In this study we review the situation of seasoned equity offerings in Chinas stock market, and examine 2 questions. Do firms engage in earnings management prior to seasoned equity offerings? If so, can we find an easy way to find out this kind of manipulation so that the investors could make the right decisions, for example whether corporate governance factors would relate to earnings management?
We examine 102 firm-issue seasoned equity offerings in a period of 6 years. In addition, we emphasize particularly on the differences divided by the presence of state-holdings, and we do find some interesting results. Moreover, we investigate how the proceeds of SEOs are affected by financial and ownership structure, and corporate governance. We find clear evidence of earnings management around SEOs prior the issue year, but the extent of influence of managing earnings is relatively weak so that we can't point out that the proceeds of SEOs have direct links with this manipulation. However, the interesting findings come out when we explore the same relations by examining sub-samples divided by whether it exists state-holding shares. The effect of discretionary current accruals (DCAs) on the proceeds of SEO does exist significantly for the sub-sample non state-holding firms. But we fail to find any significant association between corporate governance, SEO proceeds and earnings management. Unfortunately for investors, there is not a shortcut for them to evaluate whether firm's performance is exaggerated or not prior to SEO.
Keywords: Seasoned equity offerings (SEO), earnings management, corporate governance
RÉSUMÉ
Dans cette étude, nous passons en revue la situation de l'augmentation du capital (SEO) en bourse de la Chine, et d'examiner deux questions. Est-ce que les entreprises s'engagent dans la gestion des résultats avant l'augmentation du capital ? Si oui, pouvons-nous trouver une façon simple à découvrir ce genre de manipulation pour que les investisseurs pourraient prendre les bonnes décisions, par exemple si des facteurs de gouvernance d'entreprise se rapporteraient à la gestion des résultats ?
Nous examinons 102 sujets de l'augmentation du capital dans une période de 6 ans. En outre, nous insistons particulièrement sur les différences divisées par la présence de participation de l'état, et nous avons trouvé des résultats intéressants. Par ailleurs, nous examinons comment les rapports de l'augmentation du capital sont affectés par la structure financière et de la propriété, et la gouvernance d'entreprise. Nous trouvons une preuve claire de la gestion des résultats dans l'année précédant de l'augmentation du capital, mais l'ampleur de l'influence de la gestion des résultats est relativement faible pour que nous ne puissions pas remarquer que les rapports de l'augmentation du capital ont des liens directs avec cette manipulation. Toutefois, les résultats intéressants ressortent quand nous explorons les mêmes relations en examinant des sous-échantillons divisées selon s'il existe des actions détenues par l'état. L'effet de la provision courante discrétionnaire (DCA) sur les rapports de l'augmentation du capital existe significativement pour le sous-échantillon non-participation de l'état. Mais nous n'arrivons pas à trouver une association significative entre la gouvernance d'entreprise, les rapports de l'augmentation du capital et la gestion des résultats. Malheureusement pour les investisseurs, il n'y a pas un raccourci pour eux à évaluer si la performance des entreprises est exagérée ou non avant l'augmentation du capital.
INTRODUCTION
There always exists a strong motivation for senior management to manage earnings when firms are about to execute seasoned equity offerings. Managers can increase reported earnings invisibly and benefit third party by selling their firm's share under a favorable condition. In support of this, Rangan (1998), Teoh et al. (1998) and Marquardt and Wiedman (2004) have shown that managers manipulate earnings in periods around SEOs. Their studies explore that managers overstate earnings before SEOs because of opportunism. Marquardt and Wiedman (2004) further argue that earnings of SEO firms are managed increasing by accelerating the recognition of income. Shivakumar (2000) also reports earnings management by firms making SEOs, but he suggests that investors are able to identify these discretionary accruals and price the stock according to its fundamentals. He argues that earnings management at the time of an SEO is a rational behavior by the managers as the stock market expects them to engage in this practice.1 Most research studies on earnings management and all the studies cited above use data from the US. In comparison, there is dearth of research in other countries.
It's very important for investors, analysts and regulators to investigate whether earnings are managed artificially over SEO announcements. Teoh et al. (1998) and Rangan (1998) argue that poor post-issue performance2 can be explained partly by the pre-issue earnings management of seasoned new issuers. 3 This implies investors could use the information contained in the pre-offering accounting accruals to distinguish among seasoned equity issuers4 and invest in those firms with negative or low discretionary current accruals. Of course, this is not an absolute principle for investors especially for primary investors. Also the professional skills are necessary
1 Shivakumar (2000) calls this argument the managerial response hypothesis. This hypothesis contrasts with the overvaluation hypothesis that says managers have incentives to issue new equity when the firm is overvalued. 2 Empirical studies in the US have documented poor post-issue earnings performance (Hansen and Crutchley, 1990; McLaughlin et al., 1996) and poor post-issue stock returns (Loughran and Ritter, 1995; Spiess and Affleck-Graves, 1995). Brous (1992) and Jain (1992) find that security analysts significantly lower their forecasts of short-term earnings of issuing firms in response to SEO announcements. 3 Note, however, that Shivakumar (2000) finds no association between abnormal accruals and long-run stock returns. 4 Discretionary accruals in the period surrounding the offering predict a portion of the subsequent poor earnings and stock price performance.
to reveal the earnings management. If investors can't well understand the accounting practices, they will be induced to buy the stocks of those firms at a higher price because of the exaggerated but false reported earnings.
In this study we review the situation of seasoned equity offerings (SEO) in China's stock market, and examine 2 questions. Do firms engage in earnings management prior to seasoned equity offerings? If so, can we find an easy way to find out this kind of manipulation so that the investors could make the right decisions, for example whether corporate governance factors would relate to earnings management?
