The impact of politically-connected executives in fraudulent financial reporting : Evidence based on the H shares 1

This study tries to investigate fraudulent financial reporting in China based firms listed on the Hong Kong Stock Exchange which has a high degree of officials’ involvement of China mainland and impact on audit quality and corporate governance. It intends to find out the motives behind fraudulent financial reporting and the opportunity which permits such kind of reporting; and whether the presence of politically-connected executives would likely provide an environment which would be more conducive for fraudulent financial reporting. This study hypothesizes the associations between fraudulent financial reporting and the corporate environment (from the perspectives of economic, ownership, political pressures and audit quality). These were measured by unsigned discretionary accruals, after controlling several characteristics of firm from the samples selected from a number of China based firms listed on the Main Board of the HKSE. The results show that the corporate environment that mostly leads to fraudulent financial reporting is characterized by accounting practices that are already pushing to earnings management. It is also found that firms involved in fraudulent financial reporting have significantly poor corporate governance structures, where the audit quality is lower and outside directors seem over-committed. However, no evidence was found that firm’s political connection factor or the level of board independence plays a significant role in fraudulent financial reporting. The findings of this study would lead to an understanding of corporate behaviors relating to fraudulent financial reporting, and the development of such an understanding is important for preventing frauds and improving better corporate compliance with financial reporting regulations.


INTRODUCTION
Motivated by the growing debate on the role of political institutions in the listed firms (Shleifer and Vishny, 1994;Shleifer and Vishny, 1998;Fan et al. 2007;Faccio, 2006), this study tries to study corporate governance in a political economy by exploring the politically connected executives and their impact in Hong Kong.
This study tries to investigate fraudulent financial reporting in China based firms listed on the Hong Kong Stock Exchange (hereafter referred to as the HKSE) which has a high degree of officials" involvement of China mainland and impact on audit quality and corporate governance.It provides a dynamic setting for testing the value of political connection as prior studies do not discern government"s interests and incorporate ambiguous institutions and self-selection problems by cross-section test.Using data of the China based firms listed in the E-mail: d98722006@ntu.edu.tw.Tel: (886) 02 3366-1136.
HKSE, this study intends to find out the motive (that is, the economic and political factors) behind fraudulent financial reporting and the opportunity (that is, poor corporate governance) which permits such kind of reporting; and whether the presence of politically-connected executives would likely provide an environment which would be more conducive for fraudulent financial reporting; in particular when these firms encounter economic distress, viz.poor return on assets, earning loss, high financial risk, or political distress such as SEC regulation violation.It has been assumed that the politically-connected executives may be considered helpful by the government in responding to firms" distress.Hence, this work intends to review this issue by looking into the relationship between earnings management and the occurrences of fraudulent financial reporting.All these findings are expected to support the phenomenon that politically-connected executives could serve as a disciplinary or monitoring mechanism in a political economy which lacks external market for corporate control and legal protection for investors, instead of being only a form of bail-out.Their efficacy is based on their administrative power, regulatory expertise and accountability to the government"s interests.These results will provide a better understanding of government"s interests and their impact on corporate governance.

LITERATURE REVIEW AND HYPOTHESES SETTINGS
Firms are under strong institutional pressures to prepare their financial statements in accordance with GAAP.Firms would avoid issuing fraudulent financial statements if they knew that they would get caught, or otherwise if they knew it would be undetected.According to Gereish (2003), as long as there is uncertainty about whether or not the deception will be detected, the firm would be faced with the choice to either comply with GAAP requirements or not.The current paper argues that given this uncertainty, the decision to engage in fraudulent financial reporting requires that the firm must be motivated to prepare fraudulent financial reports because of economic, ownership, and political pressures.In addition, the firm must have poor corporate governance structures that make it possible to release false financial statements to the public.The prime focus of this study is to see whether motives (that is, economic and political factors) and opportunity (that is, poor corporate governance) have an impact on fraudulent financial reporting.To ascertain the potential effectiveness of the sweeping changes to motives and opportunity factors, this work tries to examine whether firms that commit fraudulent financial reporting differ from comparable firms in the fraud year and pre-fraud year.Three questions are then to be explored: (i) Is there any significant difference in the characteristics of the executives' motives and opportunity factors between fraudulent and non-fraudulent financial reporting?(ii) What are the factors that lead to the occurrences of fraudulent financial reporting in these firms?(iii) Do fraudulent firms engage in earnings management in the years prior to fraud year?

