Accounting Standards , Goodwill Impairment and Earnings Management in Malaysia

This paper has two objectives. The first is to investigate whether the provision of MFRS 136 (that is equivalent to IAS 36) allows for an earnings management technique termed “big bath” in Bursa Malaysia and, whether the change in firms’ CEOs plays a role in goodwill impairment loss recognition. The second objective is to examine the application of goodwill impairment loss recognition under the provision of MFRS136 in Bursa Malaysia between 2011–2012, using a “Mann-Whitney U” test. Our findings suggest that companies utilize goodwill impairment to take a “bath” regarding earnings management. Moreover, we find that Malaysian companies struggle to understand the ambiguity of the Value in Use (VIU) method under the MFRS 136 as they tend to take extreme accounting treatment strategies. Finally, we find that the new CEOs in Malaysian companies are more conservative in accounting treatment than CEOs with long tenure.


Introduction
In March 2004, IAS 36 (Impairment of Assets) was reissued to ensure that the companies' assets are not carried at more than their recoverable value.IAS 36 requires an annual impairment test for the companies' Cash Generating Units (referred to as CGU) without specifying a time for the test.Thus, companies are allowed to perform an impairment test for their CGU at any time in the accounting period as long as they repeat the test in the same period in the following year.There are two methods to follow in order to estimate the recoverable amount for CGS's, namely, fair value method and Value In Use method (referred to as VIU).This study focuses on the VIU as it is the most applied method in Bursa Malaysia (Haron & Atan, 2010).
The Malaysian Accounting Standards Board was established under the Financial Reporting Act 1997 as an independent authority to develop and issue accounting and financial reporting standards in Malaysia.This Act gives the issued standards legal authority to the MASB.In August 1, 2008, the Financial Reporting Foundation (FRF) and Malaysian Accounting Standards Board (MASB) issued a statement about their plans to bring Malaysia to full convergence with International Financial Reporting Standards (IFRS) on 1 January 2012.MFRS 136 is equivalent to IAS 36-Impairment of Assets as issued and amended by the IASB, including the effective and issuance dates.Entities that comply with MFRS 136 simultaneously are in compliance with IAS 36.
The main challenge in the VIU method is the determination of the present value of future cash flow for each CGU.The determined amount is also subject to selection of the level of discount rate on the one hand, and risk assessment for each CGU on the other.Moreover, these estimations are subject to tax deductibility, depreciation treatments and some financial cost (Wine et al., 2007).All these factors should be considered and supported when preparing the Discounted Cash Flow (DCF) model.The fact that the DCF model is subject to a large degree of estimation requires a solid explanation from the firms when preparing their financial reporting.Khairi et al. (2012) found that Singapore companies fail to provide meaningful information related to basic allocation of goodwill into CGU.They also argue that the complexity of the IFRS 36 may cause this failure as the companies struggle to understand this process.Similar views were shared by Giuliani & Brannstrom (2011) who found that the concept of goodwill under IFRS3 still appears to be a "black box" in Italy and Sweden.
to certain assets, then using a CGU for a collective number of assets generates cash inflows and is capable of detecting these inflows independently from other assets or CGUs.This process relies on a wide range of estimations and judgements that seek to trace the cash inflows form a collective number of assets.According to Wine et al. (2007), the main concern about the introduction of CGU in the IAS 36 is the fact that firms are able to avoid or to defer individual assets write-down, as they can be referred to their CGU that can generate internal intangible assets.This fact raises a serious question about the prohibition of recognizing the internally generated goodwill as firms can recognize the allocated internally generated goodwill for each CGU and offset it against any decline in the value of other assets within the CGU.Moreover, IAS 36 does not prevent changing the CGUs each year.This allows the management to make a flexible decision about shifting the profitable CGUs to other CGUs in that their value might be at some future date subject to impairment write downs.
The VIU method has been under heavy criticism.For example, Lonergan (2010) argues that VIU should be banned until its problems are resolved as they may contribute to creative accounting.A more optimistic view has been pointed out by Wines et al. (2007) as they argue that the numerous valuation of CGU could be the main reason for this ambiguity and more effort should be made to simplify the estimation method.