It is important that national studies should be undertaken because the prevalence and extent of earnings management are likely to be functions of the specific environment that the firm operates in. For example, good corporate governance mechanisms and effective monitoring activities will decrease possibility or frequency of earnings management, as well as the effect of these factors differs across firms and across national legal system. Because corporate governance, the legal environment and monitoring activities are far different in China than in the US, research studies using American data have limited relevance for China. Firms in China are predominantly controlled by state or families. This control is enhanced through pyramid structures and cross-share holdings among firms (Claessens et al., 2000). The major agency problem in China exits between the controlling stockholder and the minority stockholders. Comparing to the situation in US, as its capital market is very mature and few of firms have a relatively dominated shareholder, therefore its main agency problem lies between the professional managers and the stockholders. For public firms in China, if an owner has absolute power to control the creation of firm's accounting information and its statement policies; as a result he could report accounting information more for self-interested purposes than to reflect the firm's true underlying economic performance.
It is often claimed that board of directors structured to be independent of management is best able to perform its independent oversight functions (Fama, 1980; Fama and Jensen, 1983). But in China, the board is dominated by the controlling stockholders who also usually take the CEO and chairman positions. Companies have independent non-executive directors who are supposed to safeguard the interests of the non-controlling (or minority) stockholders. A larger proportion of independent directors should, in theory at least, be able to better control the opportunistic actions of top management and the controlling stockholders. Due to the limit of our resource data, we cite a governance index and use several basic governance factors to investigate the association possibly existing so that investors could easily find.
In this study, we examine 102 firm-issue seasoned equity offerings during a period of 6 years. In addition, we emphasize particularly on the differences divided by the presence of state-holdings, and we do find some interesting results. Moreover, we investigate how the proceeds of SEOs are affected by financial and ownership structure, as well as corporate governance. We find clear evidence of earnings management around SEOs prior the issue year, but the extent of influence of managing earnings is relatively weak so that we can't point out that the proceeds of SEOs have direct links with this manipulation. However, the interesting findings come out when we explore the same relations by examining sub-samples divided by whether it exists state-holding shares. The effect of discretionary current accruals (DCAs) on the proceeds of SEO does exist significantly for the sub-sample non state-holding firms. But we fail to find any significant association between corporate governance, SEO proceeds and earnings management. Unfortunately for investors, there is not a shortcut for them to evaluate whether firm's performance is exaggerated or not prior to SEO.
In this section, we will briefly introduce the background of our research topics, also we will present some featured numbers to make you have a general idea of the significance of this study.
In the second section, we will discuss respectively the situations and the problems in the context of China about our three topics, earnings management, corporate governance and seasoned equity offerings.
In the third section, we will present the statistics results.
In the final section, we will review our hypothesis and results, and conclude the meaningful findings about state-holding effect in China's stock market.
Stock market in China
The stock market is a barometer of the national economy, while it is also a product of economic development and plays an important role in economic development. The development of new economic growth theory, as well as the rapid development of the stock market, makes the study of the stock market and its impact on economic growth mechanism seems more important than ever.
Especially for China, China's stock market had played a critical role in helping
state-owned enterprises transform into joint stock companies that allow non-state capital investment.
Shanghai Stock Exchange (referred to as SSE), one of two stock exchanges in Mainland China (another is Shenzhen Stock Exchange, opened on December 1 1990, located in Luohu District, Shenzhen), located in Pudong New Area. Shanghai Stock Exchange established on November 26, 1990, approved by the People's Bank of China, officially opened on December 19 in the same year. Shanghai Stock Exchange is non-for-profit legal person (corporation), directly managed China Securities Regulatory Commission (CSRC).
Its main functions include:
? providing of securities trading venues and facilities;
? development of stock exchange business rules;
? accept of listing application, arrangement of listed securities; organization, supervision of securities transactions;
? regulation of its members and listed companies; management and publication of market information.
Shanghai Stock Exchange is the member of the International Organization of Securities Commissions, the Asian-Oceania Federation of Exchanges, World Federation of Exchanges.
In 2010, the trading volume of Shanghai Stock Exchange ranks third in the world after New York Stock Exchange and NASDAQ (the fourth for Tokyo Stock Exchange). Following types of securities can be traded in Shanghai Stock Exchange:
? Stocks
? Funds
? Bonds
? Warrants
Due to the national strategy for Shanghai to improve its position in the international financial market, we focus on the activities in Shanghai Stock Exchange. Here are some data to point out its importance in China.
Security size of Shanghai Stock Exchange (until August 2011)
? Number of listed companies: 922
? Number of A-share: 912
? Number of B-share: 54
? Total market value of A-share and B-share: 16,747.164 billion yuan
? Float market capitalization of A-share and B-share: 13,73427 billion yuan
? Average P/E ratio (price/earnings): 15.2
? Number of listed funds: 34
? Number of listed bonds: 627
? Number of listed warrants: 0
Table 1 provides the total amount of funding-raising in Shanghai Stock Exchange from its establishment, and Figure 1 provides the evolution of year fund-raising amount.
* Subject to the listing date, statistics include IPO, SEO, allotment and bonus shares.
Figure 1: Evolution of fund-raising amount
I. Literature and Background
I.1. Situation of refinancing in China
Refinancing of the listed company refers to the activities of direct funding in the exchange market through the approaches of allotment of shares, placement of shares and issue of convertible bonds, etc.
Figure 2 reports the proportion of total financing amount from January 1991 to July 2011. Refinancing has played a great role (nearly 50% of funding amount) for Chinese listed companies in promoting China's securities market, and it has attached more and more importance.
Figure 2: Distribution of the amount of financing from 1991 to 2011
Following two tables give financing details respectively for the period 2000 to 2009 and period 2002 to 2007.
Table 2: Historical data during sample period
* Subject to the listing date, statistics include IPO, SEO, allotment and bonus shares.
* Subject to the listing date, statistics include IPO, SEO, allotment and bonus shares.
Along with the refinancing boom in China, some problems have occurred.
I.1.1. Problems of refinancing of listed companies
a. Single financing method, mostly by equity financing.
Listed companies have a strong preference for equity financing. Until 1998, the allotment of shares is the only way for listed company to refinance; since 2000, placement issuing became an alternative choice of re-financing for public company; from 2001, the financing method of convertible bonds turned to be the object of pursuit by public company.
The factors that China's listed companies concern about when choosing re-financing methods are mainly the difficulty of financing, the threshold level and the size of financing amount.
