Amsterdam Business School
The influence of family ownership on goodwill impairment
Name: Sabina Dobriakova Student number: 11434376
Thesis supervisor: dhr. dr. Pouyan Ghazizadeh Date: 25 June 2018
Word count: ______
MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam
Statement of Originality
This document is written by student Sabina Dobriakova who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Contents
Abstract ... 4
1. Introduction ... 5
2. Literature review ... 7
2.1. Earnings management and goodwill impairment ... 7
2.1.1 Earnings management: definition, methods, factors and incentives ... 7
2.1.2 Earnings management through goodwill impairment accounting under IFRS... 9
2.2 Family ownership and agency theory ... 11
3. Motivation and hypotheses development ... 13
4. Research methodology, sample selection and data sources ... 16
Abstract
Accounting practices for goodwill impairment require the exercise of considering management judgment and reporting discretion. (Amiraslani, Iatridis, & Pope, 2013).
In addition, this reporting discretion leads to a greater likelihood of earnings management.
Managers and controlling owners have a possibility to manage their firm’s economic performance by means of goodwill impairment: they may move future write-offs to current periods or move current write-offs to future periods to manipulate their firms’ current earnings.
A level of ownership concentration affects earnings management engagement of the
firms(Achleitner, Günther, Kaserer, & Siciliano, 2014; Carrera, 2017; Goh, Lee, & Lee, 2013; Martin, Campbell, & Gomez-Mejia, 2016; Prencipe, Markarian, & Pozza, 2008; Tsao, Chang, & Koh, 2017; Wang, 2006). In family firms, concentrated ownership is in the hands of a controlling family. Building on the agency theory, alignment effect and entrenchment effect, I expect that family firms’ accounting behavior regarding goodwill impairment write-offs differs from that of nonfamily firms.
To test this, two regression models are developed including firm-specific factors and proxies for big bath accounting and income smoothing practices, as well as a factor of institutional settings (weak and strong legal enforcement). The first model measures the impairment decision as a dummy variable. The second model measures the impairment decision as the amount of the impairment deflated by total assets.
1. Introduction
This research examines whether or not family ownership affects accounting behaviour regarding earnings management through goodwill impairment method.
Earnings management occurs when there is an incentive for managers to structure transactions and mislead some stakeholders about the economic performance of the company. Aggressive use of earnings management is questionable from a moral hazard and ethical perspective (Kaplan, 2001). A substantial body of research investigates earnings management accounting practices used by
executives. However, the role of the dominant shareholder in the earnings management decision receives sparse attention (Goh et al., 2013; Martin et al., 2016).
Family firms play a significant role in the world economy. According to the Family Firm Institute statistics, an estimated 70%−90% of global GDP annually is created by family businesses. Namely, European family businesses represent 1 trillion Euros in turnover which is 60% of all European companies (KPMG, 2013). Moreover, family businesses account for 9% of the European Union’s GDP and they create over 5 million jobs in Europe (40−50% of all employment) (KPMG, 2013). Researchers argue that family firms have their own specific features and act in a different manner as compared to non-family firms. It is a business phenomenon. In the last few years, the interest in accounting research within family firms’ context grew a lot (Carrera, 2017). The distinguishing features raise certain issues regarding the extent of earnings management in family firms relative to non-family firms.
Deterioration in performance may increase the likelihood of recognition of goodwill impairment in firms. It is believed that the magnitude of discretion given in the accounting standards IFRS 3 and IAS 36 gives managers an additional incentive to manage the perception of users of financial statements using the impairment of goodwill item. This problem can be exacerbated when there is a high concentration of family ownership and when family has control over the management.
Since the subject of interest is the earnings management through goodwill impairment method, the existing research models are used. An important research performed by Van de Poel et al. (2008) demonstrates that the goodwill impairment decision is highly associated with financial reporting incentives. Their findings support the argument that companies take their impairments when earnings are “unexpectedly” high (smoothing) or when they are “unexpectedly” low (big bath accounting).
The focus of the research by Van de Poel et al. (2008) was on 15 EU countries under IFRS, however the level of ownership concentration is not taken into account.
Building on the agency theory, findings of Greco, Ferramosca and Allegrini (2015) provide
evidence that\ nonfamily firms use write-offs for earnings management purposes while family firms report write-offs coherent with the firm performance.
Hence, it is interesting to understand how the family as a dominant shareholder influences accounting behaviour in the form of earnings management through the goodwill impairment
method. Does family firms’ accounting behaviour regarding goodwill impairment differ from that of nonfamily firms?
The sample of public EU-based industrial firms in the period 2006-2017 is used for the research. The findings can contribute to the investor decision-making at family firms as well as to the literature addressing the impact of family ownership on earnings management.
The remainder of this paper is organized as follows. In the next chapter, earnings management will be defined and the focus will be brought on different conditions, incentives and forms of earnings management. In addition, there will be discussed key characteristics of family ownership and how the agency theory explains accounting behaviour and management decision within family firms. Based on insights from the prior research literature, there will be discussedthe link between managing earnings, the impairment of goodwill and family ownership.