This study investigates whether Malaysian managers utilize the new MFRS136 to practise "big bath" behaviour by reducing their earning level during the adoption of MFRS 136, for many studies find a strong association between the IAS36 and big bath behaviour (Khairi et al., 2012;Guler, 2007;Jordan & Clark, 2004;Sevin & Schroeder, 2005;Brütting, 2011).Moreover, we believe that assigning new CEOs to the companies will increase the chance of utilizing the MFRS 136 toward big bath behaviour (Rees et al., 1996;Beneish, 2001).This study focuses on these two issues by collecting a random sample of 250 listed Malaysian firms and building a comparison between companies which reported goodwill impairment losses and those which did not, between the period of 2011 and 2012; that is, before and after the full convergence of the MRFS 136.In addition, we trace the earning level for these two groups to measure their negative and positive earning level as indicators for big bath behaviour (Jordan & Clark, 2004;Sevin & Schroeder, 2005;Lapointe-Antunes et al., 2009;Swanson, 2007;Alfons, 2009).Lastly, we make another comparison between companies which assigned a new CEO and those which did not, with both reporting a goodwill impairment loss.
The first finding of this study suggests that "Big bath" accounting occurs just before implementing the MFRS 136.The second finding suggests that when companies change their CEO, goodwill impairment is not utilized for "Big bath" accounting in earnings management; it also indicates that new CEOs in Malaysian companies are more conservative in accounting treatment than CEOs with long tenure.
The rest of the study proceeds as follows: a review of previous studies related to our topic, methodology of the study, data analysis, study conclusion, and recommendations for future researcher.

Material and Methods
Prior to IFRS 3 Business Combinations (2004) and related amended versions of IAS 36 and IAS 38 issues, the most common approach was to amortize the goodwill on a systematic basis over the useful life of the assets.Sproul and Higson (2005) find that 94% of UK companies apply the goodwill amortization before the implementation of the IFRS3.Other approaches provided by the IFRS such as "Fair Value" were not preferable due to the lack of an active market that allowed for an accurate estimation of the CGU.Equally important, the simplicity of the goodwill amortization approach and the similarity of the treatment of other intangible assets contributed to the prevalence of the goodwill amortization over other approaches.However, many studies argued that a regular and systematic decrease of the goodwill may not be consistent with the fixed annual deduction of the goodwill (Zucca & Campbell, 1992;Sevin & Schroeder, 2005).
According to Brütting (2011), goodwill impairment occurs in a time length between the dates of acquiring the goodwill until shortly before disposal.She suggests that the beginning of the acquirement can cause impairment activities due to overpayment in the first year such as (Hubris, agency conflict, competition between acquirers).The period between the first year of acquisition and the day before disposal can trigger impairment activities if the management faces negative economic conditions (either before or after the acquisition).Lastly, the end period of the goodwill can be triggered by the event of assets disposal.
Previous studies showed severe disagreement about the effect of goodwill impairment loss under the IAS 36 or its US equivalent, and about whether the new rule contributes or not to opportunistic behaviour by managers.While many studies support this view, other studies argue that the results' variations are more closely related to economic factors rather than to opportunistic behaviour.
Using impairment recognition for opportunistic behaviour by managers is not a new topic (see Strong & Meyer, 1987;Elliott & Shaw, 1988;Zucca & Campbell, 1992).Francis et al. (1996) were among the first to examine whether managers are motivated to utilize assets write-off to control the firms' economic circumstances.They found that asset impairment and managerial incentives are important determinants of asset write-offs.Meanwhile, Rees et al. (1996) examined the effect of abnormal accruals of firms, which recognized continuous asset impairment in their financial reports, in order to detect any systematic pattern of earning management in the year of the write-down.They found that pre-write-off earnings are, on average, significantly worse than industry medians, which is consistent with the "big bath" reporting behaviour.Moreover, Riedl (2004) argues that long-lived assets impairment is more likely to reflect opportunistic reporting by managers than provision of their private information.