For now, equity financing would be the first choice for refinancing of listed companies. The ownership structure of listed company is rather special in China, the legal person shares that can't be traded accounted for 60% on average. In this case, equity financing has a certain effect on improving the ownership structure. However, a single equity financing does not take into account the differences of capital structure of companies, which does not meet the financing principle of optimal capital structure. Some studies on the listed companies conducting the allotment in 1997 in Shenzhen stock exchange found that the average asset-liability ratio was 43.29%, the equity financing of such companies tends to make their capital structure be much more unreasonable.
b. Financing amount exceeds its actual needs.
In theory, there is a dialectical relationship between investment needs and financing method, only when the necessity of investing and the possibility of financing combine well, it would generate better investment results. However, most listed companies usually refinance in accordance with the upper limit of policies, instead of determining the amount of financing by the investment demand. Raising as much money becomes an important goal for listed companies and the ground to choose re¬financing options and formulate the program of issuance, therefore, the financing amount often surpasses its actual capital needs, resulting in inefficient use of funds and other issues.
c. Blindness and uncertainty of the investment in use of financing.
Over the years, public companies generally do not pay attention to feasibility study for investment projects, resulting in frequent changes in raised funds and a low investment income, which is obviously a sign of expropriation.
Number of listed companies are lack of adequate research on the investment projects so that the raised fund can't be put into as planned, resulting in varying degrees of idle fund, and some have to change the investment orientation.
According to statistics, in a sample consisting of public companies that execute the IPO, placement and allotment in the first half of 2000, to the end of 2000, experiencing on average more than a six months' time, the mean using of the proceeds achieves only 46.15%.
From the prospectus, we can find out that the construction period of investment projects for majority of enterprises is about six months or one year. Many companies even put the raised fund into government bonds, or in the bank, else wise participate in the placing of new shares.
A considerable number of enterprises due to the new re-financing, they have to rapidly use the funds raised last time. As raised fund can't be completely used as planned, for seeking short-term returns, lots of public companies have started trust management business. If things continue in this way, this kind of refinancing of listed companies would not only promote the development of enterprises, but also result in the deviation of fund-raising orientation and the inefficient use of funds etc.
d. Dividend distribution policy at random.
Public companies do not develop a dividend policy that could ensure the company development as well as a stable return to investors; the dividend distribution program that management launched has a strong randomness. Dividend policy of listed companies failed to combine long-term development strategy, the majority of public investors cannot get a higher dividend yield through the dividend distribution.
e. Post-financing issue.
In recent years, the decline of listed companies' benefits after refinancing becomes a serious problem. According to incomplete statistics, among 34 companies that conduct the allotment and placement in 2000, 26 of them get their annual net return on assets declined, 13 of them get a significantly lower net profit than the previous year, only 8 companies get yield improved.
2001 annual report shows that among companies conducting allotment or placement in 2000 and early 2001, there are 30 companies publishing loss in annual reports or announcements of estimation of loss.
Inefficient financing of listed companies, and performance landslide, would gradually weaken investors' will. If this problem cannot be resolved in a long-term, it would extremely harm the listed companies and securities market.
I.1.2. Reasons for problems above
a. Low cost of capital of the equity financing.
The choice of financing is largely restricted by the cost of capital. The cost of equity financing is the dividend, and the cash dividend is the actual cost of capital that requires companies to pay.
In a long-standing in China's securities market, it always exist an emphasis on stock dividend instead of cash dividend. In China, the listed company's dividend distribution policy is mainly formulated by the board of directors being elected by big shareholders, a dividend distribution policy becomes a poise that authorities could free control, so the external actual cost of equity financing would be the costs that authorities can control. Relative to the hard constraints of the interest returns on bond financing, the administration of listed companies prefers the soft constraints of
dividend distribution. This is an important reason why China's listed companies prefer
equity financing.
b. Corporate bond market is not yet mature.
Corporate bond market immaturity is reflected in the following two aspects:
Firstly, lagged legislation. The current legislation about bond issue, "Corporate Bond Management Regulation", was enacted in 1993. The regulation provides that the corporate bond interest rate shall not be higher than the 40% of fixed-term deposit rate in the corresponding time period. Because of this provision, even if the corporate bond was issued at the highest interest rates limited by policy, furthermore it needs to deduct the interest tax on returns of corporate bond for investors. As a result, the real returns of corporate bond are almost the same as the low-risk government bond yields; the corporate bond has no attraction for investors.
Secondly, the size of corporate bond market is small. At that time (2001), there are only 10 corporate bonds in Shanghai Stock Exchange, and the issuers are 6 large state-owned enterprises. Many enterprises issuing bonds are not traded in the security market, the liquidity of bonds is not high enough.
Based on the two reasons above, in regard with safety, liquidity and profitability, investors are not optimistic about corporate bond, making this the most important financing method in Western countries as it is still in its infancy in current China's security market. Therefore, more listed companies choose allotment or placement.
c. Special ownership structure.
Statistics show that at the end of 2001, the percentage of tradable shares for public investors over total shares is only about 34%, while the state-holding shares account for 47% and other non-tradable shares account for 19%.
Overall, the state shareholders have absolute control in listed company. Even if the state shareholders only have relative control in some listed companies, but due to very scattered public investors, it results in that the majority on the number of public investors is difficult to gain control of listed companies, public investors cannot really participate in decision-making.
In this case, management's decision does not represent the interests of most public shareholders; largely represent the interests of a few large shareholders.
As the premium distribution of refinancing, the equity of existing shareholders has grown rapidly through financing, but for new shareholders, their interests are diluted.
As the larger shareholders can obtain additional growth of their equity through
SEO, so the major shareholders with the power of decision-making possess stronger incentive to the equity financing. Furthermore, some of listed companies would launch cash dividend program soon after the equity financing, and take the large part of dividend in accordance with its proportion of shares.
d. Policy-oriented role.