Further, the hypothesis development and research design will be discussed. Chapter 4 contains information regarding the research methodology and sample selection. Chapter 5 contains descriptive statistics and correlations.
The results of the research conducted based on the explanatory power of the model and regression coefficient as well as a comparison of results for the two models are discussed in Chapter 6.
2. Literature review
2.1. Earnings management and goodwill impairment
2.1.1 Earnings management: definition, methods, factors and incentives
In the existing literature, there are many different insights in defining earnings management. The term earnings management is associated with opportunism and deception.(Schipper, 1989; Stolowy & Breton, 2004)
Healy and Wahlen (1999) state that “earnings management occurs when managers use judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about underlying economic performance of the company or to influence contractual outcomes that depend on reported accounting numbers” (p. 368).
Additionally, Leuz, Nanda and Wysocki (2003) argue that “incentives to misrepresent firm
performance through earnings management arise, in part, from a conflict of interest between firms’ insiders and outsiders” (e.g. controlling largest shareholder and minority shareholders). (p.506). Schipper (1989) defines earnings management as behavior that occurs when managers intervene in the financial reporting process to reach their own gains (e.g. maximize compensation and benefits). Ronen and Yaari (2008) categorize earnings management into “white”, “gray” and “black”.
Earnings management is taking advantage of the flexibility in the choice of accounting treatment to signal the manager’s private information on future cash flows. (p. 25, definition of the “white” earnings management).
Earnings management is choosing an accounting treatment that is either opportunistic
(maximizing the utility of management only) or economically efficient (p. 25, definition of the “gray” earnings management).
Earnings management is the practice of using tricks to misrepresent or reduce transparency of the financial reports (p. 25, definition of the “black” earnings management).
Following Ronen and Yaari (2008), the gray earnings management refers to accounting decision discretion allowed by accounting standards, when firms' insiders have a possibility to manage their firm’s economic performance within Generally Accepted Accounting Principles (GAAP).
Researchers distinguish two types of earnings management, namely accrual and real, depending on possible manipulations through either the accruals or the cash flow component of earnings.
The other classification provided by Dechow and Skinner (2000) includes four groups of earnings management methods based on the subsequent effect on the financial figures: conservative
accounting, neutral earnings, aggressive accounting and fraudulent accounting.
The first three methods do not violate GAAP and are considered legal. Conservative accounting includes such accounting practices as an overly recognition of reserves or provisions, overstating of charges for restructuring and write-offs on assets. Neutral accounting refers to a neutral operation of the process. Aggressive accounting includes overly aggressive reducing the provisions and reserves. As to fraudulent accounting, it includes different fictitious practices and is considered as a violation of GAAP. In practice, the line between legal and fraudulent accounting is thin.
A large body of archival research examines the incentives to manage earnings. Managers’ incentive
Healy (1985) suggests that short-term bonus incentives are associated with earnings management. Prior research has also documented the association between equity incentives related to managerial ownership and earnings management.(Cheng & Warfield, 2005).
Incentive of controlling ownership
Leuz, Nanda and Wysocki (2003) agree that controlling owners have incentives to manage reported earnings in order to mask true firm performance and to conceal their private control benefits from outsiders. (p. 506).
Institutional factors such as the type of the market (developed or emerging equity market),
prevalence of dispersed ownership structures or concentrated ownership structures, strong or weak investors’ rights, and the level of legal enforcement (efficiency of the judicial system and corruption index) influence on the firm engagement in earnings management.(Leuz, Nanda, & Wysocki, 2003, p. 506)
For the purpose of this research, the earnings management definition of Leuz, Nanda and Wysocki (2003) will be used, considering that incentives to misrepresent firm performance through earnings management arise from conflict of interest between firms’ insiders (managers and controlling owners) and outsiders. (Leuz et al., 2003, p. 506) Thus, the value of benefits is exclusively owned by controlling insiders and is not shared with non-controlling outsiders. For example, in case of family ownership, the minority non-family shareholders could be misled.
2.1.2 Earnings management through goodwill impairment accounting under IFRS
International Financial Reporting Standards (IFRS) 3 (Business Combinations) defines goodwill as an asset which represent the future economic benefits arising from other assets acquired in a business combination that are not individually identified and separately recognized.(LLP, 2015, p. 1190).
Former International Accounting Standard (IAS) 22, which was superseded by IFRS 3 and a revised version of IAS 36 (Impairment of Assets) in March 2004, required that acquired goodwill should be amortized on a systematic basis over its estimated useful life. IFRS 3 and the revised IAS 36 replaced systematic amortization with annual goodwill impairment tests, in order to increase of the financial statement quality and harmonize of recording and financial reporting of business
combination and intangible assets in Europe and the USA, as part of their joint initiative on accounting for business combinations.
