Docstoc

How far would management go to manage earnings

Document Sample
How far would management go to manage earnings Powered By Docstoc
					               How Far Would Management Go to Manage Earnings?
                          ---The Evidence from China




                           Guohua Jiang Liyan Wang

                      The Guanghua School of Management
                              Peking University
                               Beijing 100871
                                    China




                                 December 2003




                     Very preliminary, comme nts welcome.




The authors are assistant professor and professor, respectively, of the Guanghua
School of Management, Peking University, Beijing 100871, China. We thank Wenwen
Chen, Chen Li and Jingqi Lu for excellent research assistance. Please send comments
to gjiang@gsm.pku.edu.cn and lywang@gsm.pku.edu.cn.




                                         1
I. Introduction

    Earnings management has been one of the most widely studied corporate

behaviors in accounting literature. Basing on this large body of research, accounting

scholars have largely agreed that company management manages corporate earnings

to meet certain targets. The incentives for earnings management include bonus

contracts, debt covenants, initial public offerings and seasonal public offerings, etc. In

terms of the means of earnings management, although accounting scholars recognize

the possibility that management may manipulate real activities (company operation)

to meet earnings targets, the vast majority of studies identify accrued earnings as the

mean through which earnings management is achieved. One recent exception is

Roychowdhury (2003).

    Earnings management (EM hereinafter) leads to misleading accounting

information being communicated to information users. Various groups of capital

market participants may suffer as a result, such as shareholders (bonus-driven EM),

creditors (debt covenant-driven EM), and potential shareholders (IPO-driven EM).

However, other than under its extreme form (accounting frauds or crimes),

accrual-based EM does not seem to do serious harm to its victims. The reason is that

the discretion on accruals is bounded by GAAP (Barton and Simko, 2002). Given the

underlying economic transactions of a firm, the management’s ability to report

accrued earnings is limited. Overtime, accruals will reverse as well, so management

has to consider the implications of their discretion on current accruals for future

earnings. This consideration in turns constraints accrual-based EM as well.


                                            2
    What if management can manipulate real activities (company operation) to

manage earnings? That is--- will managers go such a long distance to achieve their

EM targets? Will in this case myopic EM behaviors do significant harm to its

―victims‖? Surveying the EM studies with U.S. firms as samples, we find little

research on these two questions. The reason is simple. The United States has one of

the best-developed capital markets in the world. The complexity of its institutions

guarantees that the market participants, including company management, do not

engage in self-serving activities that is beyond a certain acceptable limit.

    However, what if the limit is loose? Will managers engage in more real

activity-based EM in this case? Our paper answers this question with data from

China’s stock market.

    China opened its current stocks markets at Shanghai and Shenzhen in early

1990’s. By September of 2003, there are more than 1,200 company stocks traded on

the two exchanges, with a total market capitalization of 1,266 billion RMB (153

billion US dollars).

    In almost all aspects, China’s stock market is an emerging market. One

significant difference between China’s stock market and other world markets is in the

ownership structure of listed companies. When China embarked on economic reform

in the late 1970’s, its main purpose was to improve the operating efficiency of

state-owned enterprises (SOEs). The introduction of stock market was one step in this

reform. Therefore, China’s government involvement in the development of stock

market is instrumental and the government’s top priority was to list SOEs on the new


                                            3
stock markets.

    After a SOE went public, the government usually retained a controlling interest in

the listed company. Moreover, the shares retained by the government are not traded.

This practice in fact divides shareholders into two groups: one group with un-traded

shares who in most cases controls the listed company, and the other group with traded

shares who lacks control over the company.

    Government regulation of the stock market is very strict. As in an emerging

market, the right to get listed on stock exchanges is a scarce commodity. China ’s

government actually sets a very high standard and even a Quota System to list

companies. For example, the Quota System limits the book value of equity of the

tradable shares to a certain amount. As a result, most SOEs that would like to get

listed can’t actually list their whole company. In addition, the SOE itself may not be

able to meet the past profitability requirement of listing. To list at least a part of the

SOE, the SOEs usually carve out the best parts of the corporation to form a new

subsidiary and then apply to list the subsidiary. Therefore, most listed China ’s

companies are actually subsidiaries of larger SOEs. SOEs hold controlling interests in

the listed companies, and government in turn controls SOEs. Central and local

governments also directly own listed companies through their Bureau of Asset

Management. As a result, Chinese government, through its various levels and

different branches, still retain controlling interests in most listed companies. And

listed companies, though nominally independent, face government interference in

their internal operation.


                                              4
    Another feature of China’s stock market is that most management regards equity

capital raised from the stock market as cost- free, and rightly so. Because parent SOEs

and/or various government agencies still own the majority of the common shares, and

due to the lack of protection of minority shareholder rights, company management is

still only responsible to the parent SOEs and government, brushing aside the rights of

minority shareholders. In fact, equity capital raised on stock market is a substitute for

bank loans. In the case of bank loan, companies will have to answer to banks, but the

companies do not have to answer to minority shareholders at all.