Under the system of approval, the refinancing conditions become much more tighter, the examination period gets longer, and listed company hopes to raise as much cash as it could. With the promulgation of IPO management practices in 2001, many companies have introduced additional issue programs, booming the SEOs. In the second half of 2001 due to market reasons, the approval of additional more difficult, many companies turn to issue convertible bonds that is less stringent than SEO issuing.
In March 2001, the promulgated "Management measures for listed companies to issue new shares" put the dividend payout as the key concerns while refinancing.
During that time, in the guide of "China security supervisory committee equity issuance examination committee, instructions on the equity issue of public companies", it is proposed to should concern about the previous dividend payment since the last three years, in particular the proportion of cash dividends over allocated profits, as well as the reasons the board of directors stated. As a result, majority of listed companies starts to choose cash dividends from year 2001.
These circumstances indicate that the policy play an important guidance role for the refinancing.
In the part of refinancing, seasoned equity offerings account for more than 70% of the total amount. We focus on this part and examine the possible relations between SEO, earnings management and corporate governance. First of all, we specify the definition of SEO in China.
I.1.3. SEO in China
Seasoned equity offerings include new equity placement and allotment in China. The allotment is kind of rights offering, that listed companies issue new shares to existing holders in terms of their holding proportion. It refers to the activities of financing through issuing shares in a certain price and in terms of holding proportion when listed companies need capital to extend operation.
For rights issuing, there is a priority for existing shareholders to subscribe for new
shares according to their stake of existing share proportion. If existing shareholders do not participate in rights issue, their interests will be "diluted".
For example, if the proportion of a shareholder's holding is 5%, then he has the right to subscribe for 5% shares during the issue of new shares.
The placement, which is called cash re-issue in foreign countries, is an offering to the public for capital financing.
Figure 3 and Figure 4 show directly the comparison result that equity placement becomes the preferred refinancing method in recent years no matter on the amount of refinancing or on the proportion of refinancing.
Figure 3: Refinancing amount from 2000 to 2009
Figure 4: Evolution of preferred refinancing method
The reasons why the allotment is not as popular as before are as follows:
? There is an upper-limit for allotment of 30% of total shares amount; by contrast no limit on shares or financing amount for placement. Besides, companies should also admit the share amount of allotment and issue it in the way of consignment, which would increase the allotment cost.
? Allotment will damage the interests of tradable-share holders, they pay for new shares but normally, after the allotment, the profit rate will decline; by contrast non-tradable shareholders don't have any actual pay out, and they will benefit from allotment.
I.2. Earnings management in China
Healy and Wahlen (1999, P.368) define earnings management as follows:
"Earnings management" occurs when managers use judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of a company or influence contractual outcomes that depend on reported accounting numbers.
Accrual-based accounting provides options for managers on reporting earnings. Managers can report higher earnings without breaking accounting rules in a variety of ways, such as accelerating the recognition of profits, shifting income from future periods to the present, and deferring the recognition of certain expenses. Such earnings manipulation makes firm's financial statement less informative to investors who rely on firm-reported information when assessing the "true" financial situation of a firm (Marquardt and Wiedman, 2004).
Lots of similar studies report large positive earnings and abnormal accruals prior to equity offerings, after that those firms experience a poor long-run earnings and weak stock return performance corresponding to non-issuing industry peers (e.g., Teoh, Wong, and Rao, 1998; Teoh, Welch, and Wong, 1998a, 1998b; Rangan, 1998; Kim and Park, 2005). Investors appear to misunderstand reported high earnings at the time of equity offerings and consequently overprize the new issues.
Analysis of earnings management of listed companies
Earnings management exists commonly in listed companies over developing and developed countries. In recent years, the earnings management of listed companies has also become the focus of people's attention in China.
Excessive earnings management, once beyond the requirements of accounting standards, it will be called as earnings manipulation, and its accounting information will lead to not truly reflect the company's economic strength and prospects for development, to mislead investors, endanger the optimization function of resource allocation in security market.
I.2.1. Purposes of earnings management
First, the purposes of earnings management for listed companies are as follow:
a. to obtain qualification of shares issuance.
The greatest advantage for companies to list in the stock market is that they can raise lots of money, so they have a great strong incentive.
According to the "Company Law" and other related securities regulations, companies must be consecutively profitable in the past three years, and have outstanding business performance in order to apply for listing. For achieving the purpose of publicly listed, some companies gloss over the financial statements through earnings management, and legally "cheat" for the qualification of going public.
b. to keep the eligibility of allotment.
The ability to obtain eligible for allotment is very critical for listed companies, which will affect the subsequent injection of funds for them. However, the policy provides that a listed company's recent three-year average return on equity cannot be less than 10%, and its single year return on equity cannot be less than 6%, only so it can get the eligibility of allotment. As a result, companies try lots of ways to take earnings management measures and adjust net return on assets in order to accomplish the purpose of allotment.
c. to avoid of being delisted.
According to related regulations, if listed company has two years losses, or less net assets per share than par value, or financial condition exception, it will be specially treated. If so, the company will face not only great damage of its image, more importantly, the company will face the danger of being delisted. In order to avoid this from happening, there is a strong motivation to manage earnings for the listed companies.
I.2.2. Methods of earnings management
Second, the methods of earnings management
a. the use of the accounting policies.
Accounting policy refers to the principles and specific treatments adopted by company while accounting. Since the flexibility of choices of accounting policy, company can adjust their accounting policy along with their own need. However, the frequent adjustment in accounting policy will on the one hand undermine the authenticity of corporate financial statements, on the other hand it will increase the difficulty for investors to understand business conditions. Therefore, it opens the door of "earnings manipulation" to the corporate management.
1) to change accounting method
2) to use the provision of assets depreciation preparation
3) to use capitalized interest to adjust profits
b. the use of related party transactions.
Related party transactions refer to the transactions carried out between a company and its subsidiaries or another company having an interest, directly or indirectly, or a stake with it. This kind of transactions has lower costs and maximized profits, etc., which are prevalent among China's listed companies, the majority is concentrated between listed company and its parent company or other subsidiaries under the parent company.
1) related party purchase and sale
2) trusteeship management 3) pass cost on
4) payment for the use of state funds 5) capital lease
c. the restructuring of assets.