The impairment-only approach attracts considerable attention from standards-setters, practitioners and researchers (Abughazaleh, Al-Hares, & Roberts, 2011; Ramanna & Watts, 2012). In particular, the attention focuses on the element of managerial discretion inherent in the impairment test of goodwill, and whether firms use this discretion to manipulate the write-off (Abughazaleh et al., 2011; Ramanna & Watts, 2012).
Further, I will discuss why impairment-only approach might create an opportunity for earnings management.
Goodwill impairment testing
In accordance with IFRS 3, a firm shall measure goodwill at the amount recognized in the business combination less any accumulated goodwill impairment losses (IFRS 3:B63).
IAS 36 states that for the purpose of goodwill impairment testing, the acquirer shall allocate the goodwill to each of its cash-generating units (CGUs) or groups of CGUs that are expected to benefit from the synergies of the business combination. A CGU to which the acquirer has allocated
goodwill shall be tested for impairment annually and whenever there is an indication that the CGU may be impaired. The impairment test involves comparing the carrying amount of the CGU with the recoverable amount (IAS 36:90). If the carrying amount of the CGU exceeds its recoverable
amount, the firm shall recognize the excess as an impairment loss (IAS 36:90). The impairment loss shall first be allocated to reduce the carrying amount of any goodwill allocated to the CGU and then to the other assets of the CGU (IAS 36:104).
Carrying amount
IAS 36 defines carrying amount as the amount at which an asset is recognized after deducting any accumulated depreciation and accumulated impairment losses (IAS 36:6).
Recoverable amount, fair value and value in use
IAS 36 defines recoverable amount as the higher of an asset’s or CGU’s fair value less costs to sell and its value in use (IAS 36:18). IAS 36 defines fair value less costs to sell as the amount obtainable from the sale of an asset or a CGU in an arm’s length transaction between knowledgeable, willing parties, less the costs of disposal (IAS 36:6). The best evidence of an asset’s fair value less costs to sell is a price in a binding sale agreement in an arm’s length transaction, less the costs of disposal (IAS 36:25). If there is no binding sale agreement but an asset is traded in an active market, fair value less costs to sell is the asset’s market price less the costs of disposal (IAS 36:26). If neither a binding sale agreement nor an active market exist for an asset, fair value less costs to sell shall be based on the best information available to reflect the amount that the firm could obtain from the disposal of the asset in an arm’s length transaction between knowledgeable, willing parties, less the costs of disposal (IAS 36:27).
IAS 36 defines value in use as the present value of the future cash flows expected to be derived from an asset or a CGU (IAS 36:6). Calculating value in use involves estimating the future cash inflows and outflows to be derived from continuing use of the asset and from its ultimate disposal, and applying the appropriate discount rate to those cash flows (IAS 36:31). Cash flows are based on the management’s most recent approved budgets and forecasts.
Difficulties and discretionarily
It is necessary to notice, that value in use takes entity-specific elements into account, while fair value less costs to sell is based on general use valuation estimates. Bollman and Joest (2010) state that the rationale behind annual goodwill impairment tests is that the assumptions and estimates involved in measuring goodwill need to be reviewed periodically because of the high level of uncertainty. For example, for projecting cash flows for the value in use purpose, managers are required to examine differences between prior cash flow projections and actual cash flows. Moreover, measuring fair value less costs to sell for CGU with goodwill can also be difficult. If there is no binding sale agreement, fair value less costs to sell should be determined in accordance with the market price. On the other hand, there is no active market for CGUs with goodwill. This leads to such situations when fair value less costs to sell for a CGU with goodwill will have to be based on the best information available criterion in IAS 36:27. However, it might be difficult to find objective and relevant information on which to base the measurement of fair value less costs to sell.
Consequently, recoverable amounts for CGUs with goodwill will often be based on value in use instead of fair value less costs to sell. Thus, value in use may often be the only viable option for determining recoverable amounts for CGUs with goodwill.
This creates an opportunity for earnings management in the sense that firms might try to take advantage of the entity-specific valuation procedure involved in estimating value in use by making opportunistic estimates of future cash flows and discount rates.(Storå, 2013).
Based on the discussion above, it seems that IFRS provide firms with an opportunity to account for goodwill based on the entity-specific valuations. The verifiability of this valuation is low as it
involves uncertainty. Because of it, firms have significant discretion over IFRS goodwill impairment accounting. It gives rise to a relatively high level of subjectivity in the impairment test (Lemans, 2010, p. 106) The prior studies also agree that there is existence of managerial discretion in the impairment test of goodwill.(Abdul Majid, 2015; Abughazaleh et al., 2011; Bollman & Joest, 2010; Korošec, Jerman, & Tominc, 2016; Lemans, 2010; Ramanna & Watts, 2012)
Previous studies have suggested that the managers and controlling owners may move future write-offs to current periods or move current write-write-offs to future periods to manipulate their firms’ current earnings. Goodwill impairment decision affects earnings and provides information about decline in the asset values, thus sharing some private managerial information about a firm’s future profitability with the outside world. Unnecessary write-offs by means of goodwill impairment are therefore used as an earnings management tool.