In the same line, a number of studies supported the view of utilizing goodwill impairment by managers to enhance earnings.Beatty and Weber (2006) found that both contracting and market incentives affect firms' accounting choices relating to the trade-off between the timing and the presentation of goodwill impairment loss recognition on the income statement.Haman and Jubb (2008) found that discretionary accruals of goodwill firms are higher than those of non-goodwill firms in the adoption year of the IAS 36.They suggested that the IAS 36 rule may provide flexibility to managers in dealing with any impairment loss.Ramanna and Watts (2012) found some evidence of association between goodwill non-impairment and CEO compensation, CEO reputation, and debt covenant violation concerns.Lastly, Jahmani et al. (2010) found signs of income smoothing been deliberately conducted by management through selecting the timing of goodwill impairment recognitions.
In contrast with these studies, Godfrey and Koh (2009), Jarva (2009), Schultze andWeiler (2010), AbuGhazaleh et al. (2011), Lee and Yoon (2012) focused on goodwill impairments appearing as operating expenses after the application of IAS 36.Godfrey and Koh (2009) found a strong negative association between firms' investment opportunities and the amount of goodwill impairment reported.However, they also found an inverse relation between the firms' accounting returns and the reported goodwill impairment amounts.They concluded that firms faring well economically have less reason to record large impairment losses.Jarva (2009) examined whether SFAS 142 goodwill write-offs are associated with future expected cash flows and found that SFAS 142 have significant predictive ability for one-and two-year-ahead cash flows.His results are consistent with the notion that goodwill write-offs are more closely related to economic factors than opportunistic behaviour.Schultze and Weiler (2010) argued that information required by IFRS and US-GAAP to evaluate a firm's goodwill can be used to design a performance measurement system which provides information about both value creation and realization of value.AbuGhazaleh et al. (2011) suggested that IFRS No. 3 improved the quality of reported goodwill impairment losses, and hence provided support to the IASB's contention that IFRS 3 allows firms to reflect their underlying economic attributes.Lastly, Lee and Yoon (2012) found evidence that the ability of earnings to predict future operating cash flows and earnings persistence improved after the enactment of SFAS No. 142.They argued that the results of their study contradicted the pervasive evidence of opportunistic reporting.
In the international context, goodwill impairment loss recognition under IAS 36 has been also considered by many studies in Italy, Australia, New Zealand, Finland, Holland, Romania, and others.
In Italy, D'Alauro (2013) found a significant positive association between the level of mandatory disclosure about goodwill impairment tests and the magnitude of goodwill write-offs and earnings performance.However, they also found that managers tend to avoid goodwill write-offs despite market indications of impairment, which is consistent with income smoothing techniques.Carlin and Finch (2011) found evidence from Australian and NewZeland firms that are consistent with opportunism in the selection of discount rates.Their results suggest the existence of a bias among Australian and New Zealand firms towards the application of lower than expected discount rates.This is interpreted as evidence of the opportunistic exercise of discretion to avoid unwanted impairment losses.Pajunen and Saastamoinen (2013) found that managers of listed Finnish companies behave opportunistically in goodwill write-off decisions.Lemans (2009) failed to find evidence that indicates that goodwill impairments are indeed being used to manipulate earnings in Holland.The results for the total sample are heavily influenced by the transition year to IFRS (2005) and the credit crisis (2008), and no strong evidence is found which indicates that management indeed uses goodwill impairments to manipulate earnings.Alfons (2009) found that managers do not use goodwill impairments to take "big baths" in the Dutch market.Vladu et al. (2014) found consensus among Romanian professional accountants that creative accounting techniques are presented for profit over/undervaluation, depreciation and provisioning in the private sector.
At another level, Van de Poel et al. (2009) and Stumple (2012) used the European Union as a sample for their study.Van de Poel et al. (2009) found that companies typically take their impairments when earnings are unexpectedly high (smoothing) or when they are unexpectedly low (big bath accounting).Stumple (2012) found that firms with unexpected negative earnings seem to be accelerating the impairment recognition, which indicates a big bath accounting behaviour.
