the yearly mar- e,it it it–1 it–1 ketisation index of the province where company i is headquartered during period t–1, and MI_Dum is a dummy variable indicating whether the institutional settings around com- it–1 pany i are strong or weak during period t–1. All variables are defined in Table 1. This paper measures the institutional settings using the average score of the various dimensions of the degree of marketisation in die ff rent provinces of China (Fan, Wang, & Zhu, 2011). As mentioned above, this paper uses continuous variables MI instead of MI_Dum it–1 it–1 outside of the interaction term. Apart from institutional factors, management incentives dominate accounting standards in determining accounting quality (Christensen & Lee, 2008). Ball et al. (2003) find evidence that even in the relatively strong institutional settings, such as common law countries, incen- tives of preparers affect the financial reporting quality. Better accounting standards are helpful only for companies with proper reporting incentives (Wang & Yu, 2009). Therefore, another hypothesis to be tested is as follows: H3b: Management incentives are significant for the beneficiary companies to benefit from the IFRS convergence, and the lower the earnings management, the greater reduction in the cost of equity capital. The following model is for the empirical test of management incentives: R = + POST + TYPE + DA e,it 0 1 it 2 it−1 3 it−1 + POST ∗ TYPE + POST ∗ TYPE ∗ DA_Dum (4) 4 it it−1 5 it it−1 it−1 + CONTROLS + j it where R , POST , CONTROLS and TYPE are defined as previously, DA stands for discre- e,it it it–1 it–1 tionary accruals of company i for period t–1, and DA_Dum is a dummy variable indicating it–1 whether the magnitude of discretionary accruals of company i is high or low during period t–1. Discretionary accruals are the manipulated un-reported earnings through the abuse of accounting judgement discretion. This paper measures the management incentives by using the discretionary accruals. As mentioned above, this paper uses continuous variables DA it–1 instead of DA_Dum outside of the interaction term. All variables are defined in Table 1. it–1 4. Research design 4.1. Implied cost of equity capital Due to the availability of forecasted EPS data, this paper uses the PEG (price/earnings to growth) model (Easton, 2004) and the amended OJ (Ohlson-Juettner) model (Ohlson, 1995) a fter systematic analysis and evaluation, the book of China’s Marketisation Index (f an et al., 2011) reports the marketisation indexes of all provinces in China from year 1997 to 2009, which show relative differences among provinces in the process of marketisation. Chinese scholars widely use the data to measure the institutional settings. f orecasted ePS data by financial analysts have been available from the Wind financial data centre in China since 2004. h owever, the available data only cover the next two to three years for listed companies. a gl S (g ebhardt, l ee, & Swaminathan, 2001) model is not used because it needs more than 12 periods of forecasted ePS data. CHINA JOURNAL OF ACCOUNTING STUDIES 9 to compute the implied cost of equity capital. Under the PEG model, the implied cost of capital is computed as follows: FEPS − FEPS it+1 it R = (5) e,it it where R is the implied cost of equity capital of company i on 30 April of period t, FEPS and e,it it FEPS are the averages of the forecasted EPSs of company i for the periods t and t+1 by ana- it+1 lysts on 30 April of period t, and P is the closing price of company i on 30 April of period t. it The OJ valuation model is shown in Equation (6), and the implied cost of capital model is displayed in Equation (8): g − g FEPS it st ,it lt P = (6) it R R − g e,it e,it lt FEPS − FEPS + R ∗ FD it+1 it e,it it g ≡ (7) st,it FEPS it FEPS − FEPS FEPS it+1 it 2 it R = A + A + − g (8) e,it lt P FEPS it it g − FD ∕P lt it it A ≡ (9) where g is the long-term (asymptotic) growth rate in expected earnings, g is the short- lt st,it term growth rate in expected earnings of company i at the expectation date, and FD is the it forecasted dividend per share for period t. Ohlson and Gao (2006) suggest choosing the long-term growth rate in GNP (Gross National Product) as g . In this paper, to eliminate the lt influence of inflation, the difference between the average annual GNI (Gross National Income) and average annual CPI (Consumer Price Index) in China during the period from 1985 to 2009 is used to approximate g . The g and g are both defined in Table 1. lt lt st,it Because of the unavailability of forecasted dividend per share in China, based on the prior studies (Gode, 2003; Shen, 2008), this paper approximates g by ignoring FD . The OJ model st it then is amended as follows: 4FEPS it g + g + g − g lt lt st,it lt it (10) R = e,it 4.2. Control variables To test the influence of IFRS convergence on the cost of equity capital, based on the risk- and-return theory, both systematic risk and non-systematic risks are taken into consideration. national Bureau of Statistics of China started to release gni from 1985. 10 M. CHEN ET AL. Systematic risk is triggered by macro-economic and institutional factors. The Chinese stock market went through drastic fluctuations around the 2008 financial crisis. The more drastic the market fluctuation, the higher the systematic risk, pushing up the cost of equity capital. To control this unavoidable systematic market risk, this paper includes the monthly standard deviation of daily market stock returns during April, MRV (Daske, 2006; Daske et al., 2008; Li, 2010). Non-systematic risks are induced by firm specific characteristics and events. To control the firm-specific characteristics, this paper chooses BM , LEV EPS , SIZE and IND as it–1 it–1, it–1 it–1 it the control variables (Daske, 2006; Daske et al., 2008; Li, 2010). Table 1 summarises the var- iables adopted in this paper. 4.3. Sample selection and descriptive statistics The institutional background of the Chinese stock market has to be taken into account for the selection of samples. A unique feature distinguishing the Chinese stock market is the dichotomy of stock circulation, which refers to the phenomena of equity division that nearly two-thirds of outstanding shares in China were not tradable (Green, 2003). In April, 2005, the Equity Division Reform (EDR) was launched to convert all non-tradable shares into tradable shares, a significant institutional reform with the goal of achieving equality of rights and interests. By the end of 2006, the process had been essentially completed, with over 90% of the affected companies completing the reform. By the end of 2007, 1254 firms had completed the reform, representing over 97% of the market capitalisation at the time. Owing to the overlap period between the Equity Division Reform and CAS (2006) imple- mentation, the samples used in this paper have to be delicately selected to eliminate noise. This paper determines the status of every firm–year observation, judging whether the obser - vation completed Equity Division Reform or not. The EDR completed samples are selected for major empirical tests, and the pooled samples are used in robustness