Motivation for fraudulent financial reporting
The reason that the executive has a strong motivation to adopt an aggressive accounting policy reflects the "factors" that arise and direct behavior, and it is often characterized as arising from unfulfilled needs, desires or deficiencies that spur corrective actions.The concept of motive represents the reasons or purposes of accounting manipulation and it serves as an answer to questions of "why?" Hence, it is hypothesized that a firm is motivated to commit fraudulent financial reporting when the firm has a strong economic need to report results more favorably than they would if the firm followed GAAP requirements (Rosner, 2003), and when the ownership arrangements encourage a short-term orientation to financial performance (Shleifer and Vishny, 1997).Furthermore, it is argued that a firm is motivated to commit fraudulent financial reporting due to political connection factors too (Chaney et al. 2007).
A firm under financial distress (economic factor) and poor financial condition (Bell et al. 1991) may motivate unethical insiders to take actions intended to improve the appearance of the company"s financial position, perhaps to reduce the threat of loss of employment or to acquire as many resources as possible before termination.In addition to motivating the commission of fraud, poor financial condition may indicate a weak control environment, a condition that allows the perpetration of a fraud (AICPA 1997).The proxies for financial distress, decreasing sales and profitability have been linked to various forms of illegal behavior such as collusion, pricefixing and anti-trust violations (Baucus, 1994).Executives of firms with weak financial condition are more likely to window dress in an attempt to disguise what may be temporary difficulties (Rosner, 2003).Hence, under severe financial distress, a firm might fraudulently report more favorable results than would be reported by successful firms.
An important variable unique to Hong Kong that may affect the dependent variables and the interrelationship between them is the political connections of firms.In this study, the politically-connected firms are defined as those owned and controlled by major shareholders and top management that are linked to China mainland based stated owned officials.Proponents of agency theory describe managers (politicians) of politically-connected firms as self-interested, risk-averse, rational actors who try to exert less effort and project higher capabilities and skills than they actually have (Lim, 2002).The connections with politicians give firms preferential access to government"s contracts and subsidies, distress lending, preferential bailouts and barriers to competition.However, political connections of firms are not without their drawbacks.A few empirical studies argued that political relationships are potentially detrimental to shareholder"s value (Cheung et al., 2005, Cull andXu, 2005;Shleifer and Vishny, 1998).Frye and Shleifer (1997) and Shleifer and Vishny (1998) posit that politician"s "helping hand" may also be a "grabbing hand", which leads them to expropriate shareholders" wealth.Cheung et al. (2005) also document that resources are tunneled away from publicly listed firms to the government due to social objectives such as giving bribes, providing vote-buying fund during election years, and providing unnecessary employment in economically depressed states even though they may be costly to the firm and shareholders.More often than not, these studies stress that political relationships do not enhance shareholders" value but rather fulfill political goals of politicians.
Given the above scenario, concerns about the quality of reported earnings may be especially salient for politically-connected executives since these executives typically derive gains from their connections over and above the payments that they make.The nature of these payments and gains may create additional incentives to expropriate, or at least obscure information from the firm"s minority shareholders.Hence, firms with political connections have more tendencies to misreport and overstate earnings since firms with political connections may need to suppress firm-specific information to hide expropriation activities by politicians and their cronies.Further, Bushman et al. (2004) argue that there is a possibility that politicians exploit their control over regulatory policies including relaxing regulatory oversight of the company in question to favor cronies in return for bribes, nepotism and political support.Thus, it is obvious to say that politically connected firms might care less about the quality of the information they disclose since politicians provide protection to their related companies and may help the firms not to be sanctioned.It is therefore predicted that political connection firms will be more likely to engage in fraudulent financial reporting because the benefit of committing fraud is higher than the expected cost and penalty that follows upon detection.The above discussion leads to the first hypothesis: H1: There is a significant relationship between motives (economic and political connections factors) and occurrences of fraudulent financial reporting.