In general, previous studies carried many arguments and counter-arguments about utilizing the goodwill impairment loss for opportunistic behaviour.The current study focuses on the Malaysian equivalent amended from the IAS 36, namely, MFRS 136, to see whether there are signs of big bath behaviour following its application.
Big bath accounting is a phenomenon which occurs when managers use the financial distress of the company to 'take a bath' and conduct-sometimes overstated-asset write-downs in the hope of improving future earnings expectations (White et al., 2003).
Previous literature focused on investigating signs for creative accounting in the US and among international firms.The reason that researchers took interest in investigating management behaviour toward the impairment is due to huge degree of complex estimation that needs to be done when assigning the firm's goodwill, especially when using the VIU method (Khairi et al., 2012).
For example, Riedl (2004) found a strong relation between assets impairment and big bath reporting behaviours among US firms.He argued that big bath is more likely to reflect opportunistic reporting by managers rather than private information.Jordan and Clark (2004) suggested the presence of big bath earning management in US market during 2002, as they applied a regression analysis of 100 companies between 2001 and 2002 in order to study the effect of the new SFAS 142.Additionally, Sevin and Schroeder (2005) argued that smaller firms are more likely to use big bath as they reported greater negative earnings under the SFAS 142.Guler (2007) conducted a conditional logistic regression on 130 firms to explore the goodwill impairment losses' decision when the earnings are 'unexpectedly' high/low after applying the SFAS 142.He also investigated the relationship between top executives' exercisable options and top executives' earnings, based bonuses and the strength of the board with impairment losses' decision.His findings indicated negative associations between bonus grants and in-the-money option holdings for executives with the write-off reported under SFAS 142.Moreover, he argued that managerial incentives do affect the implementation of standards that expand managerial discretion and highlight the importance of corporate boards in the monitoring of discretion allowed by such standards.In addition, the period after implementing the IFRS3 showed an increase in using an impairment testing approach to create big bath accounting behaviour (Brütting, 2011).In Canada, a study conducted by Lapointe-Antunes et al. (2009) showed that goodwill impairment losses are associated with management incentives in overstating and understating earnings.
Overall, there is a lot of evidence that supports utilizing the goodwill impairment to take big bath behaviour (Khairi et al., 2012;Guler, 2007;Jordan & Clark, 2004;Sevin & Schroeder, 2005;Brütting, 2011).Most, if not all, of these studies conducted their analysis in the period before and after the implementation of the IAS 36/SFAS 142 (Jordan & Clark, 2004;Sevin & Schroeder, 2005;Lapointe-Antunes et al., 2009;Swanson, 2007;Alfons, 2009).Indicators measurement for big bath behaviour focused on companies' earnings such as (positive vs. negative earning, earning per share, return on assets, return on sales and key benchmarks) before and after the implementation of the IAS 36/ SFAS 142 as the major indicator of big bath activities (Lapointe-Antunes et al., 2009).Altogether, we expect that Malaysian companies will show a response similar to those countries which adopted the IAS 36, by showing an increasing negative earning for companies which reported goodwill impairment losses after the application of MFRS 136 compared to companies which did not.To test this claim, the following hypothesis has been developed: H1: Companies that impaired goodwill between 2011-2012 will show a negative earnings level compared to those which did not.