tests. All the data are obtained from the financial databases of WIND and CSMAR, the website of National Bureau of Statistics, and also are manually collected from Fan et al. (2011). Forecasted EPS data by financial analysts have been available since 2004. The latest MI data from Fan et al. (2011) is data of 2009. Therefore, the sample periods cover the years of 2004 to 2010. Table 2 reports the distribution of observations for both the EDR completed samples and the pooled samples in Panel A, and the descriptive statistics of continuous variables for the EDR completed samples in Panel B. It is obvious that the yearly number of companies fol- lowed by analysts is increasing, only 26 in 2004, 323 in 2005, and 407 in 2006, and so on. In 2005, among the 323 observations, only four companies completed the Equity Dividend Reform. In 2006, 244 out of 407 companies completed the Equity Dividend Reform. Table 3 presents the Pearson and Spearman correlation coefficients matrix. There is a significant negative correlation between the cost of equity capital and the percentage of state-held share, and a significant positive correlation between the cost of equity capital and the discretionary accruals. There is no multicollinearity among the continuous variables. t he major non-tradable shares are held by the state shareholders or the biggest shareholders, who lack interest in whether market price reflects intrinsic value, because of the non-tradable status of the shares held by these investors. t he devel- opment of the Chinese financial markets and the protection of the interests of small investors had been seriously constrained by the equity division. CHINA JOURNAL OF ACCOUNTING STUDIES 11 Table 2. distribution of observations and descriptive statistics for continuous variables. Panel A: Distribution of observations 2004 2005 2006 2007 2008 2009 2010 Sum edr-completed 0 4 244 514 563 650 824 2,799 Pooled 26 323 407 529 565 650 824 3,324 Panel B: Descriptive statistics of continuous variables Variable N Mean Std. d ev. Min P25 Median P75 Max _oj 2799 0.12 0.04 0.02 0.09 0.11 0.13 0.39 R _peg 2799 0.10 0.04 0.01 0.08 0.10 0.12 0.38 BM 2799 0.44 0.33 –0.82 0.22 0.35 0.59 4.41 LEV 2799 0.53 1.05 0.01 0.38 0.51 0.64 55.41 EPS 2799 0.40 0.49 –2.86 0.15 0.31 0.54 6.28 SIZE 2799 15.40 1.09 12.92 14.61 15.30 16.05 21.22 STATE% 2799 0.23 0.24 0.00 0.00 0.20 0.44 0.85 MI 2799 7.60 1.68 0.38 6.36 7.38 8.77 11.80 DA 2799 0.06 0.07 0.00 0.02 0.05 0.08 0.46 5. Empirical results and analysis 5.1. Basic analysis results Table 4 reports the regression results for basic analysis of IFRS convergence influence on the cost of equity capital. In both the OJ model and PEG model, POST presents a consistent significantly negative association with the cost of equity capital at the level of 1%. It reveals that the IFRS convergence of CAS (2006) facilitates the reduction of the cost of equity capital, which supports hypothesis H1. The coefficients of control variables are in line with predictions. MRV presents significant positive association with the cost of equity capital at the level of 1%, showing the influence of systematic risk. LEV shows significant positive association with the cost of equity capital, indicating higher financial risk also pushing up the cost of equity capital. The significant negative association between EPS and the cost of equity capital signifies that the companies with better operating performance benefit. 5.2. Heterogeneity analysis results Table 5 reports the regression results for the heterogeneity analysis of IFRS convergence influence on the cost of equity capital. In all cases of testing the heterogeneity of cross-border financing versus domestic financing, state-owned versus non-state-owned or Big 4-audited versus non-Big 4-audited, and under both the OJ model and the PEG model, POST has a consistently significant negative association with the cost of equity capital. The significance is mostly at the 1% level, except for testing the heterogeneity of cross-border financing versus domestic financing under the OJ model. This evidence supports hypothesis H1. The first two columns of Table 5 are the results of testing the heterogeneity of cross-border financing versus domestic financing. The sign of explaining variables and control variables are consistent with expectations. OnlyA presents a significant positive association with the cost of equity capital and both of OJ model and PEG model are at the level of 1%, indicating 12 M. CHEN ET AL. Table 3. Correlation analysis. Re_oj Re_peg MRV BM LEV EPS SIZE STATE% MI DA *** ** * *** *** ** *** *** Re_oj 0.991 0.047 0.033 0.097 –0.050 0.023 –0.061 –0.002 0.149 *** ** *** *** ** *** *** Re_peg 0.991 0.042 0.055 0.096 –0.005 0.047 –0.054 –0.002 0.154 *** *** *** *** *** MRV 0.009 –0.011 –0.294 –0.016 0.133 0.251 0.135 –0.017 0.051 *** *** *** *** *** *** BM –0.027 –0.001 –0.186 –0.031 –0.171 –0.335 0.091 –0.084 –0.104 *** *** * ** LEV 0.234 0.234 –0.012 0.023 –0.030 –0.019 –0.018 0.013 0.044 *** *** *** *** *** ** *** EPS 0.028 0.083 0.115 –0.216 –0.106 0.404 0.029 0.048 0.111 *** *** *** *** *** *** * *** SIZE 0.059 0.082 0.125 –0.413 –0.012 0.460 0.131 0.035 0.107 *** *** *** *** *** *** *** STATE% –0.074 –0.070 0.191 0.118 0.020 0.020 0.088 –0.132 –0.033 *** *** *** *** *** * MI –0.012 –0.011 0.017 –0.055 0.061 0.054 0.052 –0.135 0.025 *** *** *** *** *** *** *** DA 0.108 0.112 0.013 –0.158 0.113 0.089 0.078 –0.054 0.009 notes: t he upper right part of the matrix is Pearson correlation coefficients, and the lower left part of the matrix is Spearman correlation coefficients. *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. CHINA JOURNAL OF ACCOUNTING STUDIES 13 Table 4. Basic analysis of ifrS convergence on the cost of equity capital. OJ model PEG model Variables Coeff. t-value Coeff. t-value *** *** intercept 0.080 6.43 0.058 4.76 *** *** POST − –0.016 –5.88 –0.022 –8.34 it *** *** MRV + 0.295 4.26 0.302 4.45 it ** BM + 0.003 1.35 0.006 2.41 it–1 *** *** LEV + 0.003 4.15 0.003 4.29 it–1 *** ** EPS − –0.007 –4.25 –0.003 –2.18 it–1 *** *** SIZE + 0.003 3.66 0.003 4.64 it–1 f ixed effects industry industry a dj. R 0.10 0.11 F value 18.53 20.99 num. 2,799 2,799 notes: *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. Table 5. heterogeneity analysis. Cross-border financing State-owned Big4 audited Variables OJ model PEG model OJ model PEG model OJ model PEG model *** *** *** *** ** intercept 0.043 0.020 0.070 0.048 0.070 0.049 (2.92) (1.38) (5.64) (3.94) (5.30) (3.74) * *** *** *** *** *** POST − –0.011 –0.017 –0.010 –0.016 –0.018 –0.024 it (–1.70) (–2.67) (–3.33) (–5.37) (–6.22) (–8.46) *** *** OnlyA + 0.018 0.019 it–1 (2.75) (2.88) POST * − –0.005 –0.005 it OnlyA (–0.76) (–0.79) it–1 STATE% − –0.002 –0.003 it–1 (–0.65) (–0.79) *** *** POST * − –0.008 –0.008 it STATE_Dum (–4.01) (–4.23) it–1 ** ** BIG4 − –0.015 –0.014 it–1 (–2.45) (–2.35) POST * − 0.010 0.009 it BIG4 (1.56) (1.46) it–1 *** *** *** *** *** *** MRV + 0.306 0.314 0.309 0.318 0.296 0.303 it (4.45) (4.66) (4.41) (4.64) (4.29) (4.48) *** *** ** *** * *** BM + 0.007 0.010 0.005 0.007 0.004 0.007 it–1 (2.87) (3.98) (1.97) (3.09) (1.81) (2.85) *** *** *** *** *** *** LEV + 0.003 0.003 0.003 0.003 0.003 0.003 it–1 (4.23) (4.38) (4.14) (4.29) (4.17) (4.32) *** ** *** ** *** ** EPS − –0.007 –0.004 –0.007 –0.004 –0.007 –0.004 it–1 (–4.38) (–2.30) (–4.59) (–2.54) (–4.35) (–2.27) *** *** *** *** *** *** SIZE + 0.004 0.005 0.003 0.004 0.004 0.004 it–1 (5.04) (6.08) (4.35) (5.39) (4.30) (5.21) f ixed effects industry industry industry industry industry industry a dj. R 0.11 0.12 0.10 0.12 0.10 0.14 F value 18.44 20.86 18.07 20.52 17.10 19.28 num. 2,799 2,799 2,799 2,799 2,799 2,799 notes: *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. that the cost of equity capital of the domestic financing companies is higher than that of cross-border financing companies. However, the coefficient of the intersection term of POST*OnlyA is negative, but not statistically significant, which signifies that the magnitude of the reduction in the cost of equity capital for domestic financing companies is not really higher than that for cross-border financing companies. The hypothesis H2a is not supported. 