    Given these features of China’s stock market, there exist two strong incentives for

companies to manage earnings.

    First, it is difficulty to get listed on the stock exchanges, and it is also easy to get

delisted from the exchanges. China’s SEC sets specific requirements for companies to

stay listed. Once any of these requirements are not met, companies would be put into

probation, and if the companies do not satisfactorily improve their performance, they

will be delisted at the end of the probation period. Section two of this paper provides a

chronicle of these requirements. Delisting is a big deal on any stock markets, but it is

a bigger deal for China’s listed companies. As discussed earlier, these listed

companies went through a very tough process to get listed, once they got listed, they

normally achieved star status in their respective local areas. Listing not only helps

these companies raise capital, which is basically cost-free in contrast to other

financing channels such as bank loans, it is also been regarded as a performance

measure of the local government heads. Therefore, companies, once facing the


                                              5
possibility of delisting from the stock exchanges, have stronger than usual incentive to

find ways to meet requirements to stay listed. The SOEs behind the listed companies,

and the government agencies behind SOEs, also have strong incentive to support the

listed companies (Chen, Lee and Li, 2003). 1

      Second, most listed companies also raise additional capital from the stock market

through rights offerings. China’s SEC places even stronger constraints on the

privilege to do rights offering. A company must meet return-on-equity (ROE)

thresholds in the years prior to a rights offering. Section two of this paper provides a

chronicle of these requirements. Given that management regards equity capital raised

on stock market as cost- free, it is in their interest to pursue rights offering even that

means they have to manage company earnings. Plus, there is evidence that the parent

SOEs encourage and help listed companies to raise more capital and then tunnel the

money into the parent companies (Jian and Wong, 2003).

      Compared to the incentives that prompt U.S. managers to engage in earnings

management (bond contracts, debt covenants, IPO and SEO, etc.), the incentives of

China’s company management to avoid delisting from the stock exchanges and to

raise more capital through rights offerings are much stronger. If accrual-based EM is



1
    On August 13, 2003, Sohu business news reports a story. Zong Heng Corporation of Jian Su province had two
consecutive loss years in 2001 and 2002. The stock exchange put the company stock under Special Treatment (*ST,
i.e., probation). If the company does not make profits in 2003, the stock will be delisted. M oreover, the semiannual
financial report to be released is going to report losses and predict a loss of the third quarter. To avoid delisting of
Zong Heng stock, it is reported that Jian Su gov ernment gets involved in restructuring of the company in the hope
to turn profit in 2003. One company insider was quoted as saying: ―although time is short, if the
(government-backed) restructuring moves fast, it is entirely possible that we have a profit year and avoid
delisting.‖


                                                            6
not enough to meet respective qualification requirements, we conjecture that with the

backing of SOEs and governments, China’s listed companies may engage in radical

form of EM such as managing real activities of the company to meet these

requirements. There is no guarantee that this type of EM will benefit shareholders. It

only keeps the companies listed or helps them loot more capital, but potentially could

have damaged the company in the long run.

    It is the purpose of this paper to examine real activity-based EM in China’s listed

companies.

    This paper adds to the EM literature in the following ways. First, we identify two

EM incentives that have been less commonly studies before, managerial incentive to

avoid delisting from stock exchanges and to raise more capitals. Second, we identify

some real activities-based EM techniques in addition to conventional accrue-based

EM techniques, namely asset sales/purchase/exchanges and equity sales/purchase.

Third and most importantly, our paper shows that company ma nagement could go as

far as altering real activities to meet earnings targets. The evidence of EM in

accounting literature has been abundant, but how far would management go to do EM

is not clear. This paper provides evidence that management would go further than

accrual management to manage earnings.

    A few recent papers also study earnings management in China’s stock market.

Jian and Wong (2003) find that listed companies manage earnings through

related-party sales, and parent SOEs of listed companies tunnels resources into or

away from the listed companies to extract benefits from the listed companies. Liu and


                                            7
Lu (2002) find that corporate governance is an important factor in deciding whether

companies manage earnings. Weak corporate governance companies with a large

controlling shareholder tend to manage earnings to exceed regulatory thresholds and

at times tunnel resource away from the listed companies. Chen, Lee and Li (2003)

find that local government uses its tax and fiscal policies to help listed co mpanies

from its jurisdiction manage earnings to circumvent central government regulatory

requirements.

    Our study differs from the three papers in that we look at more dramatic forms of

earnings management. That is, we look at real activities that in essence change the

operation of the listed companies. Our study presents a striking phenomenon of

earnings management at high costs.




II. A chronicle of CSRC requirements on stock listing and rights offering

    In this section, we provide a chronicle of CSRC req uirements on stock listings

and rights offering. During the short history of Chinese stock markets, regulation has

to change frequently to keep pace with the fast development of the markets.