Ramanna and Watts (2012) find that reporting incentives influence the reporting of the impairment losses and goodwill impairment is used as a tool of the earnings management.
Majid (2015) finds that managers’ reporting incentives are significantly associated with the reported impairment losses. Majid (2015) reveals also that this effect is moderated by the ownership
concentration factor.
2.2 Family ownership and agency theory
Level of ownership concentration influences on the firms’ earnings management engagement (Gilson, 2006). In family firms, concentrated ownership is in the hands of a controlling family. Family firms have unique characteristics that could explain the difference expected in accounting and accountability practice from non-family firms (Anderson and Reeb 2003; Gomez-Mejia et al. 2011):
- involvement of the family in the governance of the company; - interest in long-term survival;
- very often close relationship between management and the family;
- high relevance of the non-economic factors (reputation, emotions, duality of firm and family social capital).
In accounting literature, four theoretical frameworks are used within family firms’ context: agency framework, stewardship theory, Resource-Based View (RBV) of the firm and Socioemotional Wealth (SEW) approach (Carrera, 2017). Agency theory is the dominant theoretical framework. Research based on the agency theory has two opposing views about the association between family control and the quality of accounting decisions. The interest of the controlling family in the long-term survival of the firm to supervise and monitor managers may lead to better accounting decisions, by reducing managerial interest in reporting information that deviates
from such firms’ underlying performance, which in turn will lead to higher reporting quality, (the alignment effect, mitigation of a Type I owner-manager agency conflict) (Carrera, 2017). Hence, family ownership concentration is associated with increased shareholder monitoring of management. The interest-alignment effect results in higher earnings quality and in fewer earnings management practices.
However, the controlling family may manipulate financial performance information to expropriate the wealth of other (minority) shareholders (Carrera, 2017). From this perspective, the presence of a controlling family is expected to be associated to accounting decisions of lower quality (the
entrenchment effect, increasing Type II owner-owner agency conflicts): the conflict is between controlling owners (family) and minority shareholders.
Thus, on one hand, in family firms, the mitigation of owner-manager agency conflict limits the management’s interest in manipulation write-offs by means of goodwill impairment for private benefits. On the other hand, the family entrenchment effect might result in the use of write-offs by means of goodwill impairment to manipulate the current earnings and expropriate wealth from the minority shareholders.
3. Motivation and hypotheses development
Family owners are often the firm managers, or they directly supervise their firm management thus ensuring a substantial alignment of interest between shareholders and their managers. In addition, family owners often select such managers because of their personal relationships.(Greco,
Ferramosca, & Allegrini, 2015) Such close ties prevent managers’ opportunistic behavior and the expropriation of shareholder wealth.
Besides different forms of earnings management, different types can be also distinguished in practice. Big bath accounting and income smoothing are the examples of the earnings management use to decrease or increase the earnings of the firm.
According to Mohanram (2003), big bath accounting is used by firms that cannot achieve their targets in a year. When these firms miss their targets they engage in accounting methods to make the firm’s results even worse. Two reasons for this argumentation are given. At first it is very unlikely that the firm can reach the targets set for that year, implying the year is “lost”. Secondly, the costs arising from missing the targets are incurred anyway. The costs the firm will incur from performing even worse will be minimal since the biggest damage is done by missing the targets. The extra incurred losses can be used to increase or smooth income in a future year.
By virtue of income smoothing the management wants to report a consecutive line of increasing earnings.
In order to achieve this, earnings management that both increases and decreases income can be used. If the firm’s income is higher than a target, income can be decreased by using earnings management (also called cookie-jar accounting). As Mohanram (2203) points out, this kind of accounting has two purposes. The first purpose is to save some income for future when the firm may not be able to meet its targets. The earnings from the previous periods are used later. Earnings management can then be considered as an intertemporal transfer of income between periods. The second purpose of
decreasing theincome (if income is higher than it was targeted) is to prevent expectations about the firm to rise.If the expectations about future firm earnings increase, future targets will be more difficult to reach. The consequence of this can be that the consecutive line of increased earnings ends because of one exceptional good result.
Smoothing earnings in case of family firms
Managers may undertake unnecessary write-offs through goodwill impairment in periods of positive pre-impaired earnings performance; that is, when the level of earnings before any write-off
The positive pre-impaired earnings refer to the current performance regardless of the managers’ belief that there will be a future need for write-offs. By anticipating goodwill impairment the
management achieves multiple objectives: to maximize their bonus in future years while keeping the current profits above the threshold specified in earnings-based bonus plans, and to decrease the volatility of earnings (management engagement in signaling activities).