14 M. CHEN ET AL. Cross-border financing companies are required to prepare financial reports in accordance with IFRS for overseas markets prior to IFRS convergence. It is also the first time for them to prepare financial reports in accordance with IFRS-converged CAS (2006) for domestic mar - kets. Compared with no incremental information for overseas investors, there might be incremental information for domestic investors. This could be the explanation of the unsup- ported result for testing the heterogeneity of cross-border versus domestic financing. The last two columns of Table 5 are the results of testing the heterogeneity of Big 4-audited versus non-Big 4-audited. The signs of control variables are consistent with expectations. In terms of explaining variables, BIG4 shows significant negative association with the cost of equity capital at the level of 5%, in line with the prediction, indicating that the cost of equity capital of the Big 4 audited companies is lower than that of non-Big 4-audited companies. However, the coefficient of the intersection term of POST*BIG4 is positive, opposite to the prediction. Hypothesis H2c is not supported. In contrast with western literature, many Chinese studies argue that the Big 4 accounting firms do not provide better audit quality in China (Guo, 2011; Liu & Zhou, 2007). This could be the reason to explain the unsupported result for testing the heterogeneity of Big 4-audited versus non-Big 4-audited. The two columns in the middle of Table 5 are the results of testing the heterogeneity of state-owned versus non-state-owned. The signs of the explanatory variables and the control variables are consistent with expectations. The coefficient of state% is negative, but not statistically significant. The intersection term of POST*STATE_Dum is a significantly negative association with the cost of equity capital at the level of 1%, which indicates that the differ - ence in the reduction of the cost of equity capital between the state-owned companies and non-state-owned companies is significant. Therefore, hypothesis H2b is supported. Furthermore, this paper conducts the two-by-two difference-in-difference analysis of IFRS convergence influences for state-owned versus non-state-owned companies across the pre- versus post-convergence periods. Table 6 reports the average costs of equity capital, the numbers of observations in the subsamples, the coefficients of subsamples regressions and the significance levels. Table 6. t wo-by-two analysis of state-owned versus non-state-owned companies. Panel A: OJ model Variables Pre-convergence period Post-convergence period difference ( olS) *** State-owned R = 0.125 n = 200 R = 0.112 n = 1,828 Coeff. = –0.018 e e companies t = –6.05 *** non-state-owned R = 0.128 n = 48 R = 0.121 n = 723 Coeff. = –0.015 e e companies t = –2.63 *** *** difference ( olS) Coeff. = –0.002 Coeff. = –0.009 Coeff. = –0.008 t = –0.28 t = –5.26 t = –4.01 Panel B: PEG model Pre-convergence period Post-convergence period difference ( olS) *** State-owned R = 0.118 n = 200 R = 0.098 n = 1,828 Coeff. = –0.024 e e companies t = –8.37 *** non-state-owned R = 0.118 n = 48 R = 0.107 n = 723 Coeff. = –0.020 e e companies t = –3.62 *** *** difference ( olS) Coeff. = –0.002 Coeff. = –0.009 Coeff. = –0.008 t = –0.21 t = –5.54 t = –4.23 notes: t he regression model for state-owned subsample and non-state-owned companies is as follows: R = β + β POS- e,it 0 1 T + Σβ CONTROLS + ε . t he regression model of pre-convergence subsample and pre-convergence subsample is as it j it follows: R = β + β STATE_DUM + Σβ CONTROLS + ε . *, ** and *** denote significance at the 10%, 5%, and 1% levels, e,i t 0 1 it–1 j it respectively, all two-tailed. CHINA JOURNAL OF ACCOUNTING STUDIES 15 According to Table 6, among the limited 248 samples before convergence, non-state- owned companies total only 48, with two reasons probably accounting for this: (1) in the first few years of forecasted EPS disclosure, analysts preferred to follow the large-scale state- owned companies; (2) most of the pilot EDR companies were state-owned. Under the OJ model, in the pre-convergence period, the average equity capital cost of 200 state-owned companies is 0.125, and is 0.128 for the 48 non-state-owned companies; in the post-convergence period, the average equity capital cost of 1828 state-owned com- panies is 0.112, and is 0.121 for the 723 non-state-owned companies. In the cell of the dif- ference between pre- and post-convergence periods for state-owned companies, –0.018 is the coefficient of POST for state-owned subsample regression, significant at 1%, which means that the reduction in the cost of equity capital for state-owned companies after convergence is statistically significant. In the cell of the difference between state-owned and non-state-owned companies before convergence, –0.002 is the coefficient of STATE_Dum for the pre-convergence subsample regression, without statistical significance. In the bot - tom-right cell of Panel A (the difference-in-difference cell), the significant coefficient of –0.008 is exactly the coefficient of POST*STATE_Dum in Table 5. According to Table 6, both state-owned and non-state-owned companies experience a significant reduction in the cost of equity capital after IFRS convergence. The magnitude of state-owned companies’ reduction (from 0.125 down to 0.112, t = –6.05 under OJ model, and from 0.118 down to 0.098, t = –8.37 under PEG model) is much bigger than that of non- state-owned companies’ reduction (from 0.128 down to 0.121, t = –2.63 under OJ model, and from 0.118 down to 0.107, t = –3.62 under PEG model). That is why the coefficient of interaction term, POST*STATE_Dum, is significantly negative in Table 5. In the pre-convergence period, there is no significant difference of the cost of equity capital between state-owned and non-state-owned companies, while, in the post-conver- gence period, there is a significant difference of the cost of equity capital between them. In summary, the results of Tables 5 and 6 suggest that state-owned companies benefit more than non-state-owned companies. Hypothesis H2b is supported. This paper now goes a step further to explore the driving factors for the heterogeneity of state-owned versus nonsstate-owned companies. 