In China, as the law and financial norms for company's acquisition are not yet perfect, additionally local government, state-owned assets management department get participated indirectly. Therefore, a listed company often progresses a series of asset restructuring within its corporation or its subsidiaries, in order to improve performance, and transfer the profits from the secondary market etc. Especially at the end of year, listed companies are under pressure of packaging the annual reports, and they would introduce a series of asset restructuring program, including the transfer or the exchange of assets, and transfer or the acquisition of shares, as well as a very China-specific way, the free transfer of assets from parent company. The restructuring programs having the nature of such transactions with related party become an important method of turning loss into gain for listed companies in China.
1) to eliminate the non-performing assets
2) to adjust the ratio of equity investment
I.3. Corporate governance in China
Corporate governance is defined as series of processes, customs, policies, laws, and institutions affecting the way a company is directed, administered or controlled. An important theme of corporate governance is the nature and extent of accountability of particular individuals in the organization, and mechanisms that try to reduce or eliminate the principal-agent problem.
I.3.1. Evaluation of governance quality
The evaluation of governance quality for China's listed companies First, corporate governance situation of listed companies
a. evaluation by sectors
According to the national industry classification standards, we classify listed companies in 13 industry categories in which the manufacturing industry has 10 categories, and analyze their governance situation.
It's normal to have some differences between industries on corporate governance.
The financial and insurance industry gets the highest average on corporate governance index. Non-financial sectors, the average index of transportation and storage industry is higher than others, the synthetical sector has the lowest value.
b. evaluation by nature of shareholders
We classify listed companies in term of the nature of the largest shareholder into different types. The result shows that, state-controlled companies have the highest governance index; the index of employee stock ownership (ESOP) is relatively lower.
Possible reasons for that State-controlled listed companies have better quality of governance than private companies are as follow:
On the one hand, state-owned listed companies with poor quality of governance in the long-term will gradually be forced to delist or transfer its control power. On the other hand, we found that the industries with relatively better governance belong to the monopoly industries or the dominant state-controlled industries, which are ones relating to "national economic lifeline".
Because of the importance of issue "national economy" for the government, it makes state-controlled companies focused by many supervisor institutions, and makes them pay more attention to the "compliance" of corporate governance.
Private companies are mainly in the manufacturing industry with a high marketing degree, as strong competition can play some alternative role in the governance, therefore, private companies pay more attention to the "efficiency" and "efficiency" as opposed to state-controlled companies.
The overall governance quality of private companies is improving year by year.
c. evaluation by regions
The evaluation reports that the average governance index is higher in the economically developed provinces such as Guangdong, Shanghai, Beijing, most of which lie in the east coast; and it's lower in less-developed regions such as Ningxia Hui Autonomous Region, Qinghai Province, Tibet Autonomous Region; of which Qinghai Province, Tibet Autonomous Region have the least value. Over general analysis, corporate governance in China's listed differs between regions.
I.3.2. Governance risks
Second, the status of corporate governance reveals governance risk
The corporate governance risks for China's companies include both internal and external risks.
a. internal risks
a) the irrationality of ownership structure
The cause for corporate governance risk, its underlying reason is the form of organization itself. The ownership structure, seen as the degree of centralization or dispersion, is the primary cause of corporate governance risk.
High degree of dispersion of the ownership tends to have moral hazard for administration. so that the administration can't act in accordance with the principle of maximizing shareholders' value.
Higher degree of concentration tends to have the risk of damage to the interests of minority shareholders. b) the irrationality of governance structure 1) general meeting of shareholders becomes large shareholders' meeting
As the dominance characteristic of ownership structure make general meeting of shareholders often become large shareholders' meeting, therefore it is difficult to protect the interests of small shareholders. Large shareholders would harm the interests of small shareholders through general meeting of shareholders.
2) too strong association between board of directors and managers
Currently, most of China's companies, the board of directors is largely composed of managers. Thus, it is difficult to form a supervisory role. On the contrary in U.S., the solution is the establishment of independent director system. Recent years, this system is introduced in China, but in reality, it doesn't play its due role. Independent directors are elected by large shareholders, they have inextricable links with large shareholders, and play a dressing role, not the supervisory role.
3) useless of board of supervisors
In reality the role the board of supervisors is very limited. First of all, it's about the design of our supervisory system. Company Law requires that, the board of supervisors only has the right to propose the recall, but no power of appointment, which makes board of supervisors at a disadvantage relative to the board of directors, the right balance is broken; Second, most employee supervisors are "promoted", so their independence can't be ensured.
b. external environment risks
a) lack of professional managers The general managers of China's joint-stock state-owned company, regardless of the ways of their appointment and remuneration level, they are more or less seen as
cadres and their position is essentially "iron rice bowl", are unlikely replaced by potential competitors. Thus it cannot form an effective proxy contest, the managers' moral hazard is difficult to implement an effective constraint.
b) the legal and regulatory systems are not perfect
China has already formulated the "Corporate Governance Guidelines", making a clear direction and goals for necessary improvements in corporate governance, but there is number of listed companies not doing like that. Moreover, there are still many loopholes in other relevant laws and regulations, some phenomena occur, for example laws are not fully observed or not strictly in enforced.
c) the monitoring mechanism by debtors is not in place
Under normal circumstances, the governance by creditors to company is mainly achieved through contractual provisions to encourage and restrict certain actions. When the business failed, unable to discharge the debt contract, the creditor may take measures to debtors in accordance with contract or relevant laws and regulations, such as liquidation or restructuring. But now, many of state-controlled listed companies with high debt ratio, they don't feel the pressure from banks and other financial institutions.
I.3.3. Countermeasures
Third, the countermeasures for the status of corporate governance
a. to disperse ownership structure.
Dispersion of the ownership structure is the urgent need to take an important measure, so that shareholders can appoint competent persons to carry out valuable work to create more benefits for the company.
b. to strengthen internal supervision system.