Opportunity for fraudulent financial reporting
Empirical evidence suggests that the opportunity to commit fraudulent financial reporting increases when the firm has poor corporate governance structures.Previous researchers claimed that firms are perceived to have poor corporate governance structures when they have a few Ngan 1877 outsiders on the board and lower audit quality (Beasley, 1996;Sharma, 2004;Woodland and Reynolds, 2003).

Lack of independent board of directors
Agency theory supports the idea that boards can more effectively carry out their duties to safeguard the interests of shareholders especially minority shareholders if there is a heavy presence of outside directors (Fama and Jensen, 1983).This is because a firm dominated by insiders may be more prone to engage in activities that enhance the position of management at the expense of shareholders.Hence, outside directors who are independent and not involved in the daily operations of the firm help monitor management, and ensure that their interests are aligned with those of the shareholders (Jensen and Meckling, 1976;Baysinger and Butler, 1985).Several studies have provided evidence to support the relationship between board independence and the occurrences of fraudulent financial reporting.Beasley (1996) and Uzun et al. (2004) found that firms with a high percentage of outside directors had less financial fraud.They argued that the higher the number of outside directors the more likely they can reduce the fraudulent behavior of the executive directors.
Even though the empirical research as a whole supports the view that increasing director"s independence will lead to more transparent and more reliable financial reports, the idea cannot be generalized to environments such as Hong Kong.The Code on Corporate Governance Practices (Appendix 14) issued by the HKSE in Hong Kong in 2004 recommends that listed firms adopt good governance practice by having a balanced board composed of at least one-third non-executive directors (hereafter referred to as NED) to monitor management.A study conducted by Jaggi et al. (2009) revealed that board dominated by NED does not affect performance.This implies that such recommendation to have at least one third of the board comprising NED may not work in the firm, because most NEDs are selected not because of their expertise and experience but for reasons such as their networking contacts and political connections.They also argued that boards dominated by NED that lack real independence and awareness of their responsibilities, and that do not have the appropriate qualifications and experience could be detrimental to companies.Cheung et al. (2006) provide evidence of a significant positive relation between board independence and earnings management, raising the issue of whether the firms" boards in Hong Kong are effective and truly independent.

Audit quality
External auditors can have a profound effect on corporate fraud by deterring it (that is, reporting the fraud) and by correcting it (that is, by forcing revisions on the financial statements).Researchers have employed various proxies for audit quality, including auditor size and audit fees.However, the validity of a simple Big 4 and non-Big 4 as a definitive quality measure has been called into question by Francis et al. (1999) who raise the possibility that reputation and expertise of individual Big 4 offices are not standard and uniform, but vary from one locale to another along with the city-specific clientele.Francis et al. (1999)"s suggestion is consistent with findings from behavioral research (Carcello et al., 1992) that "team" (local office) factors are more indicative of audit quality than are "firm" factors.Hence, apart from using audit firm size as a proxy for the quality of audit services, several studies show that audit fees can reflect better the level of audit quality.This has been confirmed by Ferguson et al. (2005), showing that auditing is a good example of a service where price may signal quality.
The positive association between audit fees and earnings manipulation is relatively well established.It is argued that fraudulent firms would have higher audit fees than non-fraudulent firms for two reasons.First, since fraudulent firms present greater audit risk, auditors are likely to extend the scope and rigor of their audits.Consequently, the additional cost of the audit may manifest itself in higher audit fees.Second, from a risk-based perspective, it is expected of auditors to increase audit effort and therefore, audit fees for firms with poor governance.This is so because auditors perceive the absence of appropriate controls and oversight in the financial reporting process of fraudulent firms, and consequently auditors are likely to enhance the audit and pass the associated costs to their clients.Although they are not definitive, these earlier results suggest that audit fees are indicative of high audit effort and higher audit quality rather than mere economic bonding.The high audit quality normally requires increase in planned audit hours, increase in audit effort, and the ability to carry out thorough examination of the accounts as well as auditor"s technological capabilities.It is only through high audit quality and high audit effort can a breach in client"s accounting system be discovered and also enables the auditors to report the breach.It is therefore predicted that there is a positive association between audit fees and the occurrences of fraudulent financial reporting.
The discussion above leads to the following second hypothesis: H2: There is a significant relationship between opportunity (poor corporate governance) and occurrences of fraudulent financial reporting.

