The relationship between top management turnover with earnings management and default risk and earnings forecast error in the Tehran Stock Exchange

Article history: Received October 27, 2012 Received in revised format 29 January 2013 Accepted 19 February 2013 Available online February 22 2013 In this paper, we present a study to measure the relationship between top management turnover with earnings management and default risk and earnings forecast error in the Tehran Stock Exchange. The proposed study selects necessary information from 117 firms from the exchange over the period 2005-2010 and, using ordinary least squares technique as well as Pearson correlation ratios, examine three hypotheses of this paper. The results of the survey indicate that there are some meaningful relationships between change in top management with earning management, default risk and earning forecast error. © 2013 Growing Science Ltd. All rights reserved.


Introduction
During the past two decades, there have been various studies on relationship between earning management and top management changes (Degeorge et al., 1995;Firt et al., 1995;Degeorge et al., 1999).Call et al. (2009) studied whether analysts' earnings forecasts were more accurate when they also issued cash flow forecasts.They reported that (i) analysts' earnings forecasts issued together with cash flow forecasts were more accurate than those not accompanied by cash flow forecasts.In addition, they stated that analysts' earnings forecasts could reflect a better understanding of the implications of current earnings for future earnings when they were accompanied by cash flow forecasts.According to Agarwal and Taffler (2008), many developed corporate bankruptcy prediction techniques use a contingent claims valuation approach.They reported that the two approaches capture various aspects of bankruptcy risk, and while there was little difference in their predictive ability in the UK, the z-score approach could somewhat lead to greater bank profitability in conditions of differential decision error costs and competitive pricing regime.Burgstahler and Dichev (1997) provided some evidence that firms manage reported earnings to prevent earnings decreases and losses.They found some evidence that two components of earnings, cash flow from operations and changes in working capital, could be implemented to reach increases in earnings.Dechow et al. (2003) investigated whether boosting of discretionary accruals to report a small profit could describe this "kink" and investigated and discussed a number of alternative explanations for the kink.Brockman and Turtle (2003) proposed a model for firm security valuation based on path-dependent, barrier option models instead of the commonly implemented path-independent approach.They argued that path dependency was an intrinsic and fundamental characteristic of corporate securities because equity could be knocked out whenever a legally binding barrier is breached.Brookman and Thistle (2009) investigated CEOs' risk of termination, its determinants and its impact on firm value based on survival analysis and reported that the risk of termination could increase for about thirteen years before decreasing slightly with CEO tenure; 82% of CEOs had tenure of less than thirteen years.Bergstresser and Philippon (2006) provided some evidence that the implementation of discretionary accruals to manipulate reported earnings was more pronounced at firms where the CEO's potential total compensation was more closely tied to the value of stock and option holdings.Hillegeist et al. (2004) made an assessment on whether two accounting-based measures, Altman's (1968) Z-Score and Ohlson's (1980) O-Score, effectively summarize publicly-available information about the probability of bankruptcy.They reported that researchers could use BSM-Prob instead of Z-Score and O-Score in their studies and reach the SAS code to calculate BSM-Prob.Gong (2011) examined the relationship between CEO compensation and shareholder value added over CEO tenure.The research design studied two fundamental attributes of CEO compensation and shareholder value added.They reported that CEOs receiving higher nominal or realized pay generating more shareholder value.Jog and McConomy (2003) investigated the effect of one such mechanism, namely voluntary disclosure of management earnings forecasts by issuers of IPOs, as a means of reducing asymmetric data as well as ex ante uncertainty.They reported that firms whose forecasts turn out to be optimistic were penalized compared with other forecasters and non-forecasters.Ting (2011) stated that corporates with higher default risk could more likely change their top management in the next financial reporting period and firms default less than other companies.
In this paper, we present a study to measure the impact of top management change on some financial figures.The proposed study of this paper first explains details of models in section 2, the results are given in section 3 and concluding remarks are given at the end.