A mitigated Type I agency conflict means that family firm managers are less interested in undertaking unnecessary write-offs in the periods of good pre-impaired earnings performance to maximize their remunerations.(Greco et al., 2015) Furthermore, family firm managers are less interested in using write-offs as a signaling activity to the firm shareholders.
A Type II owner-owner agency conflict suggests that the controlling family may be entrenched and use goodwill impairment to manage their earnings and expropriate wealth from the minority
shareholders. The family owners’ entrenchment can motivate the use of such write-offs for earnings management purposes. Entrenched family owners can use write-offs (e.g. goodwill impairment) as an income-decreasing device to conceal related-party transactions, the extraction of private rents and other wealth-expropriation activities. (Greco et al., 2015). However, the sustained presence of the family makes reputation a key asset for family firms.(Martin et al., 2016)
Thus,
Hypothesis 1: The positive association between goodwill impairment and a positive pre-impaired
earnings performance is stronger in nonfamily firms than in family firms.
Big bath in case of family firms
If the pre-impaired earnings performance is negative, managers cannot reach the earnings target set in their bonus plans. In such cases the managers may anticipate write-offs to current periods thus “saving” future years from write-offs and maximizing their long-term bonus payments and stock related remunerations. (Greco et al., 2015). In family firms, a mitigated owner-manager agency conflict lessens the managers’ remuneration concerns and their interest in moving future write-offs during a periodof negative pre-impaired earnings (Greco et al., 2015).
The close control over the firm performance and the reduced owner-manager information asymmetry (Type 1) lessen such signaling activities’ appeal to family firm managers.
Entrenched family owners may undertake write-offs (e.g. goodwill impairment) during a period of the negative pre-impaired performance and put a blame for this on the economic trend. (Greco et al., 2015). They might hide firm-specific reasons for the decline in their assets’ value. Greco,
to significant reputational costs for family firms. The family reputational capital is a critical asset, thus the need to protect both the firm and the family reputation limits write-offs’ manipulation if the pre-impaired earnings performance is negative (Greco et al., 2015; Martin et al., 2016).
Based on the above, I expect that compared to family firms, non-family firms are more prone to undertaking larger write-offs through goodwill impairment when their negative pre-impaired earnings decrease.
Thus,
Hypothesis 2: The negative association between goodwill impairment and negative pre-impaired
4. Research methodology, sample selection and data sources
Based on the above discussion of an empirical evidence for the use of earnings management and goodwill impairment being used as a tool for earnings managementand the impact of ownership concentration, it is now possible to develop the empirical part of the reasearch.
An important model for this research was developed by Van de Poel et al (2008). Their model was especially developed to investigate whether reporting incentives played a role in the goodwill impairment decision. Their model incorporates a number of variables reflecting the reporting incentivesBig bath accounting and Income smoothing, as well as variables reflecting the economic conditions of the firm (sale, cash flows and industry rates) that also control some firm-specific aspects like the firm size. Since so many factors are incorporated, this model is an appropriate model to use for the purpose of this research since the impairment decision is complex and such factors as included in this model can all influence the impairment decision.
Greco, Ferramosca and Allegrini’s model (2015) is similar to the model created by Van de Poel et al but it takes into account the family ownership concentration.
Both models are combined in this research.
To test the hypotheses, the two regression models are developed to include firm-specific factors as well as proxies for income smoothing and so called “big bath” accounting.
Model 1
IMPAIRMENT (it)=a0+ a1 NONFFit + a2*PosPefDUMit+ a3 NegPefDUMit+ a4 SALESit+ a5
CASHFLit+ a6 MTOBit+ a7 LEVit + a8 SIZEit+ a9 Legal enforcementit+ a10 GOODWILLit + a11*PosPefDUMit * NONFFit + a12 NegPefDUMit * NONFFit +eit
Model 2
IMPAIRMENT_AMOUNT (it)= a0+ a1 NONFFit + a2*PosPefit+ a3 NegPefit+ a4SALESit+ a5
CASHFLit+ a6 MTOBit+ a7 LEVit + a8 SIZEit+ a9 Legal enforcementit +a10 GOODWILLit + a11*PosPefit * NONFFit ++ a12 NegPefit * NONFFit +eit
Dependent variables
When filling in amounts in such an analysis it is however necessary to control for the size of the firm so that the relative size of the impairment instead of the absolute amount is being measured. Otherwise the results could be distorted. Therefore, the dependent variable
IMPAIRMENT_AMOUNT (it) will be measured as the reported impairment amount deflated by total assets at the end of year t-1.
The dependent variable is IMPAIRMENT (it) representing the impairment decision. It will be
measured as a dummy variable (Lemans, 2010, p. 112)
However, this method disregards the fact that the amount of the reported impairment can also be an important issue compared to only the decision to report an impairment loss or not (Lemans, 2010, p. 112).