    In terms of delisting policy, the Chinese Company Act of 1993 ma ndates that if a

listed company suffers three consecutive loss years, its stock will be suspended from

trading on the exchanges and put into probation. If the company does not make

satisfactory improvement during the probation period, its stock will be delisted. From

February 2001, the probation period is one year and during the probation period, the

company stock is up under Particular Transfer (PT) status. However, after January 1,


                                            8
2002, CRSC removed PT status and installed Special Transfer (ST) status instead.

CSRC also shortened the probation period to half- a-year. It is clear that CRSC

monitors closely those firms who do not show profitability in consecutive years and

deal with them with the punishment of delising.

    In terms of rights offering, in 1993, CSRC requires two years of profits for

companies to be eligible for rights offering. In September 1994, CSRC stops to

require profits only, but require the average return on equity (ROE) of the previous

three years not below 10 per cent. In January 1996, CSRC further tightens the

requirements to require ROE not below 10 per cent in each of the previous three years.

In March 1999, the requirements change to average ROE not below 10 per cent AND

ROE not below 6 per cent in EACH of the previous three years. In March 2001,

CSCR requires weighed-average ROE not below 6 per cent in the previous three

years.

    The CSRC requirements create two critical thresholds for listed companies, one is

zero profit and the other is 6 per cent or 10 per cent ROE. If possible, a co mpany will

prefer to report small profit instead of small loss to avoid getting into a delising trap,

or to report a ROE above 6 per cent or 10 per cent instead of slightly fall short in case

the company needs to raise more capital from stock market in the next few years.

Prior studies have documented out-of-proportion number of firms reporting small

profits, or reporting ROE slightly higher than 6 per cent or 10 per cent, depending on

which year ’s data was used.

    Prior literature has ably documented that once a company is going to miss all


                                             9
meaningful earnings thresholds (zero profit, analysts’ forecast, earnings of the same

period in the previous year), management tends to ―take a big bath‖ to earnings

(Abarbanell and Lehavy, 2001). A ―big bath‖ to current earnings creates earnings

reserve for future years.

    Therefore, in this paper, we identify three scenarios where earnings management

is suspected: big loss firms whose ROE fall below negative 10 per cent (big loss

group), small profit firms whose ROE fall between zero per cent and 2 per cent (small

profit group), and rights offering firms whose ROE fall between 6 per cent and 8 per

cent (rights offering group). All other firms are put into one group: other firms group.




III. Identifying real activity-based earnings management techniques

    We identify five real activities that we suspect were used to manage earnings,

equity sales, equity purchase, asset sales, asset purchases, and asset exchanges. For

the sake of this paper, we call these activities earnings management activities.

    Equity sales refer to companies selling ownership in another firm, sometimes a

subsidiary. Equity purchases refer to companies buying ownership in another firm. If

these transactions are purely based on fair market prices, the room for earnings

management is limited. However, as discussed earlier, a parent SOE usually controls

the listed company, and the SOE tends to control a net of other companies.

Furthermore, The local government behind the SOE controls more companies. This

complex net of connections offer listed company opportunities to engage in equity

sales and equity purchases that generate profits. Surveying the news reports on these


                                           10
so called ―restructuring transactions‖, it at first appears that the ―restructuring

transactions‖ do not make economic sense for at least one party in the transactions,

but tracing the roots of each party, quite often, will lead to the controlling

shareholders of the listed company or local government. It is not rare that ownership

sold (purchased) by the listed companies in one year was purchased back (sold) in

another year.

      There are a few ways equity sales/purchases would affect earnings and ROE.

For equity sales, the earnings the sold equity generates from beginning of the year to

the date of sales could be recognized; for equity purchases, the earnings the purchased

equity generates would be recognized. Assuming companies have a good idea about

the earnings-generating ability of the equity, they would certainly pick subsidiary to

sell, or other company to purchase so that their earnings target is met. Also, for equity

sales, gains may be generated because selling price is higher than book value of the

sold equity. A more subtle impact of equity sales/purchases is that equity

sales/purchases may change the consolidation base of financial reporting. A subsidiary

which was consolidated in financial reporting may not need to be consolidated after

the listed company sold a certain portion of its holding in this subsidiary. After equity

purchase, the listed company may need to consolidate a new subsidiary.

    Therefore, we believe listed companies, with the backing of parent SOE and local

government, may ―strategically‖ use equity sales/purchases to manage earnings.

    Similar arguments can be made for asset sales, asset purchases and asset

exchanges. Therefore, in this paper, we examine whether Chinese listed companies


                                             11
use these five activities to manage earnings.




IV. Empirical results

   We hand-collect earnings management activities for 704 listed compa nies on the

Shanghai Stock Exchange which had announced their 2002 annual results before

April 30, 2003. Seven companies with negative ROE are removed from the sample.

Our final sample consists of 698 companies.

   We first plot a histogram of ROE of these 698 companies. It is shown in figure 1.

Figure 1 plot the number of firms whose ROE fall into each one-per cent interval.

Firms whose ROE fall below negative 10 per cent are put into one group, so are firms

whose ROE are above 20 per cent.