Relationship between earnings management and fraudulent financial reporting
Earnings management has become a concern not only in the developed countries but also in the highly capitalized economies such as Hong Kong.A number of studies conducted by Ball et al. (2003), Leuz et al. (2003) and Gul et al. (2002) reported that the practice of earnings management is prevalent in the Hong Kong listed firms.This is alarming because previous research indicates that while earnings management practices may start in a small scale, pressures and incentives can later heighten these actions, which may then lead to fraudulent financial reporting (Powell et al., 2005).Dechow et al. (1996) provide evidence suggesting that firms would rather turn to fraudulent financial reporting when they have limited opportunities to change to more aggressive earnings management tactics.Frequently, earnings management involving revenue adjustments generate the need for more sophisticated accounting techniques to ensure analyst"s earnings expectation are met.Eventually, companies must engage in blatant fraudulent activities by creating artificial reserves, understating reserve liabilities, using creative acquisition accounting practices or otherwise manipulating GAAP to perpetuate myths involving company"s "growth".Argenti (1976) noted that manager may resort to fraud and proceed to overstate earnings in failing firms, when firm"s troubles no longer seem temporary and earnings management cannot sufficiently disguise the firm"s failing condition.Therefore, this leads to the final hypothesis as follows: H3: There is a significant positive relationship between earnings management and the occurrences of fraudulent financial reporting.

Sample
The numbers of firms listed on the Main Board of the HKSE at year-ending were 95 (2006), 104 (2007) and 110 (2008).The full list of the H shares is up to 2008.For this study, the enforcement reporter from the Securities and Futures Commission of Hong Kong (hereafter referred to as the SFC and the full list of the H shares up to 2008 can be found from the following web site: http://www.sfc.hk/sfc/html/EN/speeches/public/enforcement/enforcement.html) were reviewed and the sample of firms where fraudulent financial reporting occurred was determined.For the period between 2006 and 2008, a total number of 70 companies were found experiencing fraud and the whole number was taken as the sample of the study.However, ten (10) private companies were excluded from the sample because they are not subject to the same governance and disclosure requirements as listed firms.A fraudulent firm identified from the SFC enforcement releases was included in the sample if the appropriate annual report data and corporate governance information from one of the following sources (that is, corporate annual reports, the HKSE and SFC online search database) were available.Due to the unavailability of the required information (5 companies), misrepresentations in the prospectus of IPO (5 companies) and unknown fraud year (3 companies), the final sample comprised 10 and 99 for fraudulent and non fraudulent firms related to the H shares, respectively.
The data for earnings management cover five years period that started from year 5 prior to fraud year until year before the first year of the accounting fraud.This study predicts that income increasing earnings management to take place within those five years period prior to the fraud year.Basically, earnings management is the discretionary accrual for each year for the 5 years periods.Since the actual occurrences of earnings management are unknown, the formula of performance-adjusted modified Jones model (Kothari et al., 2005) was carried out yearly for those five year periods in order to identify the existence of earnings management.The financial data for earnings management were obtained from the DataStream database, and any missing financial data were obtained from the annual reports.Other data on the research variables were mainly extracted from the annual reports and the CD-ROM database of Hong Kong listed companies: Corporate Documents.This database is published annually by the HKSE and contains copies of corporate documents regarding the board of directors, prior violations, share distributions and financial data.

Models and variables
The following logistic regression model is used to analyze the relationship between various determinants of fraud and the occurrences of fraudulent financial reporting: Model 1 examines a variety of previously suggested determinants of fraudulent financial reporting.Consistent with Beasley (1996), the dependent variable FFR is measured dichotomously.Similar to the SEC in US, the HKSE in Hong Kong only pursues material fraud.Further, due to the nature of the fraud and legal risks, the agreements in settled cases are sealed, and the details of the resolution are not known publicly.The independent variables in Model 1 consist of four potential factors that could lead to the occurrences of fraudulent financial reporting, namely the level of financial distress (DIS) and political factor (POL), percentage of board independence (BDI) and audit quality (AUQ).Specifically, the fraudulent firms predisposed to issuing fraudulent financial reports are more strongly motivated to engage in fraudulent financial reporting (economic and political connections factors), and have a better opportunity to issue fraudulent financial reports due to poor corporate governance.