Therefore, the second independent variable will be IMPAIRMENT_AMOUNT (it) which will be
measured as the reported impairment amount deflated by total assets at the end of year t-1.(Lemans, 2010, p. 112)
Independent variables
The first two variables are related to reporting incentives by the management. The way of their inclusion in the model depends on the use of the dependent variable as discussed above. When the regression is performed by using IMPAIRMENT (it) as a dummy variable, the variables (PosPef)
and (NegPef) will also be measured as dummy variables. This is consistent with the model of Van de Poel et al. (2008) that has proven that these variables have a significant influence on the
impairment decision at the 5% and 1% level, respectively.
1. Positive pre-impaired earnings performance (PosPef) (proxy for income smoothing) – positive
current firm performance before goodwill impairment. It will be measured as a change in the firm’s pre-impaired earnings from period t to t − 1, divided by the assets total at the end of period t − 1 if above the median of nonzero positive values, 0 otherwise. When the regression is performed by using IMPAIRMENT dummy variable, the variable PosPef will be a dummy variable. The variable PosPefDUM is included in the model to test Hypothesis 1 and will have a value of 1 if the change in the firm’s pre-impaired earnings from period t to t − 1, divided by the assets total at the end of period t − 1 is above the median of nonzero positive values, and 0 otherwise. In this case, earnings
are unexpectedly high. Therefor this variable is used as a proxy for income smoothing (earnings management).
2. Negative pre-impaired earnings performance (NegPef) (proxy for big bath accounting) is a
negative current performance the firm had before goodwill impairment. It will be measured as a change in the firm’s pre-impaired earnings from period t to t − 1, divided by the assets total at the end of period t − 1 if below the median of nonzero negative values, 0 otherwise. When the
regression is performed by using IMPAIRMENT dummy variable, the variable NegPef will be a dummy variable. The variable NegPefDUM is included in the model to test Hypothesis 2 and will have a value of 1 if the change in the firm i’s pre-impaired earnings from period t to t − 1, divided by the assets total at the end of period t − 1 is below the median of nonzero negative values, and 0 otherwise. In this case, earnings are unexpectedly low. Therefore, this variable is used as a proxy for the use of big bath accounting (earnings management).
The reasoning for including these two variables in the two models is to test whether specifically the higher or lower level of unexpected earnings has a positive effect on the impairment decision. The reported impairment loss has been deducted from the expected earnings for this variable because the reported impairment amount influences whether the level of earnings can be classified as
unexpectedly high or just high. Only when the earnings are unexpectedly high (in excess of the impairment amount), the reporting incentive of income smoothing could play a role since then the earnings can/need to be smoothed to present a consecutive line of increasing earnings. This
reporting incentive therefore would not play a role when earnings are high like thiswas expected to a certain extent. It is noticeable that write-off has not been deducted from earnings for the
measurement of big bath accounting since then the management is not concerned with a trend in earnings that can be broken.
Variable for creating interaction terms and studying the difference in accounting behavior in family and nonfamily firms.
Non-family firms (NONFF) are a subgroup of nonfamily firms within the sample, used to create interaction terms and study family firms and nonfamily firms’ differing accounting behavior. Dummy, 1− if the firm is classified as non-family, 0 − if classified as a family firm.
Control variables
Control variables represent economic factors, a firm’s specific economic performance and potential determinants of goodwill impairment. The said variables are incorporated as proxies for the
economic conditions the firm is facing.
Change in sales (SALES) − Sales performance, change in total sales from period t to t − 1. This variable is associated with measuring the economic condition of the firm (Van de Poel et al., 2008). This variable has proven to have a significant effect at the 1% level by Vna de Poel et al. (2008) and it measures as the year-to-year change in sales.
Change in the operating cash flow (CASHFL) − Cash flow from the operations performance, change in the operating cash flows from period t to t − 1, divided by the assets total at the end of t – 1. This variable has also proven to have a significant effect at the level 1% (Van de Peol et al., 2008). This variable is included because it is also a firm-specific factor that is associated with the economic condition of the firm, as described previously for SALES. It is therefore expected that a negative relation will be found between this variable and the impairment decision as the recognized
impairment loss will be higher when the firm is performing poor. Also it can be a useful explanatory variable since it is afirm-specific factor which can be used to explain differences between firms and industries when the analyses proves that it has a significant effect on the impairment decision. This is the main reason why this variable is included.
Growth options (MTOB) − Market estimate of the firm’s future growth opportunities. Market-to-book ratio at time t. This variable controls for the economic condition of the firm as related to the market value. The firms having a lot of growth opportunities will be less likely to record goodwill impairment for earnings management purposes. It also indicates the market expectation about the firm’s future performance and whether the firm’s assets are overvalued.
Leverage (LEV) −Amount of debt used to finance the firm’s assets. Firm’s debt to assets ratio at time t. Firms with more debt may be subject to stricter covenants and external lenders’ close control. More indebted firms may record fewer write-offs to avoid deeper scrutiny of their financial health.