   As documented by earlier studies, it is apparent from figure 1 that there is an

out-of-proportion clustering of companies in the ROE intervals (0, 1 ) and (6, 7). There

are very few companies with ROR between negative 10 per cent and zero percent, but

there are 46 companies with ROE between zero percent and one per cent—that is,

these firms managed to report a small profit. There are 106 companies with ROE

between 6 per cent and 7 per cent—that is, they slightly meet the CSRC rights

offering requirement on ROE. In contrast, the two adjacent intervals contain much

fewer companies, with 36 companies in the interval (5, 6) and 60 companies in the

interval (7, 8). We believe that the out-of-proportion clustering of companies in the

ROE intervals (0, 1) and (6, 7) indicates possible earnings management by these

firms.


                                           12
   Figure 2 plots number of earnings management activities by ROE intervals.

Similar to figure 1, firms in the ROE interval (0, 1) report 49 cases of earnings

management activities, while firms in with ROE between negative 10 per cent and

zero percent reports only a few. Firms in the ROE interval (6, 7) reports 170 such

cases, while those in the ROE intervals (5, 6) and (7, 8) only report 51 cases and 78

cases, respectively.

   A more meaningful statistic is the average number of earnings management cases

in each ROE interval. Figure 3 shows this. Apparently, in terms of average number of

earnings management activities, ROE intervals (0, 1) and (6, 7) do not exhibit a

higher frequency than their adjacent intervals.




   Table 1 gives a detailed analysis of earnings management activities. We show in

this table number and average number of earnings management activities by group,

and number and percentage of firms in each group reporting earnings management

activities by zero time, one-to-three times and more than three times. We use other

firms group as a benchmark to analyze whether big loss group, small profit group, and

rights offering group report higher frequency of earnings management activities.

   Out of 698 firms, 58 are big loss firms, 88 are small profit firms, 168 are rights

offering firms, and 384 are other firms. From the top panel of table 1, these 698 firms

in year 2002 reported 951 cases of earnings management activities, 1.36 cases per

firm on average. Only rights offering group reported a higher average number of

earnings management activities than the benchmark other firms group (1.48 versus


                                           13
1.34), while small profit group actually reported a smaller average (1.25). In terms

number and percentage of firms in each group reporting earnings management

activities by zero time, one-to-three times, and more than three times, there is still no

clear difference between big loss group, small profit group, rights offering group and

other firms group.

    The other panels of table 1 report similar statistics for each individual earnings

management activity. Once again, significant differences between groups are not

evident. However, it appears that small profit group and rights offering group report

higher average number of earnings management activities in December, the last

month of a fiscal year. Small profit group reports 0.4 cases per firm, rights offering

group reports 0.45 cases, while other firms only reports 0.29 cases. This is a

preliminary indication that small profit firms and rights offering firms, trying to meet

ROE threshold, engage in more earnings management activities right before fiscal

year end.

    In table 2, we test whether there are significant differences in the percentage of

firms reporting at least one earnings management activities. The lower panel reports

the differences between big loss group, small profit group, rights offering group, and

other firms group. We use a K-squared test to test whether the differences are

statistically significant.

    Statistically significant differences are few. Small profit group reports

significantly higher percentage of firms reporting asset sales (by 6 per cent), higher

percentage of asset exchanges (by 7 per cent), but lower percentage of asset purchase


                                            14
(by 8 per cent). Big loss group reports a significantly lower percentage of firms

reporting equity purchase (by 18 per cent).

   Overall, table 1 and table 2 indicate that contrary to our expectation, big loss firms,

small profit firms, and rights offerings firms do not report earnings management

activities more often than other firms.

   One possibility to explain the lack of differences in table 1 and table 2 is that

number of earnings management activities may not be as important as the amounts of

earnings these transactions generate if firms were to use these activities to manage

earnings—for example, to turn a loss year into a profit year, a small profit firms might

have used only one earnings management activity which generates a larger amount of

earnings than more activities of another company combined.




   To explore this possibility, we hand-collect the impact of earnings management

activities on earnings. A better approach is to examine the impact on ROE. However,

firms do not report the impact of these activities on equity, so we could not determine

the impact on ROE.

   Firms usually report the following items in their annual reports: the contributions

to current year earnings of sold assets or sold equity; the contributions to current year

earnings of purchased assets or purchased equity; and the gains on asset sales or

equity sales. So we separate the contributions to current year earnings into these three

types. Table 3 reports the results. The numbers reported are contributions to current

year earnings standardized by total equity.


                                              15
    Other than in one case, we can see earnings management activities contribute

larger amounts to earnings of big loss firms, small profit firms, rights offering firms

than to other firms. The only case where other firms group report a larger contribution

from equity/asset sales than small profit group is due to one extreme observation.

Getting rid of this extreme observation small profit group would report a larger

contribution than other firms group by 0.027.

    However, using a t-test, all these differences in contributions to earnings are not

statistically significant.