Motives [Economic factor: financial distress and political connections factor]
The motivation to issue fraudulent financial reporting can be due to either economic factor or political connection factors.Prior literature provides evidence that a firm would issue fraudulent financial statement when it is financially distressed (Rosner, 2003;Beasley et al., 1999).The financially distressed firms (DIS) in this study are measured by using the Altman Z-score.According to Foster (1986), the Altman Z-score has been found to be a good predictor of financial distress.Hence, the variable ZSCORE is calculated for each firm using Altman"s (1993) formula as follows: where: X1 = working capital to total assets.X2 = retained earnings to total assets X3 = earnings before interest and taxes to total assets X4 = market value of equity to total liabilities, and X5 = net sales to total assets A dummy variable coded 1 for low Altman Z-Score (<2.073) and 0 Ngan 1879 otherwise (Altman, 1993).Altman (1993) finds that Z-Score of less than 2.073 is highly stressed and therefore, it is assumed that firms that have Z-score of less than 2.073 have a stronger economic motive to issue fraudulent financial reports than the less financial distressed firms.Specifically, the political connections variable is a dummy variable that indicates whether political connections were present in an organization or not.In this study, POL variable is coded "1" if the firms are owned and controlled by individuals, next of kin, relatives, or associates linked to the top government officials of the political parties and leadership in Hong Kong and China mainland as well; and "0" otherwise.

Opportunity [Poor corporate governance]
Firms have a greater opportunity to engage in fraudulent financial reporting when firms have poor corporate governance structures proxied by lack of independent board and lower audit quality.Consistent with Sharma (2004), the board independence (BDI) is measured as the proportion of independent non-executives directors on the board.Independent directors are those that have no vested interest in the firm and their only observable connection is their appointment as a non-executive director.
Another important factor of corporate governance is the audit quality (AUQ).Audit quality is measured by the ratio of audit fees to total asset and is expected to provide a more robust result compared to the dummy variable used for audit firm size.
Model 2 adds the predictive variable earnings management (EM) to the same independent variables in Model 1. Prior literatures (Dechow et al., 1996;Richardson et al., 2002) document that firms with high discretionary accruals are more likely to be subject to Securities Enforcement Actions.Hence, it is reasonable to expect that discretionary accruals will help predict the likelihood of fraudulent financial reporting too.It is expected to find greater evidence of earnings management among fraudulent firms in the period before the scandal years as well as in the fraud year itself.

Measurement of earnings management
To measure earnings management, this study uses Kothari et al. (2005)"s earnings management model that captures one aspect of the quality of accruals and earnings.Dechow et al. (1996) reported that among the various models, the Jones and modified-Jones models perform the best in measuring discretionary accruals.However, Kothari et al. (2005) find that these models can be improved by controlling the return on assets for the effect performance can have on measuring discretionary accruals.Thus, following Kothari et al. (2005) and measuring performance adjusted modified total accruals for the Jones model.First, total accrual is calculated as the change in non-cash current assets minus change in current liabilities excluding the current portion of long term debt, depreciation and amortization, which is scaled by lagged total assets.Second, the Jones model discretionary accrual is estimated cross-sectionally using all-firm year observations for each industrial sector.
TACCit/TAit-1 = α0 +α1(1/TA it-1) +α2ΔREV it/TA it-1 +α3PPE it/TA it-1 +α4ROAit + εit where: TACC = total accruals TA = total assets ΔREV = change in revenue PPE = gross property, plant and equipment ROA = return on assets Kothari et al. (2005) documented that the inclusion of a constant in the model and the use of assets as the deflator are intended to mitigate heteroskedacity in the residuals.Accordingly, the estimated values of TACC in the above model are normal accruals given the sales and firms" assets.Therefore, the residuals in the regression equation are supposed to be the discretionary accrual in the sense that they are not motivated by either sales or depreciation of assets, which would arise mainly due to the discretionary decisions of managers.The estimates for coefficients α0, α1, α2, α3 and α4 are obtained by sector classification from the regressions and then used to estimate performance-adjusted discretionary accruals (DAC) as follows: Audit quality is measured by the ratio of audit fees to total asset and is expected to provide a more robust result compared to the dummy variable used for audit firm size.