5.3. Driving factor analysis results 5.3.1. Influence of institutional factors Table 7 presents the results of the regression analyses to test the influence of institutional factors. OnlyA and BIG4 are added in as control variables since Table 5 shows that OnlyA and BIG4 have significant associations with the cost of equity capital. Except for BIG4, the other control variables show predicted significant associations with the cost of equity capital. Based on Table 7, in terms of the explaining variables, POST presents significant negative association with the cost of equity capital at the level of 1%, being consistent with the above basic analysis and the heterogeneity analysis. The interaction term of POST*STATE_Dum*MI_Dum t he regression model of the state-owned subsample is as follows: R = β + β POST + β MRV + β BM + β LEV + β EPS + β SIZE + ΣIND + ε e,it 0 1 it 2 it 3 it–1 3 it–1 4 it–1 5 it–1 it it t he regression model of the pre-convergence subsample is as follows: R = β + β STATE_DUM + β MRV + β BM + β LEV + β EPS + β SIZE + ΣIND + ε e,it 0 1 it–1 2 it 3 it–1 3 it–1 4 it–1 5 it–1 it it 16 M. CHEN ET AL. Table 7. institutional factors influence analysis. OJ model PEG model Variables Coeff. t-value Coeff. t-value ** Intercept + 0.032 2.28 0.008 0.60 *** *** POST − –0.012 –3.96 –0.018 –6.08 it STATE% − –0.002 –0.61 –0.003 –0.73 it–1 ** *** MI − 0.001 2.38 0.002 2.90 it–1 POST *STATE_Dum − –0.002 –0.78 –0.002 –0.71 it it–1 *** *** POST *STATE_Dum *MI_Dum − –0.008 –3.75 –0.009 –4.12 it it–1 it–1 *** *** MRV + 0.324 4.65 0.335 4.92 it *** *** BM + 0.009 3.44 0.011 4.58 it–1 *** *** LEV + 0.003 4.22 0.003 4.37 it–1 *** *** EPS − –0.008 –4.96 –0.005 –2.94 it–1 *** *** SIZE + 0.004 5.43 0.005 6.41 it–1 *** *** OnlyA + 0.012 4.57 0.012 4.85 it–1 BIG4 − –0.001 –0.41 –0.001 –0.23 it–1 f ixed effects industry industry a dj. R 0.12 0.13 F value 17.03 19.38 N 2,799 2,799 notes: *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. presents a significantly negative association with the cost of equity capital (−0.008, t = −3.75, under OJ model, and −0.009, t = −4.12, under PEG model), in line with the expectation, sup- porting hypothesis H3a. Table 8 presents the results of two-by-two difference-in-difference analysis in settings of high marketisation and in settings of low marketisation respectively. Those observations with MI_Dum=1 are attributed to the high marketisation subsample, and the others to the low marketisation subsample. For illustration, in the pre-convergence period, among 48 non-state-owned companies as presented in Table 6, 33 companies are in the subsample of high-marketisation settings, while 15 companies are in the subsample of low-marketisation settings. Based on Panel A of Table 8, in the cells of differences between pre- and post-convergence, the coefficients of POST for both state-owned subsample regressions (−0.017, t = −4.59 under the OJ model, and −0.023, t = −6.47 under the PEG model) and non-state-owned subsample regressions (−0.019, t = −2.90 under the OJ model, and −0.024, t = −3.72 under the PEG model) indicate that, in high-marketisation settings, both state-owned companies and non- state-owned companies experience a reduction in the cost of equity capital after IFRS con- vergence. In the bottom-right cells, both under OJ model and PEG model, −0.008 and −0.008 are the coefficients of POST*STATE_Dum for subsample of high-marketisation settings, with a significance level at 1%, which means state-owned companies enjoy more cost reduction of equity capital than non-state-owned companies, supporting hypothesis H2b. Based on Panel B of Table 8, state-owned companies also enjoy a greater reduction in the cost of equity capital than non-state-owned companies, judging from the significant coef- ficients of POST*STATE_Dum for subsample regressions of low-marketisation settings (−0.008, t = −2.52 under the OJ model, and −0.008, t = −2.51 under PEG model). However, in low-marketisation settings, a reduction in the cost of equity capital is not found for the non-state-owned companies, judging from the insignificant coefficients of POST for non- state-owned subsample regressions (−0.008, t = −0.66 under the OJ model, and −0.013, t = −1.17 under the PEG model). Therefore, it is indicated that the institutional environment CHINA JOURNAL OF ACCOUNTING STUDIES 17 Table 8. t wo-by-two analysis of institutional factors influence. Panel A: High marketization settings o J model Peg model Pre-convergence period Post-convergence period difference ( olS) Pre-convergence period Post-convergence period difference ( olS) *** *** State-owned R = 0.121 n = 129 R = 0.109 n = 1,115 Coeff. = –0.017 R = 0.113 n = 129 R = 0.095 n = 1,115 Coeff. = –0.023 e e e e companies t = –4.59 t = –6.47 *** *** non-state-owned R = 0.127 n = 33 R = 0.119 n = 468 Coeff. = –0.019 R = 0.118 n = 33 R = 0.105 n =468 Coeff. = –0.024 e e e e companies t = –2.90 t = –3.72 *** *** *** *** difference ( olS) Coeff. = –0.008 Coeff. = –0.008 Coeff. = –0.008 Coeff. = –0.008 Coeff. = –0.009 Coeff. = –0.008 t = –0.86 t = –4.10 t = –3.13 t = –0.95 t = –4.45 t = –3.42 Panel B: Low marketisation settings o J model Peg model Pre–convergence period Post–convergence period difference ( olS) Pre–convergence period Post–convergence period difference ( olS) *** *** State-owned R = 0.134 n = 71 R = 0.115 n = 713 Coeff. = –0.022 R = 0.126 n = 71 R = 0.102 n = 713 Coeff. = –0.028 e e e e companies t = –4.51 t = –5.95 non-state-owned R = 0.129 n = 15 R = 0.126 n = 255 Coeff. = –0.008 R = 0.119 n = 15 R = 0.111 n = 255 Coeff. = –0.013 e e e e companies t = –0.66 t = –1.17 *** ** *** ** difference ( olS) Coeff. = 0.006 Coeff. = –0.008 Coeff. = –0.008 Coeff. = 0.008 Coeff. = –0.008 Coeff. = –0.008 t = 0.38 t = –3.13 t = –2.52 t = 0.51 t = –3.12 t = –2.51 notes: t he regression model for state-owned subsample and non-state-owned companies is as follows: R = β + β POST + Σβ CONTROLS + ε . t he regression model of pre-convergence subsample e,it 0 1 it j it and pre-convergence subsample is as follows: R = β + β STATE_DUM + Σβ CONTROLS + ε . *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. e,it 0 1 it–1 j it 18 M. CHEN ET AL. Table 9. t wo-by-two analysis of discretionary accruals. Pre-convergence period Post-convergence period Difference (T-test) *** State-owned DA = 0.052 n = 200 DA = 0.067 n = 1,828 t = 2.92 companies ** non-state-owned DA = 0.052 n = 48 DA = 0.075 n = 723 t = 2.52 companies ** difference (T-test) t = 0.03 t = −2.19 notes: *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. Table 10. Management incentives influence analysis. OJ model PEG model Variables Coeff. t-value Coeff. t-value *** intercept + 0.037 2.70 0.015 1.08 *** *** POST − –0.012 –3.80 –0.017 –5.87 it STATE% − –0.003 –0.69 –0.003 –0.84 it–1 *** *** DA + 0.070 5.19 0.071 5.37 it–1 *** *** POST *STATE_Dum − –0.006 –2.62 –0.006 –2.79 it it–1 * ** POST *STATE_Dum *DA_Dum − –0.004 –1.90 –0.004 –1.99 it it–1 it–1 *** *** MRV + 0.306 4.40 0.315 4.63 it *** *** BM + 