The 'big bath' phenomenon is particularly suitable to explain a change in top management, as the new CEOs may find large opportunities for big bath accounting and blame their predecessors for bad performance (Zang, 2008;Masters Stout et al., 2007).In such cases, the new manager might have an incentive to take a write-down, and thus 'clean the house' (Alciatore et al., 1998).Stumple (2012) argues that signs for big bath are more likely to be found when new CEOs are assigned as they attempt to charge extra goodwill impairment.Prior research also found indications about the "big bath" timing when new CEOs arrive.Elliott and Shaw (1988) found that many write-downs occur in the fourth quarter of the year.One possible explanation that is consistent with the 'big bath' hypothesis could refer to the new managers' decision to delay their write-downs until the company earnings can be foreseen and adjusted to managers' best interest.In a subject related to the change in CEOs, Viśnevskaá (2010) found that CEOs' tenure is significantly associated with the financial reporting behaviour of the company.This changing behaviour is also strongly associated with the magnitude of goodwill impairment.Her results contradict those by many researchers who found a strong association between big bath accounting and early stages of CEO employment period, such as (Zang, 2008;Masters Stout et al., 2007).Thus, the arrival of a new CEO does indeed affect the impairment goodwill decisions.However, previous studies show different responses by the new managers to the write-off decisions.Given the fact that Malaysian companies are newly applying the MFRS 136 amended, we expect that new CEOs in Malaysia will take advantage of this rule by increasing their write-off behaviour to "clean their house" and blame their predecessors for bad performance.Therefore, the following hypothesis will be tested: H2: Companies which changed their CEO and impaired goodwill during 2011-2012 will shows a negative earning level compared to those which did not do so.
The process of constructing the research sample proceeded as follows.First, the selected 250 firms listed in the Bursa Malaysia Index on the 1st of January 2011.The final sample consisted of the following companies: 1) Firms that recognised goodwill as an assets base in their financial reporting since 31 December 2011.
2) Firms that used MFRS as their reporting base.
3) Firms that adopted the VIU method as a basis for CGU recoverable amount estimation.4) Firms that used a single discount rate for the purpose of impairment test.
A random selection of 250 companies between the periods 2011 to 2012 was conducted.The sample revealed that (58.8%) of firms recognized goodwill in their financial statement during 2011-2012.Companies were grouped as follows: 1) Companies that impaired goodwill during 2011or/and 2012 (N=17.2%); 2) Companies that changed their CEOs and impaired goodwill during 2011or/and 2012 (N=7.2%); 3) Companies which did not impair goodwill during 2011or/and2012 (N=28.3);4) Companies which failed to meet the research criteria (N=6.1%).
To test the argument that the MFRS 136 creates a chance for creative accounting, a comparison between two groups was done.One group included the companies which impaired goodwill during 2011 and the other group included companies which did not.This process was repeated in 2012 after re-organizing the groups.Jordan and Clarck (2004) investigated the role of the SFAS 142 in 100 leading companies.They divided their sample into two groups.The first group were companies which reported an impairment of goodwill and the second group included companies which did not report impairment of goodwill.They then measured the earning level for both groups in the period between 2001 (before the implementation) and 2002 (after the implementation) using two measures: Return On Assets (ROA) and Return On Sales (ROS).They found that companies which reported goodwill impermanent in 2002 showed a decrease in earning level compared to companies which did not report an impairment of goodwill.In a likewise manner, Sevin and Schroeder (2005) adopted the same two measuring tools, namely, ROA & ROS, to investigate the effect of SFAS 142 in big bath accounting behaviour using the earning level for companies as the main indicator.120 companies which reported impairment of goodwill and 82 companies which did not report such impairment were chosen as the sample size.They also considered the company's size to see whether they showed different big bath behaviour.Their results indicate that small firms tend to show more evidence of taking a bath compared to large companies.A more recent study by Brütting (2011) adopted the same two measures, ROA & ROS, and found an increase in using an impairment testing approach to create big bath accounting behaviour.In a similar fashion, this study adopts the ROA & ROS measures as indicators of earning level in search for signs of big bath behaviours among companies which impaired goodwill losses after the implementation of the MFRS 136 in Malaysia.Moreover, the same indicators are utilized to measure the earning level among the companies which changed their CEO in the same period in implementation of the MFRS 136.

Results and Discussion
To test the first hypothesis we used Sevin and Schroeder's (2005) and Jordan and Clarck's (2004) method to compare the changes in goodwill to total assets and the change in companies ROA and ROS indicators as signs for earning levels that indicate "Big bath" accounting activities.Table 1 shows the "Mann-Whitney U" nonparametric test between the two groups as follows: 1) Companies which impaired goodwill during 2011 and 2012.