On the other hand, firms with higher debt are likely to be subject to external auditors, investors and lenders’ stricter control of their asset value. This may press indebted firms to use goodwill
impairment methods for earnings management purposes to make more adjustments to their assets’ value.
Firm size (SIZE) − The size of a firm. Total assets at time t. This variable is included in the model to distinguish the size type of the firms. This variable has proven to have a significant effect at the 1% level on the impairment decision (Van de Poel et al., 2008). It is expected that there will be a
positive relation between the size of the firm and the impairment decision, meaning that larger firms will be more likely to recognize an impairment loss than smaller firms.
Additionally, the variable Cluster is included in the model. Cluster (Legal enforcement) – the type of legal enforcement inherent to the country (described in the sample selection). A subgroup of cluster 2 firms within the sample, used to create interaction terms and study the effect of institutional legal enforcement effect on the accounting behavior. Dummy, 1 for the cluster 2 firm (weak legal enforcement), 0 for the cluster 1 firm (strong legal enforcement). The European Union countries (continental part) may be divided into two institutional clusters (Amiraslani, Iatridis, & Pope, 2013, p. 26)
Cluster 1 with the following institutional settings: developed stock markets, concentrated ownership, weak investor protection, strong legal enforcement;
Cluster 2 with the following institutional settings: developed stock markets, concentrated ownership, weak investor protection, weak legal enforcement.
Finally, the tenth variable is included as a control variable that has proven to have a significant influence ( 1%) on the goodwill impairment decision (Van dde Poel et al., 2008) is GOODWILL. The reason to include this variable is that a firm with more goodwill as an asset might incur more goodwill impairment because the relative amount of goodwill exposed to the impairment test is greater (Lapointe-Antunes et al., 2008). Therefore the expected relationship between this variable and the impairment decision is positive.
The variable GOODWILL is measured as a ratio of a firm’s opening balance of goodwill on total assets, measures at t-1. Therefore, goodwill deflated by total assets will be used in the model to control for the effect of the goodwill opening balance on the impairment decision.
The following interaction terms have been included in the models developed:
a11*PosPefDUMit * NONFFit (model 1) a12 NegPefDUMit * NONFFit (model 1) a11*PosPefit * NONFFit (model 2) a12 NegPefit * NONFFit (model 2)
Goodwill is likely to have different trends per industry and time. It can be triggered by industry-specific technological changes, regulatory changes or exposure to foreign competition.
This can increase the need to control for unobserved heterogeneity per industry and time. For this reason, a pooled and ordinary least square regression with industry-by year intercepts can be used to test the hypotheses. The regression is run with normal standard errors and with standard errors robust with respect to heteroscedasticity to check whether it can affect the results.
Sample selection
The focus will be brought on the listed EU firms.
The EU countries (continental) may be divided into two institutional clusters (Amiraslani, Iatridis, & Pope, 2013, p. 26)
• Cluster 1 with the following institutional settings: developed stock markets, concentrated
ownership, weak investor protection, strong legal enforcement;
• Cluster 2 with the following institutional settings: developed stock markets, concentrated
ownership, weak investor protection, weak legal enforcement.
In the research, a sample of listed firms from the German Stock Exchange - Boerse Frankfurt (cluster 1) and from the Italian Stock Exchange -Borsa Italiana (cluster 2) is used. The focus will be brought on the period 2006-2017 (starting from 2006 mandatory use of IFRS in the EU). These two countries were chosen because of the ability to take into account the legal enforcement effect (cluster 1 vs cluster 2, strong vs weak),and the Italian Stock Exchange accounts as the largest stock market of the stock family; and around 95% (listed/unlisted) of entities in Germany are controlled by a family (Achleitner et al., 2014, p. 435)
This implies that the total sample consists of 591 firms or 7.092 firm-year observation as gathered through Osiris financial database from Bureau Van Dijk. The year 2017 has also been included as far as is known at this very moment.
The initial sample needs to be adapted to the research setting by selecting only the companies that are active in the research period. This indicates that the inactive companies are excluded, as a consequence of this it is possible that a particular firm will for instance be incorporated only once in the sample.
The sample has also been selected based on the information available. Since many different
variables in the model need to be determined, it has proven that for some firms in the sample not all necessary information is available. A reason for this is that data from two following years are being used for only one firm-year observation.
In addition, data about the family ownership is not published in Osiris. The firms from the sample have been screened through available resource to determine the family ownership and its percentage (hand-collected). The following resources have been used for the screening:
- Official site of a listed firm (annual reports at least for the last 5 years, press releases); - News and data from Bloomberg, Financial Times and other trusted sources;
- Osiris data on shareholder structures and an ultimate beneficial owner.
According to Miller and Le-Breton Miller (2006) (Miller & Le Breton-Miller, 2006), a firm has been classified as a family firm when a family owns 30% or more of the voting shares and is able to appoint the majority of directors to the board, or it is determined as an official Ultimate Beneficial Owner with more than 30% shares in Osiris database.