Descriptive statistics
The results of the analysis of the motives and opportunity variables during fraud year are listed in Table 1.
Since the descriptive statistics are for both continuous and dichotomous variables, T-stats and Chi-square tests are used to test for differences where appropriate.Management motives are operationalized by calculating the firm"s Altman Z-Score and by looking at the ownership structures and political connection factors.The results for the motive for fraudulent financial reporting indicate that firms have a significantly different profile on the level of financial distress during the fraud year, which is significantly higher for fraudulent firms.The results demonstrate that fraudulent firms are significantly more financially distressed than non-fraudulent firms, and there is no statistical difference between fraudulent and nonfraudulent firms regarding political connection factor.Table 1 also demonstrates the result of the analysis for the opportunity variables, which are measured by lack of independent board.Contrary to the expectations, the fraudulent firms have a significantly higher percentage of independent board members than the non-fraudulent firms, which are significant at the 0.07 level in the study.
Table 2 reports the Pearson correlations among the explanatory variables; there exists a strong positive correlation between independent board and level of financial distress (0.283), which suggests that companies with independent board members seem to have more financial difficulties.Thus, the recommendation by the Hong Kong Code on Corporate Governance ( 2004) issued by the HKSE to have at least one third of the board comprising of NED may not be beneficial for Hong Kong listed firms, because, typical of developing countries, most NEDs are not selected for their expertise and experience, but for their networking contacts.It could also be because it is very difficult to find outside directors who are truly independent among China based firms as they are closely held, and mostly are politically connected firms.NEDs in China based firms are normally appointed through close ties with governmental officials.Their appointment is symbolic of the government"s trust in them, and not necessarily a proclamation of their competence.Hence, such directors may not be able to independently monitor and reduce the agency conflicts among insider owner, managers and outside minority shareholders.This can potentially lead to misallocation of resources, particularly, during times of financial distress.Overall, the Pearson correlations between the factors are low indicating that multicollinearity is not likely to pose a problem in the regression analyses.

Logistic regression analysis
H 1 -H 2 predicts that there is significant relationship between motives, opportunity and the occurrences of fraudulent financial reporting.The results of Model 1 are presented in Table 3.The level of financial distress is positively and significantly associated with fraudulent financial reporting.This result supports the findings of Rosner (2003), who posits that under severe financial distress, a firm might fraudulently report more favorable results than reporting conservatively.There is no evidence to show that political connection and board independence are significantly related to fraud.
Since it is interesting to examine the environments that subsequently lead to fraud, it is also proper to examine the financial reporting characteristics of fraudulent reporting and non-fraudulent reporting firms in the year before the first year of accounting fraud.It is worth noting that the year prior to the initial year of fraud was used as it may better represent the actual perpetration of the fraud.Hence, the second regression model tests the association between predisposition, motives and opportunity; and the incidence of fraud with variables in the year before the fraud year.The results of Model 1 are provided in Table 4.This model is well specified (p=0.000) and has a R 2 value of 66.5%, with the model correctly predicting 97% of the firms as non fraudulent (98%) and fraudulent (96%).Given the higher explanatory power one could say that those monitoring the financial reporting of fraud firms (e.g.auditors) could have predicted poor financial reporting the year prior to fraud year.The results of Model 1 in the year prior to the fraud year are generally similar to those of Model 1.
The results in the year prior to the fraud year provides additional support for the hypotheses H 1 and H 2 that motives and opportunity have an impact on a firm"s propensity to commit fraud.This suggests that a corporate environment that is ripe for fraud does exist even before the year of the scandal.These results indicate that pre-fraud financial statement factors have   explanatory power in assessing the likelihood of fraud prior to its occurrence.H 3 states that there is a significant positive relationship between earnings management and the occurrences of  fraud.Table 5 reports the results for Model 2 for the fraud year and one year prior to fraud year.In general, the findings remained the same; that is similar to those of Model 1 for both fraud year and in year prior to fraud year.However, it is interesting to note that the addition of earnings management variable (DAC) increases the R 2 to 60.3% from 57.7% and to 69.2% from 66.5% respectively for fraud year and one year prior to fraud year.The difference of 2.6 (fraud year) and 2.7% (one year prior to fraud year) is significant at 5% level.The DACs are found to have a significant and positive relationship with the incidence of fraud during fraud year and one year prior to fraud year at 0.045 and 0.049 levels, respectively.This potentially suggests that accrual management occurs prior to the GAAP violation.This is consistent with Beneish (1997) who stated that GAAP violators have larger positive accruals and discretionary accruals.Beneish (1997) also revealed that fraudulent firms make income-increasing accounting choices prior to the year of GAAP violation either to avoid accrual reversals or because they have exhausted alternatives to better represent their firms" performance.Hence, it can be conclude that the results provided support for H 3 .