2) Companies that did not impair goodwill during 2011 and 2012.Both big bath indicators ratios (ROA and ROS) show a significant difference that indicates an earnings management for the overall sample.But once we divided the sample into two periods, the results showed a significant increase in goodwill impairment loss recognition during 2011 compared to 2012 (Median= 0.089 in 2-tailed); that is, one year before applying the MFRS 136 provision.Moreover, although goodwill impairment represents a small portion of the company's assets, the change in the ROA and ROS is significantly high in 2011 (Median= 0.004 & 0.013 in 2-tailed respectively) compared to 2012 (Median= 0.69 & 0.160 in 2-tailed, respectively).Similar studies that covered this topic found evidence of increasing goodwill impairment during the period that follows the transition period (Jordan & Clark, 2004;Sevin & Schroeder, 2005;Lapointe-Antunes et al., 2009;Swanson, 2007;Alfons, 2009).However, our study contradicts these studies, as we found increasing activation of goodwill impairment loss recognition just before the transition period.These results should be interpreted with caution, because transition periods may lack generalisability, as some accounting treatments are permitted during this period (Godfrey & Koh, 2009;Jarva, 2009;Schultze & Weiler, 2010;AbuGhazaleh et al., 2011).Moreover, we note that goodwill represents a small portion of Malaysian firms' assets; this may be due to the relatively small stock market compared to others in US, Canada and Europe.Another explanation could be related to Khairi et al. (2012) findings, as they conclude that companies struggle to adopt the IAS 36 due to its complexity.Thus, Malaysian companies might have significantly impaired their goodwill in order to avoid dealing with MRFS 136 treatment complexity, which may eventually lead to this impact on the "Big bath" signs.
Moreover, "Big bath" accounting activities which occurred in 2011-just before the implementations could be due to the following reasons: 1) The new MFRS 136 implementation could be the main cause for this significant difference in the companies' ROA and ROS.
2) Many companies (25%) impaired their total goodwill before the implementation of MFRS 136, which may explain the main effect of the new rules.
3) The significant difference in ROA and ROS does indicate some "Big bath" activities.However, a general conclusion could be hard to predict during this period.
In general, the findings suggest that the implementation of MFRS 136 has a significant effect on the companies that recognized impairment activities.In order to further investigate "big bath" accounting signs, the following question was asked: "Is there any difference in positive vs. negative earnings for companies which impaired goodwill, and if so, which year was that?" (See, Table 2).Messier et al. (2005) argue that an item's effect on income is considered as an important factor in determining the materiality of that item.They add that an item that affects income by 10 percent or more can be considered as material.Following this, Table 2 confirms that there is a significant difference in 2011 between positive and negative earnings for the companies that impaired goodwill and companies that did not.38% of companies which impaired goodwill in 2011 reported negative earning while just 14% of companies which did not impair goodwill reported a negative earning.In contrast, only 22.5% of companies which impaired their goodwill in 2012 reported negative earning, while 24.5% of companies which did not impair goodwill reported a negative earning.This may suggest that MFRS 136 allowed for some "big bath "activities, especially in the period before the application of the MFRS 136.
The second objective of this paper is to look for signs of "big bath" accounting activities by using the goodwill impairment in companies that changed their CEOs in 2011 and those that retained their CEOs.
Table 3 show the results of Mann-Whitney U non-parametric test that compares the "Medians" between the two groups as follows: 1) Companies which impaired goodwill during 2011 and/or 2012 without changing their CEOs.