    Nevertheless, table 3 indicates that small profit firms and rights offerings firms

might have used earnings management activities to increase their ROE above the

zero-profit, and 6 per cent ROE thresholds. Contrary to our expectation, these

earnings management activities also increase earnings for big loss group, instead of

decreasing earnings to take a big bath.




    Our final test is a probit regression analysis to detect whether big loss firms, small

profit firms, and rights offering firms report higher frequency of earnings

management activities after controlling other variables that might influence the

engagement of such activities. The dependent variable takes value one if a company

reports more than one earnings management activities, and zero otherwise. One is the

median number of earnings management activities reported by our sample firms.

    The independent variables include the following items. The variable Earnings

Management Group takes value one if a firm belongs to big loss group, small profit


                                            16
group, or rights offering group, and zero otherwise. Although table 2 indicates

earnings management group firms do not report significantly higher frequency of

earnings management activities, the probit regression will add a robustness check

after controlling other variables that might affect the frequency of these activities.

   % controlling shareholder ownership is the percentage of common shares owned

by controlling shareholders. If shareholder ownership is concentrated, the controlling

shareholders will have more leeway to manipulate the internal operation of the firms

without sufficient monitoring. We conjecture that the larger the controlling

shareholder ownership, the more likely a firm may engage in earnings management

activities. Liu and Lu (2002) documents that firms with weak corporate governance

engages in more earnings management activities, and the existence of a controlling

shareholder is frequently cited as an indication of weak corporate governance.

     % government ownership is the percentage of common shares owned by various

levels and branches of Chinese government. As Chen, Lee and Li (2003) shows,

governments are involved in helping listed companies in their own jurisdictions to

meet CSRC requirements through tax benefits, we conjecture governments may also

help companies engage in the earnings management activities identified in this paper

through their administrative power.

   Market-to-Book ration is market capitalization divided by total equity. High

market-to-book stocks are routinely regarded as glamour stocks and even small

negative news would have significant consequences for the stock prices (Sloan and

Skinner, 1999). So high market-to-book firms may tend to manage earnings to meet or


                                            17
exceed thresholds.

    Finally, we also include total assets as a control variable. Companies’ ability to use

asset sales or asset purchases depend on how much asset they control.

    Other than ROE, variables mentioned above are measured at beginning of year.

    Table 4 reports descriptive statistics of these variables. Only big loss firms show

differences from the other three groups with lower controlling shareholder ownership,

lower government ownership, and smaller amount of assets.

    Table 5 reports the probit regression results. Consistent with table 2, even after

controlling for these other variables, earnings management group still show no

deterministic power for earnings management activities. The sign of the coefficient is

actually negative. The coefficient on % controlling shareholder ownership is 0.026

and significant, indicating firms with larger controlling shareholder ownership engage

in more earnings management activities. This result is consistent with the Liu and Lu

(2002) conjecture. Contrary to our expectation, the coefficient on % government

ownership is –0.007 and significant. After controlling for the impact of controlling

shareholder ownership, more government ownership actually decreases the likelihood

of earnings management activities.

    Finally, the coefficients on market-to-book ratio and on total assets are not

significant.




    In summary, the empirical analysis does not support the notion that big loss firms,

small profit firms, and rights offering firms engage in more frequent earnings


                                            18
management activities identified by this paper, but with weak evidence, these firms

report larger amounts of income- increasing contributions to current year earnings

from earnings management activities to meet or exceed important thresholds.




V. Conclusions

    Our purpose in this paper is to test whether Chinese companies use real activities

such as asset sales/purchase/exchange and equity sales/purchase to manage earnings

to meet or exceed CSRC regulatory thresholds. Different from earnings management

through accrued earnings, real activity-based earnings management may cause more

harm to investors.

    Our results in this paper are weak. We do not observe significant difference in the

frequency of earnings management activities between big loss firms, small profit

firms, rights offering firms (firms we suspect would engage in more earnings

management) and the other firms (firms we use as benchmark). However, there is

indeed indication that these activities of big loss firms, small profit firms, rights

offering firms generate more earnings than other firms, helping companies to meet or

exceed earnings thresholds.

    We believe there are a few things we can do to enrich this paper. One is to expand

our sample to include companies on the Shenzhen Stock Exchange. Seco nd, since

CSRC delisting and rights offering requirements are both for three consecutive years,

we should further identify firms who have missed these thresholds in the past two

years. These firms should have stronger incentive to manage earnings. Third, we can


                                             19
identify firms who in fact do rights offering in the next two years. Assuming firms

decide early whether to do rights offering in the future, they would engage in earnings

management to make sure their current earnings meet CSRC requirement. We will

collect more data to further explore these possibilities.




                                            20
Reference

Abarbanell, J., and R. Lehavy. 2001. Biased forecasts and biased earnings? The role
of earnings management in explaining apparent optimism and inefficiency in analysts’
earnings forecasts. Working paper, University of California, Berkeley.

Barton, J. and P. Simko. 2002. ―The Balance Sheet as An Earnings Management
Constraint.‖ Accounting Review, Forthcoming.