The proposed study
We first present how to calculate each variables associated with the proposed study of this paper.To calculate default risk, we using the following where CEOchange it is the change in management, Debt it is the debt ratio, LnAsset it is the natural logarithm of total assets, Direct it is the natural logarithm of total assets, and all y variables are dummy variables, which receive one when we collect the information of the same year and zero, otherwise.In addition, we use the following to calculate DD, 2 ln( ) ( 0.5 ) where V A represents total assets, X t denotes book value of total debts, t is due time of debts, which is normally considered as one year, σ A is the standard deviation of total assets and finally μ is the relative growth rate of total assets.In addition, DP, is calculated as follows, where z-score is the Altman bankruptcy ratio, which is calculated as follows, where X 1 is equal to working capita divided by total assets, X 2 is equal to accumulated profit divided by total assets, X 3 is equal to earning before tax and interest divided by total assets, X 4 is equal to market value divided by total assets and finally X 5 is equal to total sales by total assets.In order to study the relationship between change in management and earning management we use the following regression model, (5) In our study, total accruals (TA) is estimated based on change in current assets (ΔCA i,t ), change in total current liabilities (ΔCL i,t ), change in cash (ΔCASH i,t ), current portion of long-term debt (ΔSTD i,t ) and depreciation cost (DEP i,t ) and A i,t-1 is total assets of firms i in year t-1 as follows, , Once we calculate TA i,t-1 we need to calculate α 1 , α 2 and α 3 are calculated as follows, where ΔREV i,t is the change in revenue from year t-1 to year t, PPE it is the growth value of equipment.Therefore, we have, where NDA represents non-discretionary accruals and finally discretionary accrual is calculated as follows, , 2 2 , , .
In order to study the relationship between change in management and earnings forecast error (FE) we use the following regression model, Note that FE is calculated as follows, where YEPS t-1 , RORE t-1 and P t-1 are actual income, predicted income and last stock price, respectively.In order to do regression analysis, we need to know the data are normally distributed using Kolmogorov-Smirnov test and Table 1 shows details of our survey..695.000 .000 As we can observe from Table 1, there are two variables maintain Sig.value of 0.000, which are less than 0.05 and we can conclude that they are not normally distributed.Therefore, we take natural logarithm and repeat the tests summarized in Table 2.The results of Table 2 clearly indicate that the data are normally distributed and we can use ordinary least square techniques.

The results
In this section, we present details of our findings on testing three hypotheses of this survey.

The results of the first hypothesis: The relationship between change management and default risk
To examine the first hypothesis of this survey we use Pearson correlation ratio between change management and default risk as follows, The result of Pearson correlation test is equal to 0.778, which means there is a relationship between these two variables

Table 3
The results of regression analysis on Eq. ( 1   As we can observe from the results of Table 5, there is a meaningful and positive relationship between change management and risk default.Therefore, the first hypothesis of this survey has been confirmed.

The results of the second hypothesis: The relationship between change management and earning management
To examine the first hypothesis of this survey we use Pearson correlation ratio between change management and earning management as follows, The result of Pearson correlation ratio between these two variables is equal to 0.863, which means there is a positive and meaningful relationship between these two variables.Table 6 shows details of our findings on basic statistics on the regression model.The result of Durbin-Watson indicates that there is no auto-correlation among different residuals.
Table 7 shows the results of ANOVA test.The results of Table 8 indicate that there is a meaningful and linear relationship between independent variable and dependent variable.

Table 8
The results of regression analysis The results of Table 8 show a meaningful and positive relationship between earning management and change in management.Therefore, the second hypothesis of this survey is confirmed.

The results of the third hypothesis: The relationship between change management and earnings forecast error
To examine the first hypothesis of this survey we use Pearson correlation ratio between change management and earnings forecast error is as follows, The result of Pearson correlation ratio between these two variables is equal to 0.882, which means there is a positive and meaningful relationship between these two variables.Table 9 shows details of our findings on basic statistics on the regression model.In addition, Table 10 shows details of our findings on ANOVA test One more time the results of ANOVA test indicate that there is a linear relationship between independent variable and dependent variable.

Table 11
The results of regression analysis Finally, the results of our regression model confirms the existence of positive and meaningful relationship between change management and earning forecasted error leading us to conclude that the third hypothesis is confirmed.

Conclusion
In this paper, we have presented an empirical study to measure the effects of change in top management on earning management, earning forecasted error and risk default.The proposed study of this paper gathered the necessary data from Tehran Stock Exchange and using two measures of Pearson correlation as well as ordinary least square techniques, we have examined three hypotheses of the survey.The results have confirmed that there were positive and meaningful relationships between top management change and three mentioned factors.

Table 1
The results of Kolmogorov-Smirnov test ) aLevel of Significance = 5%

Table 4
shows details of ANOVA test for regression analysis on Eq. (1).As we can observe from the results of Table4, F-value is statistically significant and we can confirm the results of our regression analysis.

Table 5
The results of regression analysis

Table 6
The results of some basic statistics aLevel of Significance = 5%

Table 9
The results of basic statistics for the third hypothesis aLevel of Significance = 5%