DISCUSSION AND RECOMMENDATIONS
This study documented a significant presence of politically-connected executives that provide an environment which would be more conducive for fraudulent financial reporting, in particular when these firms encounter economic distress, viz.poor return on assets, earning loss, high financial risk, or political distress such as SEC regulation violation.The politically-connected firms are defined as those owned and controlled by major shareholders and top management that are linked to China mainland based stated owned officials.Their presence is related to motive (that is, economic factor and political factors) and opportunity (that is, poor corporate governance).Firms with politically-connected executives are more likely to have boards populated with politicians.By contrast, these boards show low degrees of professsionalism, as fewer directors have relevant professional background or prior business experience.The accounting and stock return performance of the firms run by politically-connected executives is poor relative to their less politically-connected counterparts.The results in this study bring deep thinking to both academic researchers and policymakers and they should be useful when trying to understand problems in other transition economies with similar institutional features.This study thus sheds light on the benefits of political participation for private entrepreneurs in China based firms listed overseas and, more importantly, documents several channels that private entrepreneurs use to exploit political networks they develop through active political participation.
In addition to extending the evidence in the literature, this study reveals that institutional constraints and public governance fundamentally affect audit quality and corporate governance at the firm"s level.To market reformers, the findings suggest that the non-transferability of the property rights of firms and the politicians" power of intervening firms and markets are important considerations for both improving corporate governance and the functioning of product and capital markets in China based firms listed overseas.
In appraising the findings of this study, it is important to consider the following limitations.First, the sample size is limited to the firms punished by the H shares during the years from 2006 to 2008.Unpunished cases were not included in the study.Second, this study investigated the data of 3 years that can be limited, as the actual discovery of management frauds normally takes more years and some frauds that occurred in the study period may not be detected.The data of frauds in this study were limited to the cases exposed.Unexposed cases and cases that were under investigation were excluded from the study.Future research is suggested to use a larger sample size with a longer period of data to investigate the motivations, consequences and market reactions of management frauds.The findings of this study have significant implications for an understanding of corporate behaviors relating to fraudulent financial reporting.For example, there is a mixture of acts in falsifying the income statement and the balance sheet.Mostly, firms committed several frauds simultaneously; adopted more complicated approaches; and, particularly, they liked to cover up warranties, significant events and transactions with related parties.The development of such an understanding is important for preventing frauds and improving better corporate compliance with financial reporting regulations.

Conclusion
The results of the above study show that corporate environment that most likely leads to fraudulent financial reporting is characterized by accounting practices that are already pushing to earnings management.In addition, it is also documented that firms are embroiled in fraudulent financial reporting when the level of financial distress is high.The results also find that firms involved in fraudulent financial reporting have significantly poor corporate governance structures whereby the audit quality is lower and outside seem over-committed.Finally, there is no evidence found that firm"s political connection factor or the level of board independence plays a significant role in the potential for fraudulent financial reporting.

Table 1 .
Descriptive statistics comparing profiles of fraud and no-fraud firms.

Table 2 .
Correlations between test variables.

Table 4 .
Logistic regressions for one year prior to fraud year (model 1).
fraudulent financial reporting.To test this hypothesis, it includes the earnings management variable measured using DAC as an indicator of the propensity to commit

Table 5 .
Logistic regressions with addition of earnings management variable during fraud year and one year prior to fraud year (model 2).