0.009 3.54 0.011 4.68 it–1 *** *** LEV + 0.003 4.11 0.003 4.26 it–1 *** *** EPS − –0.008 –5.03 –0.005 –2.97 it–1 *** *** SIZE + 0.004 5.33 0.005 6.32 it–1 *** *** OnlyA + 0.012 4.58 0.012 4.88 it–1 BIG4 − –0.002 –0.62 –0.001 –0.47 it–1 f ixed effects industry industry a dj. R 0.12 0.14 F value 17.75 20.06 N 2,799 2,799 notes: *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. is the driving factor of the heterogeneity of state-owned versus non-state-owned, supporting hypothesis H3a. 5.3.2. Influence of management incentives The premise of hypothesis H3b is that the incentives of the management of state-owned companies to abuse discretion are lower than non-state-owned companies. Thus, before running multivariable regressions, this paper conducts a two-by-two analysis of discretionary accruals. Table 9 reports the average discretionary accruals, the numbers of observations in the subsamples, the t-test values and the significance levels. Discretionary accruals of both state-owned and non-state-owned companies increase significantly after IFRS convergence in Table 9 , consistent with the findings of He et al. (2012) and Hou et al. (2016). This is one of the twofold effects of principle-orientated, fair-value- based CAS (2006). In the pre-convergence period, there is no significant difference in dis- cretionary accruals between state-owned and non-state-owned companies. However, in the post-convergence period, state-owned companies’ DA is significantly lower than non- state-owned companies’ DA. In other words, the earnings management incentives of state- owned companies are significantly lower than non-state-owned companies, which is consistent with Bo and Wu (2009). CHINA JOURNAL OF ACCOUNTING STUDIES 19 Table 10 presents the regression results for analysing the influence of management incen - tives. OnlyA and BIG4 are added in as control variables too. Except for BIG4 and STATE%, the other explaining variables and control variables show predicted significant associations with the cost of equity capital. POST also presents a significant negative association with the cost of equity capital at the level of 1%. The interaction term of POST*STATE_Dum*DA_Dum pre- sents a significantly negative association with the cost of equity capital, in line with expec - tation, supporting hypothesis H3b. Table 11 presents the results of two-by-two difference-in-difference analysis of manage - ment incentives, respectively with high discretionary accruals and low discretionary accruals subsamples. Those observations with DA_Dum = 1 are attributed to the high DA subsample, and the others to the low DA subsample. Based on Panel B of Table 11, in the cells of differences between pre- and post-conver - gence, the coefficients of POST for both state-owned subsample regressions (−0.022, t = –6.03 under the OJ model, and −0.028, t = −7.86 under the PEG model) and non-state-owned subsample regressions (−0.013, t = −2.90 under the OJ model, and −0.018, t = −2.94 under the under the PEG model) indicate that with low discretionary accruals, both state-owned companies and non-state-owned companies experience a reduction in the cost of equity capital after IFRS convergence. In the bottom-right cells both under OJ model and PEG model, −0.006 and −0.007 are the coefficients of POST*STATE_Dum for low DA subsample regressions, with a significance level at 1%, which signifies state-owned companies enjoy a greater reduction in the cost of equity capital than non-state-owned companies, supporting hypothesis H2b. Based on Panel A of Table 11, state-owned companies also enjoy a greater reduction in the cost of equity capital than non-state-owned companies, judging from the significant coefficients of POST*STATE_Dum for high DA subsample regressions (−0.012, t = −3.45 under the OJ model, and −0.012, t = −3.58 under the PEG model). However, with high discretionary accruals under the OJ model, the reduction in the cost of equity for the non-state-owned companies is not significant (–0.016, t = –1.34); under the PEG model, the reduction of the cost of equity for the non-state-owned companies is only at the significance level of 10%. It is concluded that those companies who present the financial information with less earnings management, benefit more from IFRS convergence. It is observed that the man- agement incentives are another driving factor of the heterogeneity between state-owned and non-state-owned companies, supporting hypothesis H3b. 6. Robustness analysis This paper finally conducts a series of robustness tests to check the reliability of the above results. First, the observations from the transitional period of 2007 are excluded; second, the observations that have not completed EDR are added in. Table 12 presents the regression results of robustness analyses. For brevity, only the coefficients of explaining variables are reported in Table 12. The full sets of regression results are available on request. CAS (2006) has been implemented mandatorily since 2007 in China. Statements of earn- ings and equity adjustments from the old accounting standards to CAS (2006) were required to attach to the annual reports of 2006. By the end of 30 April 2007, the first quarter reports following CAS (2006) had been also released. During the transitional period of 2007, the information disclosed in the capital market was mixed. Therefore, this paper runs the 20 M. CHEN ET AL. Table 11. t wo-by-two analysis of management incentives influence. Panel A: High discretionary accruals o J model Peg model Pre-convergence period Post-convergence period difference ( olS) Pre-convergence period Post-convergence period difference ( olS) *** *** State-owned R = 0.124 n = 129 R = 0.115 n = 630 Coeff. = –0.014 R = 0.116 n = 129 R = 0.101 n = 630 Coeff. = –0.020 e e e e companies t = –2.66 t = –4.06 non-state-owned R = 0.132 n = 16 R = 0.129 n = 284 Coeff. = –0.016 R = 0.123 n = 16 R = 0.115 n = 284 Coeff. = –0.022 e e e e companies t = –1.34 t = –1.88 *** *** *** *** difference ( olS) Coeff. = –0.023 Coeff. = –0.011 Coeff. = –0.012 Coeff. = –0.022 Coeff. = –0.011 Coeff. = –0.012 t = –1.31 t = –3.85 t = –3.45 t = –1.34 t = –3.99 t = –3.58 Panel B: Low discretionary accruals o J model Peg model Pre–convergence period Post–convergence period difference ( olS) Pre–convergence period Post–convergence period difference ( olS) *** *** State-owned R = 0.126 n = 71 R = 0.110 n = 1,198 Coeff. = –0.022 R = 0.118 n = 71 R = 0.096 n = 1,198 Coeff. = –0.028 e e e e companies t = –6.03 t = –7.86 ** *** non-state-owned R = 0.125 n = 32 R = 0.116 n = 439 Coeff. = –0.013 R = 0.116 n = 32 R = 0.102 n = 439 Coeff. = –0.018 e e e e companies t = –2.09 t = –2.94 *** *** *** *** difference ( olS) Coeff. = 0.006 Coeff. = –0.008 Coeff. = –0.006 Coeff. = 0.006 Coeff. = –0.008 Coeff. = –0.007 t = 0.61 t = –3.92 t = –2.77 t = 0.76 t = –4.17 t = –2.96 notes: t he regression model for state-owned subsample and non-state-owned companies is as follows: R = β + β POST + Σβ CONTROLS + ε . t he regression model of pre-convergence subsample e,it 0 1 it j it and pre-convergence subsample is as follows: R = β + β STATE_DUM + Σβ CONTROLS + ε . *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. e,it 0 1 it–1 j it CHINA JOURNAL OF ACCOUNTING STUDIES 21 Table 12. robustness tests. Panel A: Excluding transition period of 2007 Variables Basic analysis heterogeneity analysis Mi influence da influence o J model Peg model o J model Peg model o J model Peg model o J model Peg model *** *** *** *** * * Intercept 0.076 0.054 0.068 0.045 0.029 0.005 0.030 0.007 (5.47) (3.96) (4.87) (3.32) (1.81) (0.32) (1.94) (0.50) *** *** *** * *** * *** POST − –0.011 –0.017 –0.004 –0.010 –0.006 –0.012 –0.006 –0.012 it (–3.85) (–6.04) (–1.35) (–3.14) (–1.89) (–3.72) (–1.95) (–3.76) STATE% − 0.001 0.001 0.001 0.000 0.001 0.001 it–1 (0.33) (0.20) (0.19) (0.06) (0.28) (0.14) MI − 0.001 0.001 it–1 (0.88) (1.27) *** *** DA + 0.066 0.066 it–1 (4.41) (4.54) *** *** *** *** POST *STATE_Dum − –0.009 –0.009 –0.004 –0.004 –0.006 –0.007 it it–1 (–4.11) (–4.29) (–1.55) (–1.52) (–2.75) (–2.88) ** *** POST *STATE_Dum *MI_Dum − –0.006 –0.007 it it–1 it–1 (–2.52) (–2.74) POST *STATE_Dum *DA_Dum − –0.004 –0.004 it it–1 it–1 (–1.62) (–1.66) o ther control variables included included included included included included included included industry controls included included included included included included included included a dj. R 0.10 0.11 0.10 0.12 0.12 0.13 0.12 0.14 f value 15.26 17.21 14.85 16.76 14.19 15.94 14.89 16.67 N 2,285 2,285 2,285 2,285 2,285 2,285 2,285 2,285 Panel B: Pooled sample with EDR uncompleted observations added-in Variables Basic analysis heterogeneity analysis Mi influence da influence o J model Peg model o J model Peg model o J model Peg model o J model Peg model *** *** *** *** ** *** Intercept 0.078 0.054 0.070 0.046 0.032 0.007 0.035 0.011 (0.66) (4.77) (5.93) (4.00) (2.42) (0.53) (2.70) (0.85) *** *** *** *** *** *** *** *** POST − –0.015 –0.021 –0.011 –0.017 –0.013 –0.019 –0.013 –0.018 it (–5.85) (–8.47) (–3.73) (–5.92) (–4.38) (–6.64) (–4.23) (–6.45) *** *** *** *** *** *** *** *** EDR ? 0.010 0.011 0.010 0.0108 0.009 0.011 0.010 0.011 it (3.62) (4.13) (3.52) (4.03) (3.43) (3.92) (3.52) (4.02) * ** * ** * ** STATE% − –0.006 –0.006 –0.006 –0.006 –0.006 –0.007 it–1 (–1.82) (–2.02) (–1.82) (–2.02) (–1.95) (–2.18) ** *** MI − 0.001 0.0014 it–1 (Continued) 22 M. CHEN ET AL. Table 12. (Continued) Panel B: Pooled sample with EDR uncompleted observations added-in Variables Basic analysis heterogeneity analysis Mi influence da influence o J model Peg model o J model Peg model o J model Peg model o J model Peg model (2.57) (3.04) *** *** DA + 0.075 0.076 it–1 (5.96) (6.26) *** *** * ** POST *STATE_Dum − –0.006 –0.007 –0.000 –0.000 –0.004 –0.004 it it–1 (–3.33) (–3.59) (–0.18) (–0.16) (–1.92) (–2.08) *** *** POST *STATE_Dum *MI_Dum − –0.008 –0.009 it it–1 it–1 (–3.90) (–4.21) ** ** POST *STATE_Dum *DA_Dum − –0.005 –0.005 it it–1 it–1 (–2.15) (–2.27) o ther control variables included included included included included included included included industry controls included included included included included included included included a dj. R 0.09 0.10 0.10 0.11 0.11 0.12 0.101 0.13 F value 19.22 22.60 18.68 22.00 17.62 20.73 18.65 21.81 N 3,324 3,324 3,324 3,324 3,324 3,324 3,324 3,324 notes: *, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, all two-tailed. CHINA JOURNAL OF ACCOUNTING STUDIES 23 regressions with a subsample excluding the observations from the transitional period of 2007 to check the sensitivity of the above findings. In Table 12, the key explaining variables of POST, POST*STATE_Dum, POST*STATE_Dum*MI_Dum and POST*STATE_Dum*DA_Dum present significant negative associations with the cost of equity capital, supporting hypoth - eses H1, H2b, H3a and H3b. Taking account of the time overlap between Equity Division Reform and CAS (2006) imple- mentation, a dummy variable of EDR is added in the pooled sample to control the impact of EDR. Panel B of Table 12 shows a significant positive association between EDR and the cost of equity capital, indicating that EDR pushes up the cost of capital. After the EDR, pre- viously non-tradable shares become tradable. Although these newly tradable shares are not allowed to be traded within certain periods, the market is also concerned with the effect of drawing the market down. This could explain the significant positive correlation between EDR and the cost of equity capital. In Panel B of Table 12, controlling EDR and other factors, the signic fi antly negative coec ffi ients of POST, POST*STATE_Dum, POST*STATE_Dum*MI_Dum and POST*STATE_Dum*DA_Dum ensure the reliability of the above findings. This paper also conducts other sensitivity tests, such as replacing EPS with ROA, measuring SIZE with the natural logarithm of total assets and winsorising the top and the bottom at 1%. The coefficients of explaining variables are still highly significant, consistent with the above. 7. Conclusions In summary, the cost of equity capital reduces significantly after implementation of the IFRS-converged accounting standards in China. Compared with non-state-owned companies, the state-owned companies enjoy a greater reduction in the cost of equity capital. The two- by-two difference-in-difference analysis documents that non-state-owned companies in low-marketisation environments or with high discretionary accruals do not experience a reduction in the cost of equity capital. Institutional settings and management incentives are critical driving factors for state-owned companies to benefit more from IFRS convergence. The contributions of this paper are summarised as follows. First, this paper extends the literature on the IFRS convergence influence on the reduction in the cost of equity capital by analysing the heterogeneity and exploring the driving factors of heterogeneity in China. Second, explicit influence evidences of institutional settings and management incentives on IFRS convergence have been documented, which are emphasised but not yet examined in prior studies. Third, transitional economies might wish to consider the evidence relating to China. And above all, the findings of this paper have policy implications for regulators, contributing to the appraisal of the convergence approach and highlighting the critical factors in the implementation of IFRS-converged accounting standards in transitional econ- omies. The process of development of external marketisation and the governance of incen- tives for earnings management should not be ignored, in order to achieve the benefit of IFRS convergence. In addition, this paper is still subject to the following caveats. First, apart from MI and DA, it is better to find other proxy variables to represent institutional settings and management incentives. Second, it may be better to find another way, rather than the employment of a dummy variable, to control for the overlap period between the Equity Division Reform and the application of CAS (2006). Third, the significant positive correlation between MI and the cost of equity capital should be reinvestigated. 