Owners and managers make arrangement on the company's responsibilities, rights, benefits, to reach a series of contracts through bargaining. Due to the incompleteness of business contract, the information asymmetry of contractors and the incompatibility of incentive mechanism, operators' self-interest and moral hazard of opportunism, above all for which the separation of ownership and operation will have a serious agency problems, and thus causes high operating costs and low efficiency for the company. This requires the establishment of a comprehensive incentive and restraint mechanisms for modern company during operation process to ensure that the interests of owners and managers can be fully realized.
c. to strengthen the agency's supervision, and strengthen the information disclosure
system.
To strengthen the information disclosure system, and require the scope, form, content, frequency of information disclosure. Meanwhile, the strengthening the supervisory role of intermediaries (accounting firms, auditing firms, law firms, news media, etc.) on information disclosure, enhance corporate transparency, prevent internal operations.
d. to actively train and develop manager market.
The manager market can establish a framework of scientific selection and evaluation system for companies, form efficient competition mechanism of manager market. Past actions of managers will be reflected over the manager market, and this information will affect managers' future career prospects. Therefore, managers will naturally restrain themselves over pursuing their own interests.
e. to improve governance of creditors.
In order to effectively prevent the company from financial difficulties, as the company's creditors, they should follow the principles of whole-process monitoring, and reconstruct creditors' meeting and strengthen creditors' right to information.
After knowing about the situation of refinancing and other related factors such as earnings management accounting for the momentum of a successful SEO and corporate governance accounting for the potential risk of SEO, we can tell a general idea that, under the background of low qualified and less developed corporate governance in China, it exists great possibility for public firms to conduct earnings management over the refinancing period. The following analysis will reveal the facts of China's capital market whether our hesitation exists. Before that, we will talk about the sample selection and our analysis methods.
II. Sample and Methods
II.1. Sample selection
The sample consists of all seasoned equity offerings by listed Chinese firms from 2000 to 2009 from the Resset financial database. To be included, seasoned issues must meet the following criteria:
1) the company is listed on Shanghai Stock Exchange (SSE);
2) the company must have completed corporate governance information regarding to the SEO during the sample period;
3) for analyzing the trend of earnings management of the company, we exclude the first two and last two years sample issues;
4) if an issuing firm has multiple SEO issues in the sample period, we use the first issue observation to avoid influence of previous SEO.
After applying these criteria, we obtain a sample of 102 seasoned issues, and the total issue size achieves 194.94 billion yuan.
For each seasoned issue, we include another 4 year-observations, 2 years prior to the issue year and 2 years next to the issue year in order to analyze whether firms manipulate earnings prior to the seasoned issue and the post-issue performance, therefore our final sample consists of 510 observations in a five-year period.
Table 3 panel A provides the time distribution of issue-year observations. More than 70% seasoned issues occur after 2005, because since then there is a strong bullish market rally in Shanghai Stock Exchange (SSE) until the crisis in early 2008, which could be the reason why issues soar in 2006 and 2007. Figure 5 points out the Shanghai composite index history, and Figure 6 provides distribution of seasoned issue proceeds by year during our sample period. We can find the similar trend for them at the corresponding timing.
Table 3 panel B provides the subtotal of issuers' share type, only 4.90% sample firms are listed on other markets as well, therefore for simplifying the analysis we ignore the effect of cross-listing on A-share seasoned issue.
Table 3 panel C provides issuers' industry distribution. As shown, the manufacturing industry has a large proportion of 65.69% of issues comparing to other industries. Due to maintain the production and cash flow, manufacturing industry has great demand for capital more than other industries. Figure 7 shows the proportion of seasoned issue proceeds by different industries.
Table 3: Summary of SEO issues
Panel A: SEO issues by years: 2000-2009
Our sample consists of all secondary equity offerings by Chinese firms from 2000 to 2009 satisfying the following: 1) the company is listed on Shanghai Stock Exchange market; 2) the company must have completed corporate governance information regarding to the SEO during sample period; 3) for analyzing the trend of earnings management of the company, we've exclude the issues before year 2002 and that after year 2007; 4) if an issuing firm has multiple SEO issues in our sample period, we use the first issue observation to avoid the influence of the previous SEO issue. Our final sample contains 102 issue-year observations.
Issue year # of firms % of firms in the sample
Most firms in our sample are listed only on the A-share market, which is priced and traded on Chinese yuan; B-share market is priced on Chinese yuan as well but traded on US dollar, this is the particular market for foreign investors to trade Chinese firms' shares; H-share is related to the shares of Chinese firms listed on the Hong Kong stock exchange market, which is priced and traded on Hong Kong dollar.
Share type # of firms % of firms in the sample
Figure 5: Historical data of Shanghai composite index
Figure 6: Seasoned issue proceeds by year Figure 7: Seasoned issue proceeds by industry
Page 23 of 50
II.2. Measuring discretionary current accruals
Earnings management is often exposed by abnormal levels of accruals. The literature suggests that short-term accruals are more likely to cause management manipulation than long-term accruals. We calculate the unexpected discretionary current accruals (DCAs) using the modified Jones (1991) model,5 its calculation process is as followed.
Current accruals (CA) are usually defined as the changes of short-term assets and liabilities that relate to daily operations of the firm. They are calculated by the annual change in non-cash current assets deduct the annual change in operating current liabilities. Total current accruals are the sum of both discretionary and non¬discretionary current accruals.
We degenerate the current accruals on the change in sales from the previous year for all non-sample firms with the same SSE industry code, industry j. In order to
5 Studies that employ a modified Jones model for abnormal accruals include Teoh, Welch, and Wong (1998a, 1998b); Rangan (1998); Hribar and Collins (2002); Xie, Davidson, and DaDalt (2003); Kim and Park (2005); Crutchley, Jensen, and Marshall (2007); Johnston and Madura (2009) and Chou, Gombola, and Liu (2006).
ease the heteroskedasticity effect, we deflate each variable in the model by the book
value of total assets TAj,t-1 from the prior year and we get the formula for current accruals:
CA 1 ISales
jt jt
= Y+ Y
01
TATA TA
j,t 1 j,t 1 j,t 1
To detect the non-discretionary part, we use the estimated y0 and y1, we calculate non-discretionary current accruals for each sample firm. The non-discretionary current accruals are the part of current accruals caused by firm's sales growth and are "viewed as independent of managerial control" (Teoh, Welch, and Wong, 1998a).