Burgstahler, D. and I. Dichev. 1997. ―Earnings Management to Avoid Earnings
Decreases and Losses." Journal of Accounting and Economics, Vol. 24, No. 1
(December), pp. 99-126.

Chen, X., C. Lee, and J. Li. 2003. ―China’s Tango: Government Assisted Earnings
Management.‖ Working Paper, Tsinghua University and Hong Kong University of
Science and Technology.

Jian, M., and T. J. Wong. 2003. ―Earnings Management and Tunneling Through
Related Party Transactions: Evidence from China’s Corporate Groups.‖ Working
Paper, Hong Kong University of Science and Technology.

Liu, Q., and J. Lu. 2002. ―Earnings Management to Tunnel: Evidence from China’s
Listed Companies.‖ Working Paper, The University of Hong Kong.

Roychowdhury, S. 2003. ―Management of Earnings Through the Manipulation of
Real Activities that Affect Cash Flow From Operations‖. Working Paper, University
of Rochester.

Skinner, D. and R. Sloan. 2002. ―Earnings Surprises, Growth Expectations, and Stock
Returns or Don’t Let an Earnings Torpedo Sink Your Portfolio." Review of
Accounting Studies 7, 289-312.

郭宪英、张炳发:               ,
        《企业盈余管理的手段方法分析》《山东经济》2002 年第 3 期。

郭艳华、崔玲玲:               ,
        《试论我国企业的盈余管理行为》《山西财经大学学报》2001 年第 4 期。


潘琰、刘秋明:                ,
       《上市公司利润操纵动机比较研究》《交通财会》2000 年第 4 期。

王啸、邵寰:                    ,
      《上市公司的资产重组与盈余管理行为分析》《上海会计》2000 年第 9 期。


王跃堂:                   ,
    《我国证券市场资产重组绩效之比较分析》《财经研究》1999 年第 7 期。


邹小凡、陈雪洁:                   ,
        《我国上市公司盈余管理研究的实证发现》《浙江学刊》2001 年第 6 期。



                                         21
    Figure 1 Histogram of ROE for 698 Companies in Shanghai Stock Exchange, 2002



                               Histogram of ROE for 698 Companies in Shanghai Stock
                                                 Exchange, 2002


                         120
                         100
           Frequency




                          80
                          60
                          40
                          20
                           0
                               <=-10




                                                                               10
                                                                                    12
                                                                                         14
                                                                                              16
                                                                                                   18
                                                                                                        20
                                                           0
                                                               2
                                                                   4
                                                                       6
                                                                           8
                                       -8
                                            -6
                                                 -4
                                                      -2



                                                                    ROE




Figure 2 Histogram of Earnings Management Acti vi ties of 698 Companies in Shanghai Stock
                                             Exchange, 2002, by ROE Groups



                                Histogram of Earnings Management Activities of 698
                               Companies in Shanghai Stock Exchange, 2002, by ROE
                                                      Groups


                         200
             Frequency




                         150
                         100
                          50
                           0
                                <-10




                                                                               10
                                                                                    12
                                                                                         14
                                                                                              16
                                                                                                   18
                                                                                                        20
                                                           0
                                                               2
                                                                   4
                                                                       6
                                                                           8
                                       -8
                                            -6
                                                 -4
                                                      -2




                                                                    ROE




                                                               22
Figure 3 Histogram of Average Number of Earnings Management Acti vi ties of 698 Companies in
                                       Shanghai Stock Exchange, 2002, by ROE Groups



                               Histogram of Average Number of Earnings Management Activities
                                of 698 Companies in Shanghai Stock Exchange, 2002, by ROE
                                                         Groups


                             6. 00
         Average Frequency




                             5. 00
                             4. 00
                             3. 00
                             2. 00
                             1. 00
                             0. 00
                                     <-10




                                                                                     10
                                                                                          12
                                                                                               14
                                                                                                    16
                                                                                                         18
                                                                                                              20
                                                                0
                                                                    2
                                                                         4
                                                                             6
                                                                                 8
                                            -8
                                                 -6
                                                      -4
                                                           -2