2) Companies which impaired goodwill during 2011 and/or 2012 while changing their CEOs.The results indicate that there is no significant deference between the two groups in both 2011 and 2012 (Median= .052..409 in 2-tailed, respectively) Therefore, there is not enough evidence to confirm "Big bath" accounting techniques under the MFRS 136 when companies change their CEO's.However, the significant difference in the ROS indicator in 2012 (Median = .144in 2011) compared to (.829 in 2012) raises a question about the reason for this shift from the normal trend.As a result, a year-to-year comparison for each group was conducted for 2012.The reason for this comparison was to detect the new managers' attitude towards the new MFRS 136, and to determine whether they took the same path that managers with longer tenure took as they heavily impaired their goodwill in 2011.The comparison relied in the change that occurred on the impairment loss on goodwill, impairment loss on total assets, impairment loss on total sales and impairment loss on operational income during 2012 compared to 2011.The two groups were arranged as follows: 1)-Companies that impaired goodwill during 2012 without changing their CEOs.
2)-Companies that impaired goodwill during 2012 and changed their CEOs.
The results in table 4 suggest that there is a significant difference between the 2 years activities when the companies changed their CEOs in 2011.The new CEOs in Malaysian companies did increase their impairment activities in 2012; that is, after the adoption of MFRS 136.However, their activities did not show a significant effect on negative earning levels.This result is consistent with Godfrey and Koh (2009), Jarva (2009), Schultze andWeiler (2010), AbuGhazaleh et al. (2011), Lee and Yoon (2012), who assumed that goodwill write-offs are more closely related to economic factors than to opportunistic behaviour when new CEOs are assigned.The results also indicate that new CEOs show a better commitment to the adoption of the MFRS 136 compared to managers with long tenure.One possible explanation is that the new CEOs in Malaysian companies are more conservative in accounting treatment than CEOs with long tenure.However, signs for "big bath" accounting were not found.

Conclusion
This paper has studied MFRS 136 convergence in Malaysian market in order to expand the accounting literature on the "big bath" theory and to provide additional evidence of its existence.This study may provide evidence to support previous findings that companies utilize goodwill impairment to take a "bath" regarding earnings management (Stumple, 2012;Viśnevskaá, 2010;Sevin & Schroeder, 2005;Jordan & Clark, 2004).However, the study focused on the transition period for MFRS 136 convergence.Therefore, theresults should be interpreted with caution, because transition periods may lack generalisability, as some accounting treatments are permitted during this period (Godfrey & Koh, 2009;Jarva, 2009;Schultze & Weiler, 2010;AbuGhazaleh et al., 2011).Moreover, we notice that goodwill represents a small portion of Malaysian firms assets, due to the relatively small stock market compared to others in US, Canada and Europe.The findings also suggest that if "Big bath" accounting was utilized by managers, signs would be more likely to appear just before implementing the MFRS 136.This paper further suggests that this behaviour is consistent with Khairi et al. (2012) who found that companies struggle to apply the IAS 36.Therefore, companies went through major goodwill impairment activities in order to avoid this complexity (25% of those companies fully impaired their goodwill in 2011).
Based on these findings, we conclude that Malaysian companies struggle to understand the ambiguity of the VIU method under the MFRS 136 as they tend to take extreme accounting treatment strategies.
Another finding of this paper suggests that when companies change their CEO, goodwill impairment is not utilized for "Big bath" accounting in earnings management.Surprisingly, our results are different from those by Stumple (2012), Zang (2008), andMasters Stout et al. (2007), as we found that new managers avoid recognizing goodwill impairment in their first year.Rather, they tend to delay changing the impairment until their second year, as we also found that companies which changed their CEOs impaired their goodwill in 2012.This finding is consistent with Viśnevskaá (2010), whereby although she found a significant relationship between CEO tenure and "big bath" accounting behaviour for goodwill impairment, she also failed to capture any "big bath" signs for new CEOs.We conclude that new CEOs in Malaysian companies are more conservative in accounting treatment than CEOs with long tenure.
Based on our results and findings, future studies could investigate the following: first, the impact of goodwill impairment in income smoothing when new CEO is appointed; second, the relationship between CEO's tenure and "Big bath" accounting behaviour for goodwill impairment in Malaysian listed companies; finally, different measurement tools such as Jones (1991) model, a modified Jones model used In Dechow et al. (1995), or regression analysis may provide more solid evidence about utilizing goodwill impairment for earning management.

Table 3 .
Goodwill impairment and change of CEO