We estimate the non-discretionary current accruals for firm i at time t, NDCAit, as
_Y1
NDCAit = Y0 TAå·¥ +
i,t1
_
1
1Salesitå·¥1 ARit
TAi,tå·¥1
AR represents the trade receivables. We define the discretionary current accruals, DCAit, as the remaining part of the current accruals:
DCAit = CAit å« NDCAit TAi,tå«1
Therefore, positive (negative) DCAs imply income increasing (decreasing) earnings management.
II.3. Measuring the seasoned issue proceeds
Due to the issue price is not comparable factor to measure which issue is more successful, therefore we use the seasoned issue proceeds as the yardstick. As the larger size of seasoned issue, the stronger motivation for issuer to manipulate the earnings prior the issue year, so we try to established kind of relation between these two factors and verify whether this kind of relation is significant and reliable.
If it does exist this close relation, what kind of firms has it? State-holding ones, or 100% private firms? If we go further, are the firm size and the issue size the important elements for this relation?
II.4. Other control variables
In our multiple regressions, we control for both firm-and SEO-specific variables. We include firm size, as measured by book value over total assets, since larger firms tend to be less risky and can realize a relative large size of SEO. We use the accounting-based ratios of debt-to-assets (TotalLev) and return on assets (ROA) to proxy for firms' default risk, kind of credit level. In addition, we include their capital intensity (Collateral) to account for differences in firms' asset structure; firms with greater capital intensity present lower risk to investors and are thus expected to be more acceptable during SEO. We also control for the market-to-book ratio (Bhojraj and Sengupta, 2003; Nikolaev and Van Lent, 2005), since seasoned issuers with lower market-to-book ratio are more likely to be welcomed by investors because of the continuous growth potential. We further control for firm's desire of manipulation of earnings by measuring the variance of the stock returns (RetVar) over the year prior to seasoned issue.
We include governance factors as well, because we also want to establish a relation among governance, seasoned issue proceeds and discretionary current accruals. If they have a straight and apparent relation, we could summarize simpler approach for investors to invest in SEO. We include number of board members and independent board members, the number of board of supervisors' members and independent supervisor members, as well as their corresponding percentage. We include ownership concentration and Z-index since they are indicators of centralized power or dispersed power. We include shareholders' type to distinguish whether the state is one of shareholders. We include tradable shares percentage to show the unrestricted holding part. At last, we also cite a governance index issued by Resset Date Company that shows us a general idea of the quality of corporate governance.
SEO-specific variables in the analysis include the price, volume and cost of SEO.
Finally, in each regression, we will eliminate unnecessary variables according to corresponding analysis. Table 4 summarizes the definitions and measurements of each variable in our study.
III. Analysis of results
III.1. Descriptive statistics
Table 5 reports that seasoned equity offerings issuers tend to be large firms, with average (median) sales of 5.79 (1.82) billion yuan and assets of 18.41 (3.28) billion yuan. However, there is a considerable variation in firm size such as each 75th percentile of firms' sales and assets is less than its average value, thus we would get a long right tail on both of these two variables. In our sample firms are profitable with an ROA of 5.90% and have a market-to-book ratio of 1.60. Firms' mean (median) total leverage ratio is 0.48 (0.48) while the long-term debt ratio is 0.10 (0.07), as we know the long-term debt seems to be a very important source of financing for firms, but a relative low long-term debt ratio for our sample firms might show us some financial operation characteristics of Chinese firms.
Table 5 also reports that our sample firms get a mean (median) variance of stock returns of 0.35 (0.07) yuan prior to the seasoned issue, and our sample firms have an average (median) fixed assets ratio over the total assets of 0.27 (0.25). There appears as well an apparent variation in variance of stock return variable like firm size.
For the governance factors, Table 5 indicates several simple variables, which are the number of board members with a mean (median) value of 12.73 (12), the number of independent board members with mean (median) value of 4.04 (4), the percentage of independent board members with mean (median) value of 30.54% (32.29%), the number of board of supervisor members with a mean (median) value of 5.46 (5), the number of independent board of supervisor members with a mean (median) value of
1.33 (1), the percentage of independent board of supervisor members with a mean (median) value of 22.22% (25.00%).
Ownership concentration is a key indicator that measures whether the ownership is centralized or dispersed due to the different stake of shares by shareholders; it's also an important indicator of firm's stability. In our sample firms, they tend to have a large proportion of shares held by the largest shareholder with a mean (median) percentage of 38.33% (35.95%), while the mean (median) proportion of top 5 shareholders is 53.30% (51.51%) and part of top 10 shareholders of 59.23% (57.93%). Z-index shows us discrepancy of the power between the largest shareholder and the second largest shareholder; the larger the value is, the more the power differs. In our case, firms appear to have a large difference of power between the largest two shareholders, however there is also a considerable variation in this variable that its representativeness would be questioned.
As we mentioned before, whether shares being held by state is an influential factor to firms' seasoned equity offerings, this is an important question we need to find out. Our sample firms have an average (median) state-holding share of 26.20% (26.87%). Also our sample firms have an average (median) legal person share-holding of 23.45% (16.38%), and tradable shares proportion of 87.15% (100.00%).
G-index is a kind of governance index introduced by Resset Data Company that measures the ability of corporate governance. Our sample firms own a mean (median) index value of 27.21% (30.85%).
Regarding to the SEO characteristics, the average (median) issue price is 12.21
(8.79) yuan, the mean (median) issue volume is 219.02 (56.05) million shares, the means (median) seasoned issue proceeds is 1.91 (0.55) billion yuan, the mean (median) cost is 21.52 (14.16) million yuan. The variables above except issue price, each of them differs too much so that they have the long right tail.