                                                                         ROE




                                                                    23
                          Table 1 Summary of Earnings Management Acti vities


                                                                 Rights Offering
                         Big Loss Firms     Small Profit Firms         Firms        Other Firms         All Firms
# of firms                     58                   88                 168                384                698
# of All Activities            78                   110                249                514                951
Average per firm               1.34                 1.25               1.48               1.34               1.36
                                Number and Percentage of Firms Reporting Earnings M anagement Activities by
Zero times                27          47%      36          41%    75          45%   163          42%   301          43%
One to three times        24          41%      46          52%    69          41%   180          47%   319          46%
M ore than three times    7           12%      6           7%     24          14%   41           11%   78           11%
# of Equity S ales             45                   39                  81                154                319
Average per firm               0.78                 0.44               0.48               0.4                0.46
Zero times                45          78%      66          75%   135          80%   299          78%   545          78%
One to three times        9           16%      20          23%    25          15%   77           20%   131          19%
M ore than three times    4           7%       2           2%      8           5%    8           2%    22           3%
# of Equity Purchase            8                   27                  90                174                299
Average per firm               0.14                 0.31               0.54               0.45               0.43
Zero times                53          91%      70          80%   116          69%   283          74%   522          75%
One to three times        4           7%       18          20%    49          29%   95           25%   166          24%
M ore than three times    1           2%       0           0%      3           2%    6           2%    10           1%
# of Asset S ales              12                   19                  16                56                 103
Average per firm               0.21                 0.22                0.1               0.15               0.15
Zero times                49          84%      73          83%   152          90%   345          90%   619          89%
One to three times        9           16%      15          17%    16          10%   37           10%   77           11%
M ore than three times    0           0%       0           0%      0           0%    2           1%     2           0%
# of Asset Purchase             7                   13                  46                100                166
Average per firm               0.12                 0.15               0.27               0.26               0.24
Zero times                51          88%      78          89%   134          80%   309          80%   572          82%
One to three times        7           12%      10          11%    33          20%   74           19%   124          18%
M ore than three times    0           0%       0           0%      1           1%    1           0%     2           0%
# of Asset Exchange             6                   12                  16                30                 64
Average per firm               0.1                  0.14                0.1               0.08               0.09
Zero times                53          91%      76          86%   157          93%   356          93%   642          92%
One to three times        5           9%       12          14%    10           6%   28           7%    55           8%
M ore than three times    0           0%       0           0%      1           1%    0           0%     1           0%
# of December
Activities                     16                   35                  76                111                238
Average per firm               0.28                 0.4                0.45               0.29               0.34
Zero times                49          84%      67          76%   127          76%   307          80%   550          79%
One to three times        8           14%      20          23%    38          23%   76           20%   142          20%
M ore than three times    1           2%       1           1%      3           2%    1           0%     6           1%


                                                          24
# of Acti vities With
Subsidiaries                    20                 27                 85                  132                264
Average per firm                0.34               0.31               0.51                0.34               0.38
Zero times                 48          83%    73          83%   121          72%    296          77%   538          77%
One to three times         9           16%    14          16%    44          26%    83           22%   150          21%
M ore than three times     1           2%      1          1%      3          2%      5           1%    10           1%


This table summarizes earnings management activities by groups. It presents average number of each
activity by groups, and number and percentage of firms reporting zero time, one -to-three times, and
more-than-three times such activities. For 698 2002 Shanghai Stock Exchange companies who reported
a valid ROE, those with ROE less than –10% are classified into Big Loss Group, those with ROE
between 0% and 2% into Small Profit Group, and those with ROE between 6% and 8% into Rights
Offering Group. All others are classified into Other Firms Group. Equity sales and equity purchases
refer to companies selling or buying ownership of other companies; asset sales, asset purchase, and
asset exchange refer to selling, buying, and exchanging of fixed assets; December activ ities refers to
any equity activities that occur in December of calendar year (all Ch inese companies have December as
fiscal year end); activities with subsidiary refer to dealings with a subsidiary.




                                                        25
                                               Table 2 Di fferences in Reporting Earnings Management Acti vities among ROE Groups
                                                             ---Number of Firms Reporting At Least One Earnings Management Activity


                                                                                                  Earnings M anagement Activities
                                                                     Equity                                                                                December               With
                                             Equity Sales        Purchase          Asset Sales    Asset Purchase Asset Exchange Any Activities              Activities      Subsidiary
                                            Freq.   %         Freq.     %        Freq.   %        Freq.       %        Freq.   %        Freq.   %        Freq.   %        Freq.     %
                       Big Loss
                       Firms (1)              13    23%          5          9%      9     16%        7        12%         5        9%     31    53%         9     16%       10       18%
                       Small Profit Firms
                       (2)                    22    25%         18       20%       15     17%       10        11%        12    14%        52    59%        21     24%       15       17%
                       Rights Offering
                       Firms (3)              33    20%         52       31%       16     10%       34        21%        11        7%     93    55%        41     25%       47       28%


                       Other Firms (4)        85    22%         101      27%       39     11%       75        19%        28        7%     221   58%        77     20%       88       23%
                                                                         Difference in Frequency of Earnings M anagement Activities Between Groups
                       (1) – (4)                        1%              -18%                 5%               -7%                  2%           -5%                -4%               -5%
                       (2) – (4)                        2%               -7%                 6%               -8%                  7%               1%               4%              -6%
                       (3) – (4)                    -2%                     4%            -1%                     2%               0%           -3%                  5%                 5%


This table presents differences in frequency of occurrence of earnings management activities. For all 2002 Shanghai Stock Exchange companies who reported a valid ROE, those with ROE less
than –10% are classified into Big Loss Group, those with ROE between 0% and 2% into Small Profit Group, and those with ROE between 6% and 8% into Rights Offering Group. All others are
classified into Other Firms Group. This table for each group reports the number of firms and percentage of firms in respective group reporting at least one activity of equity sales, equity
purchase, asset sales, asset purchase, and asset exchanges. We use K-squared test to test the difference of the percentages between groups. Statistically significant differences are highlighted in
bold.