24 M. CHEN ET AL. Funding This work was supported by National Social Science Foundation of China [13AJY005], Humanities and Social Sciences Key Base of the Ministry of Education [16JJD790035], Fujian Federation of Social Science [FJ2016B243] and Natural Science Foundation of China [71402114]. References Armstrong, C., Barth, M.E., Jagolinzer, A.D., & Riedl, E.J. (2010). Market reaction to the adoption of IFRS in Europe. The Accounting Review, 85(1), 31–61. Atwood, T.J., Drake, M.S., Myers, J.N., & Myers, L.A. (2011). Do earnings reported under IFRS tell us more about future earnings and cash flows? Journal of Accounting and Public Policy, 30, 103–121. Ball, R., Robin, A., & Wu, J.S. (2003). Incentives versus standards: Properties of accounting income in four East Asian countries. Journal of Accounting and Economics, 36, 235–270. Ball, R., Li, X., & Shivakumar, L. (2015). Contractibility and transparency of financial statement information prepared under IFRS: Evidence from debt contracts around IFRS adoption. Journal of Accounting Research, 53(5), 915–963. Barth, M.E., Landsman, W.R., & Lang, M.H. (2008). International Accounting Standards and accounting quality. Journal of Accounting Research, 46, 467–498. Bartov, E., Goldberg, S.R., & Kim, M. (2005). Comparative value relevance among German, U.S. and International Accounting Standards: A German stock market perspective. Journal of Accounting, Auditing and Finance, 20, 95–119. Berker, C., DeFond, N., & Subramanyam, K. (1998). The effect of audit quality on earnings management. Contemporary Accounting Research, 15(1), 1–24. Bo, X., & Wu, L. (2009). The governance roles of state-owned controlling and institutional investors: A perspective of earnings management. Economic Research, 2, 81–91 (in Chinese). Botosan, C.A. (1997). Disclosure level and the cost of equity capital. The Accounting Review, 72, 323–349. Botosan, C.A., & Plumlee, M.A. (2002). A re-examination of disclosure level and the expected cost of equity capital. Journal of Accounting Research, 40(1), 21–40. Burgstahler, D., Hail, L., & Leuz, C. (2006). The importance of reporting incentives: Earnings management in European private and public firms. The Accounting Review, 81, 983–1016. Cang, Y., Chu, Y., & Lin, T.W. (2014). An exploratory study of earnings management detectability, analyst coverage and the impact of IFRS adoption: Evidence from China. Journal of Accounting and Public Policy, 33, 356–371. Capkun, V., Cazavan-Jeny, A., Jeanjean, T., & Weissc, L.A. (2008). Earnings management and value relevance during the mandatory transition from local GAAPs to IFRS in Europe. SSRN eLibrary. http:// papers.ssrn.com/sol3/papers.cfm?abstract_id=1870007 Chen, H., Tang, Q., Jiang, Y., & Lin, Z. (2010). The role of international financial reporting standards in accounting quality: Evidence from the European Union. Journal of International Financial Management and Accounting, 21, 220–278. Chen, M., & Qu, X. (2014). The influence of international convergence of accounting standards on accounting information quality: A systematic test. Contemporary Accounting Review, 7(1), 1–27 (in Chinese). Chen, T., Chin, C., Wang, S., & Yao, W. (2015). The effects of financial reporting on bank loan contracting in global markets: Evidence from mandatory IFRS adoption. Journal of International Accounting Research, 14(2), 45–81. Christensen, H.B., & Lee, E. (2008). Incentives or standards: What determines accounting quality changes around IFRS adoption? Working paper. Christensen, H.B., Lee, E., & Walker, M. (2007). Cross-sectional variation in the economic consequences of international accounting harmonization: The case of mandatory IFRS adoption in the UK. The International Journal of Accounting, 42, 341–379. Christensen, H.B., Lee, E., & Walker, M. (2009). Do IFRS reconciliations convey information? The effect of debt contracting. Journal of Accounting Research, 20, 1–33. CHINA JOURNAL OF ACCOUNTING STUDIES 25 Daske, H. (2006). Economic benefits of adopting IFRS or US-GAAP: Have the expected cost of equity capital really decreased? Journal of Business Finance and Accounting, 33, 329–373. Daske, H., & Gebhardt, G. (2006). International financial reporting standards and experts’ perceptions of disclosure quality. Abacus, 43, 461–498. Daske, H., Hail, L., Leuz, C., & Verdi, R. (2008). Mandatory IFRS reporting around the world: Early evidence on the economic consequences. Journal of Accounting Research, 46, 1–58. DeAnglo, L.E. (1981). Auditor size and audit quality. Journal of Accounting and Economics, 3, 183–199. DeFond, M., & Jiambalvo, J. (1993). Factors related to auditor-client disagreements over income- increasing accounting methods. Contemporary Accounting Research, 9, 415–431. DeFond, M., Gao, X., Li, O., & Xia, L. (2014). Did China’s adoption of IFRS attract more foreign institutional investment? SSRN eLibrary. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2426484 DeFond, M., Hu, X., Hung, M., & Li, S. (2011). The impact of mandatory IFRS adoption on foreign mutual fund ownership: The role of comparability. Journal of Accounting and Economics, 51, 240–258. Ding, Y., & Su, X. (2008). Implementation of IFRS in a regulated market. Journal of Accounting and Public Policy, 27, 474–479. Du, X., Lei, Y., & Zhu, G. (2009). Market reactions to Chinese Accounting Standards (2006): Preliminary empirical evidence. Accounting Research, 3, 18–25 (in Chinese). Easley, D., & O’Hara, M. (2004). Information and the cost of capital. The Journal of Finance, 59, 1553–1583. Easton, P. (2004). PE ratios, PEG ratios, and estimating the implied expected rate of return on equity capital. The Accounting Review, 79(1), 73–95. Fan, G., Wang, X., & Zhu, H. (2011). China’s Marketisation Index. Beijing: Economics Science Publication (in Chinese). Francis, J., Olsson, P.M., & Schipper, K. (2004). Cost of capital and earning attributes. The Accounting Review, 79, 967–1010. Gao, F., & Fu, R. (2012). Accounting standards reform, stock liquidity and cost of equity capital: Evidence from Chinese A-share listed companies. China Management Science, 20, 27–36 (in Chinese). Gebhardt, W., Lee, C., & Swaminathan, B. (2001). Toward an implied cost of capital. Journal of Accounting Research, 39(1), 135–176. Green, S. (2003). China’s stock market: Eight myths and some reasons to be optimistic. The Royal Institution of International Affairs, University of Cambridge. Gode, D. (2003). Inferring the cost of capital using the Ohlson-Juettner Model. Review of Accounting Studies, 8, 399–431. Guo, Z. (2011). International big four accounting firms and high auditing quality-Evidence from China’s securities market. Audit Research, 1, 98–107 (in Chinese). Hail, L. (2002). The impact of voluntary corporate disclosures on the ex-ante cost of capital for Swiss firms. European Accounting Review, 11, 741–773. Hail, L., & Leuz, C. (2006). International differences in the cost of equity capital: Do legal institutions and securities regulation matter? Journal of Accounting Review, 44, 485–531. He, X., Wang, T.J., & Young, D. (2012). Challenges for implementation of fair value accounting in emerging markets: Evidence from China. Contemporary Accounting Research, 29(2), 538–562. Hong, H.A., Hung, M., & Lobo, G.J. (2014). The impact of mandatory IFRS adoption on IPOs in global capital markets. The Accounting Review, 89, 1365–1397. Hou, Q., Jin, Q., & Wang, L. (2014). Mandatory IFRS adoption and executive compensation: Evidence from China. China Journal of Accounting Research, 7(1), 9–29. Hou, Q., Jin, Q., Wang, L., & Zhang, G. (2016). Mandatory IFRS adoption, accounting quality, and investment efficiency: Evidence from China. China Journal of Accounting Studies, 4, 236–262. Hung, M.Y., & Subramanyam, K.R. (2007). Financial statement effects of adopting International Accounting Standards: The case of Germany. Review of Accounting Studies, 12, 623–658. Iatridis, G., & Rouvolis, S. (2010). The post-adoption effects of the implementation of International Financial Reporting Standards in Greece. Journal of International Accounting, Auditing and Taxation, 35, 1–11. IASB & FASB. (2010). Conceptual Framework for Financial Reporting. London: IFRS Foundation Publications Department. 