We report the descriptive statistics on DCAs in Table 5. As shown in Table 5, the mean value of DCA as a percentage of lagged total assets in year -1 is 2.02%, with variation ranging from -0.11% at the 25th percentile to 3.94% at the 75th percentile.
Table 5: Descriptive statistics
Our sample contains 102 firm-year observations. Sales is the net sales in million yuan; Assets is the total book value of assets in million yuan; ROA is calculated as net income over total assets; TotalLev is total book debt over total assets; LtLev is the long-term debt over total assets; M_B is the market value of assets divided by the book value of assets; RetVar is the variance of stock returns over one year prior to SEO issuance; Collateral is fixed assets over total assets. Dirnum is the number of the board members; Inddirnum is the number of independent board members; Inddirpct is the percentage of independent board members; Supnum is the number of the board of supervisors' members; Indsupnum is the number of independent supervisor members; Indsuppct is the percentage of independent supervisor members. Owncon1 is the ownership concentration of the largest shareholder's holdings over total shares; Owncon5 is the ownership concentration of the sum of top 5 shareholders' holdings over total shares; Owncon10 is the ownership concentration of the sum of top 10 shareholders' holdings over total shares; Zindex is calculated as the largest shareholder's holdings over the 2nd largest shareholder's; Stateshrpct is the percentage of shares held by the state; Lpshrpct is the percentage of shares held by legal person; Tradshrpct is the percentage of tradable shares over total shares; G-index is a governance index established by Resset Data Company, which could generally show
us an idea of the quality of corporate governance. All firm characteristic variables are measured at the fiscal year-end of the issue year. IssuPr refers to the issue price in yuan; IssueVol is the issue volume in million shares; TotIssuMv is the total market value of SEO in million yuan; IssuCost is the cost of issue activity in million yuan; SiProceeds is the proceeds of this seasoned issue in million yuan; SiNetProceeds is the net proceeds in million yuan calculated as the proceeds deduct the issue cost. DCA is the discretionary current accruals at the fiscal year-end before seasoned issue. DCA is caculated following Teoh, Welch, and Wong (1998a) and is reported as a percentage of lagged total assets.
25th 75thStandard
III.2. DCAs around seasoned equity offerings
We report the intertemporal pattern of DCAs around seasoned equity offerings (Year 0) in Table 6. For the entire sample, we find no evidence of abnormal use of accruals two years before seasoned issue (Year -2) because all values are negative. But at the end of the year prior to the seasoned issue (Year -1), DCAs become positive and the change in DCA from Year -2 to Year -1 is highly significant at the 1% level (t=6.51, p<0.001). Subsequent to the seasoned issue, the level of DCAs falls sharply in the year right after the issue year (Year 1).
The pattern continues to exist in subsample partitioned by state-holding. The increase in DCAs immediately prior to seasoned issue and the subsequent mean reversion toward zero is consistent with issuers engaging in pre-issue income-increasing earnings management activities.
The magnitude of DCAs around seasoned issue (2.02% in Year -1 and 2.76% in Year 0) is smaller than corresponding findings prior to seasoned equity offerings (5.37% in Teoh, Welch, and Wong, 1998b) and during the IPO issue year (9.95% in Teoh, Welch, and Wong, 1998a). As we could see, the degree of earnings management is likely to depend on the different types of issuing securities and the capital markets where they are issued.
Table 6: Discretionary current accruals around SEO
Year 0 is the year of seasoned equity offerings. Discretionary current accruals (DCAs) are calculated using the modified Jones (1991) model. Stateshare are the shares owned by the state, 0 means no shares held by state. T-statistics are in parenthesis.
Significance at the 10%, 5% and 1% levels is indicated by *, ** and ***, respectively.
Page 32 of 50
III.3. Seasoned issue proceeds and DCAs
Our results suggest that there is income-increasing earnings management prior to SEO. The next and important question is to determine whether investors fully understand that earnings have been inflated and whether they price shares accordingly, additionally state owned enterprises share a great status in China. To address this question, we partition the full sample of firms into several parts according to our supposes such as the presence of state-holding shares, firm size and seasoned issue size.
We estimate a regression model with seasoned issue proceeds as the dependent variable and DCAs in Year -1 as an independent variable. If a positive relation between DCAs and seasoned issue proceeds occurs, this result would suggest that the proceeds of SEO is higher as earnings are inflated and that the investors are unable to distinguish between firms with good earnings quality and those where earnings quality is poor. In contrast, if we find no relation between seasoned issue proceeds and DCAs, this would indicate that investors recognize and include the positive accruals in their assessment of the firm, and there would be no effect on the proceeds of SEO.
Following Anderson, Mansi, and Reeb (2004), we control for pre-issue firm-specific and issue-specific characteristics, including firm size (log assets), profitability (ROA), financial leverage (TotalLev), growth opportunities (M_B), firm risk (RetVar), and property, plant, and equipement to total assets (Collateral) (Ashbaugh, Collins, and LaFond, 2006) because they are all likely to affect the proceeds scale.
More simply, the total SEO proceeds are decided by both issue price and volume, beside the cost of SEO would affect the total proceeds if the issue size differs.
A good quality of corporate governance gives a positive image for public investors, which would enhance the acceptance level to the SEO and achieve a higher proceeds. Also we try to establish an easier connect between governance factors, seasoned issue proceeds, and discretionary current accruals, therefore we include governance-specific variables. However, whether we could find significant relation among them is still the question, if we fail to find any evidence between governance factors and other elements, for common investors, they have to regress to the technical and professional financial analysis that would not be an effortless thing.
Table 7 contains the regression results. For the entire sample (model 1), the
estimated coefficient for DCA is not significant, which means the pre-issue earnings manipulation by issuing firms will not affect the seasoned issue proceeds at all.
We also find that higher SEO proceeds are associated with larger firm size (t=5.40, p<0.001), higher firm risk (t=2.94, p<0.05), higher issue price (t=2.42, p<0.05), larger issue volume (t=5.28, p<0.001) and lower ownership concentration of top 5 (t=-1.73, p<0.1).
To examine whether the results are consistent within sub-samples partitioned by state-holding shares, we exclud