                                                                                                         26
           Table 3 The Impact of Earnings Management Acti vi ties on Earnings by Group


                                          Impact on Earnings of Earnings Management Activities
                                                           Equity and        Gains on
                                          Equity and           Asset         Equity or
                                          Asset Sales         Purchase      Asset Sales         Total
         Big Loss Firms (1)                  0.030             0.006           0.020            0.029
         Small Profit Firms (2)             -0.004*            0.015           0.028            0.024
         Rights Offering Firms (3)           0.050             0.004           0.020            0.027
         Other Firms (4)                     0.002             -0.013          0.014            -0.001
         (1) – (4)                           0.028             0.019           0.005            0.029
         (2) – (4)                          -0.007             0.027           0.014            0.024
         (3) – (4)                           0.047             0.017           0.006            0.028


*The magnitude of this number is mainly due to one extreme observation. After g etting rid of this
extreme observation, it is 0.0029.


This table presents the impact of earn ings management activ ities on earnings by groups. For all 2002
Shanghai Stock Exchange companies who reported a valid ROE, those with ROE less than –10% are classified
into Big Loss Group, those with ROE between 0% and 2% into Small Profit Group, and those with ROE between
6% and 8% into Rights Offering Group. All others are classified into Other Firms Group. The second column
reports contributions to earnings from beginning of the year to the date equity or asset is sold. The third column
reports the contributions to earnings from the date of equity purchase or asset purchase to fiscal year end. The
fourth column reports gains on equity or asset sales. The last column reports the combined effects.




                                                         27
                                Table 4 Descripti ve Statistics


                                                Small       Rights
                                 Big Loss       Profit     Offering Other Firms
                                   Firms       Firms        Firms                  All Firms



Nu mber of observations              58            88        168          384         698


                                  -110.16       1.03         6.83        8.27         -2.82
ROE                               (-36.01)     (0.98)       (6.67)       (8.29)       (6.3)


                                   36.50        45.38       44.48        46.56       44.99
% Controlling Ownership           (33.02)      (45.74)     (44.19)      (46.43)     (44.90)


                                   21.81        29.91       30.03        32.00       30.32
% Govern ment Ownership           (19.09)      (31.33)     (31.58)      (33.99)     (30.49)


                                    6.60        4.89         4.80        8.44         6.95
Market-to-Book Ratio               (5.16)      (4.47)       (4.04)       (4.23)      (4.29)


                                  1242.47     2235.76      2168.09     3324.68      2717.79
Total Assets (millions)          (933.52)     (1247.90)   (1307.65)    (1281.29)   (1217.09)


This table presents summary statistics of our sample. For 698 2002 Shanghai Stock Exchange
companies who reported a valid ROE, those with ROE less than –10% are classified into Big Loss
Group, those with ROE between 0% and 2% into Small Profit Group, and those with ROE
between 6% and 8% into Rights Offering Group. All others are classified into Other Firms Group.
ROE is return on equity; % controlling ownership is the percentage of common shares owned by
controlling shareholders; % government ownership is the percentage of common shares owned by
various levels and branches of Chinese government; Market-to-Book rat ion is market
capitalizat ion divided by total equity. Other than ROE, variab les are measured at beginning of
year.




                                              28
Table 5 Probi t Regression of Companies Reporting Earnings Management Acti vities on
                                   Determinant Variables


                                        Predicted   Parameter
                                        Sign        Estimate     Chi-SQ        P-value
    Intercept                                           -0.190        0.58         0.44
    Earnings Management Group                  +        -0.049        0.08         0.78
    % Controlling Ownership                    +        0.026        18.68        <0.01
    % Govern ment Ownership                    +        -0.007        3.82         0.05
    Market-to-Book Ratio                       +        0.002         0.32         0.57
    Total Assets                               +        -0.000        0.75         0.39


     Log Likelihood Ratio: -384.41
     Nu mber of Observation: 617


This table reports Probit regression results. The dependent variable takes value one if a co mpany
reports more than one earnings management activ ities, and zero otherwise. For 617 2002
Shanghai Stock Exchange companies who reported a valid ROE, those with ROE less than –10%
are classified into Big Loss Group, those with ROE between 0% and 2% into Small Profit Group,
and those with ROE between 6% and 8% into Rights Offering Group. All others are classified into
Other Firms Group. Earn ings Management Group includes big loss group, small pro fit group and
rights offering group. ROE is return on equity; % controlling ownership is the percentage of
common shares owned by controlling shareholders; % government ownership is the percentage of
common shares owned by various levels and branches of Chinese government; Market -to-Book
ration is market capitalizat ion divided by total equity. Other than ROE, variables are measured at
beginning of year.




                                               29

				
DOCUMENT INFO