26 M. CHEN ET AL. Jaafar, A., & McLeay, S. (2007). Country effects and sector effects on the harmonization of accounting policy choice. Abacus, 43, 156–189. Jeanjean, T., & Stolowy, H. (2008). Do accounting standards matter? An exploratory analysis of earnings management before and after IFRS adoption. Journal of Accounting and Public Policy, 27, 480–494. Lee, E., Walker, M., & Christensen, H.B. (2010). Mandating IFRS: Its impact on the cost of equity capital in Europe. Journal of International Accounting Research, 9(1), 58–59. Leuz, C. (2003). IAS versus U.S. GAAP: Information asymmetry-based evidence from Germany’s new market. Journal of Accounting Research, 41, 445–472. Leuz, C., & Verrecchia, R.E. (2000). The economic consequences of increased disclosure. Journal of Accounting Research, 38(Supplement), 91–124. Li, S. (2010). Does mandatory adoption of International Financial Reporting Standards in the European Union reduce the cost of equity capital? The Accounting Review, 85, 607–636. Li, J., & Park, S. (2011). The value relevance effects of IFRS adoption: The case of the People’s Republic of China. Korea International Accounting Review, 38, 199–218. Liu, F., & Zhou, F. (2007). Does size really matter? A perspective of conservatism test. Accounting Research, 3, 79–88 (in Chinese). Liu, C. (2015). Change of accounting standards, sector heterogeneity and efficiency of resource allocation. The Chinese CPA, 10, 109–115 (in Chinese). Luo, T., Xue, J., & Zhang, H. (2008). The effect of the new accounting standards on value relevance of accounting information. China Accounting Review, 6, 129–140 (in Chinese). Ohlson, J. (1995). Earnings, book values, and dividends in equity valuation. Contemporary Accounting Research, 11, 661–687. Ohlson, J., & Gao, Z. (2006). Earnings, earnings growth and value. Foundations and Trends in Accounting, 1(1), 1–51. Ozkan, N., Singer, Z., & You, H. (2012). Mandatory IFRS adoption and the contractual usefulness of accounting information in executive compensation. Journal of Accounting Research, 50, 1077–1107. Peng, S., & Smith, J.L. (2010). CAS and IFRS: An analysis of the convergence process. Journal of International Accounting, Auditing and Taxation, 19, 16–34. Peng, S., Tondkar, R.H., Smith, J.L., & Harless, D.W. (2008). Does convergence of accounting standards lead to the convergence of accounting practices? A study from China. The International Journal of Accounting, 43, 448–468. Platikanova, P. (2007). Market liquidity effects of the IFRS introduction in Europe. Working paper. Available at: http://ssrn.com/abstract=1005364 Qu, X., & Zhang, G. (2010). Measuring the convergence of national accounting standards with International Financial Reporting Standards: The application of fuzzy clustering analysis. The International Journal of Accounting, 45, 334–355. Radebaugh, L.H., & Gray, S.J. (1997). International Accounting and Multinational Enterprises (4th ed., p. 73). New York: Wiley. Ruland, W., Shon, J., & Zhou, P. (2007). Effective control for research in international accounting. Journal of Accounting and Public Policy, 26(1), 96–116. Schultz, T.W. (1968). Institutions and the rising economic value of man. American Journal of Agricultural Economics, 50, 1113–1122. Shelton, S., Jackson, L., & Robinson, D. (2011). IFRS and U.S. GAAP: Assessing the impact of reporting incentives on firm restatements in foreign and U.S. markets. Advances in Accounting, incorporating Advances in International Accounting, 27, 187–192. Shen, H. (2008). Market segmentation, cross listing and expected capital cost. Financial Research, 2, 146–155 (in Chinese). Soderstrom, N.S., & Sun, K.J. (2007). IFRS adoption and accounting quality: A review. European Accounting Review, 16(4), 675–702. Sun, X. (2011). Foreign investment, market liquidity and accounting standards convergence—Evidence from Chinese capital market. Journal of Shanxi University of Finance and Economics, 33, 108–115 (in Chinese). CHINA JOURNAL OF ACCOUNTING STUDIES 27 Tweedie, D. (2006a). Prepared statement of Sir David Tweedie, before the Economic and Monetary Affairs Committee of the European Parliament. IASPLUS. http://www.iasplus.com/ resource/0601tweedieeuspeech.pdf Tweedie, D. (2006b). Statement of Sir David Tweedie, at the ceremony announcing the adoption of new accounting standards by Chinese Ministry of Finance. IFRS.org http://www.ifrs.org/News/ Announcements+and+Speeches/China+affirms+commitment+to+converge+with+IFRSs.htm Voulgaris, G., Stathopoulos, K., & Walker, M. (2014). IFRS and the use of accounting-based performance measures in executive pay. The International Journal of Accounting, 49, 479–514. Wang, Y., & Ye, Z. (2011). Does implementation of new accounting standards reduce the cost of equity capital: Evidence from Chinese capital market. China Industrial Economics, 3, 119–128 (in Chinese). Wang, J.W., & Yu, W. (2009). The information content of stock prices, reporting incentives, and accounting standards: International evidence. Hong Kong Polytechnic University Working Papers. Wang, Y., Xue, J., & Chen, X. (2009). Accounting choice and earnings management: An analysis on the implementation of new accounting standards. China Accounting Review, 7, 256–269 (in Chinese). World Bank. (2009). Report on the Observance of Standards and Codes (ROSC)–Accounting and Auditing: People’s Republic of China (pp. 12–13). Washington, DC: World Bank. Wu, J.S., & Zhang, I.X. (2009). The voluntary adoption of internationally recognized accounting standards and firm internal performance evaluation. The Accounting Review, 84, 1281–1309. Xin, Q., & Wang, B. (2010). Cross-border listing, big4 and accounting earnings quality. Economics Science, 4, 96–110 (in Chinese). Yu, B. (2009). Firm size, ownership type and the implementation of new Chinese Accounting Standards: From the perspective of value relevance of accounting information. Journal of Zhongnan University of Economics and Law, 5, 127–131 (in Chinese). Zhang, G. (2013). Accounting standards, cost of capital, resource allocation and welfare in a large economy. The Accounting Review, 88, 1459–1488.
Journal
China Journal of Accounting Studies
– Taylor & Francis
Published: Jan 2, 2017
Keywords: IFRS convergence; cost of equity capital; heterogeneity; institutional settings; management incentives