The next step in selecting the sample is to exclude from the firm-year observations those firm-year observations in which there was no goodwill on the opening balance at the same time that no impairment was recorded (to exclude not-related to goodwill or goodwill impairments.
References
Abdul Majid, J. (2015). Reporting incentives, ownership concentration by the largest outside shareholder, and reported goodwill impairment losses. Journal of Contemporary Accounting &
Economics, 11(3), 199-214. doi:10.1016/j.jcae.2015.07.002
Abughazaleh, N., Al-Hares, O., & Roberts, C. (2011). Accounting discretion in goodwill
impairments: UK evidence. Journal of International Financial Management & Accounting,
22(3), 165-204. doi:10.1111/j.1467-646X.2011.01049.x
Achleitner, A., Günther, N., Kaserer, C., & Siciliano, G. (2014). Real earnings management and accrual- based earnings management in family firms. European Accounting Review, 23(3), 1-31. doi:10.1080/09638180.2014.895620
Ahearne, M. J., Boichuk, J. P., Chapman, C. J., & Steenburgh, T. J. (2016). Real earnings
management in sales. Journal of Accounting Research, 54(5), 1233-1266. doi:10.1111/1475-679X.12134
Amiraslani, H., Iatridis, G. E., & Pope, P. F. (2013). Accounting for asset impairment: A test for IFRS compliance across europe. London, UK: Centre for Financial Analysis and Reporting
Research, Cass Business School.Standards, Regulations, and Financial Reporting, , 199-223.
Bollman, F., & Joest, A. (2010). Impairment testing. Guide to fair value under IFRS (In Catty, J. P.; Hoboken, New Jersey ed., pp. 201-2013) Wiley.
Carrera, N. (2017). What do we know about accounting in family firms? Journal of Evolutionary
Cheng, Q., & Warfield, T. D. (2005). Equity incentives and earnings management. Accounting
Review, 80(2), 441. doi:10.2308/accr.2005.80.2.441
Gilson, R. (2006). Controlling shareholders and corporate governance: Complicating the
comparative taxonomy
Goh, J., Lee, H., & Lee, J. (2013). Majority shareholder ownership and real earnings management: Evidence from korea. Journal of International Financial Management & Accounting, 24(1), 26-61. doi:10.1111/jifm.12006
Greco, G., Ferramosca, S., & Allegrini, M. (2015). The influence of family ownership on long-lived asset write-offs. Family Business Review, 28(4), 355-371. doi:10.1177/0894486515590017
Kaplan, S. (2001). Ethically related judgments by observers of earnings management. Journal of
Business Ethics, 32(4), 285-298. doi:10.1023/A:1010600802029
Korošec, B., Jerman, M., & Tominc, P. (2016). The impairment test of goodwill: An empirical analysis of incentives for earnings management in italian publicly traded companies. Economic
Research-Ekonomska Istraživanja, 29(1), 162-176. doi:10.1080/1331677X.2016.1164923
Lemans, J. D. (2010). Goodwill impairment as a tool for earnings management
Leuz, C., Nanda, D., & Wysocki, P. D. (2003). Earnings management and investor protection: An international comparison. Journal of Financial Economics, 69(3), 505-527.
LLP, E. &. Y. (2015). International GAAP 2016 custom generally accepted accounting principles under international financial reporting standards.
Martin, G., Campbell, J., & Gomez-Mejia, L. (2016). Family control, socioemotional wealth and earnings management in publicly traded firms. Journal of Business Ethics, 133(3), 453-469. doi:10.1007/s10551-014-2403-5
Miller, D., & Le Breton-Miller, I. (2006). Family governance and firm performance: Agency, stewardship, and capabilities. Family Business Review, 19(1), 73-87. doi:10.1111/j.1741-6248.2006.00063.x
Prencipe, A., Markarian, G., & Pozza, L. (2008). Earnings management in family firms: Evidence from R& D cost capitalization in italy. Family Business Review, 21(1), 71-88.
doi:10.1111/j.1741-6248.2007.00112.x
Ramanna, K., & Watts, R. (2012). Evidence on the use of unverifiable estimates in required goodwill impairment. Review of Accounting Studies, 17(4), 749-780. doi:10.1007/s11142-012-9188-5
Schipper, K. (1989). Commentary on earnings management. Accounting Horizons, 3(4), 91-102.
Stolowy, H., & Breton, G. (2004). Accounts manipulation: A literature review and proposed conceptual framework. Review of Accounting and Finance, 3(1), 5-92.
Storå, J. (2013). Earnings management through IFRS goodwill impairment accounting: In the context of incentives created by earnings targets.
Tsao, S., Chang, Y., & Koh, K. (2017). Founding family ownership and myopic R& D investment behavior. Journal of Accounting, Auditing & Finance, , 0148558X1770408.
Wang, D. (2006). Founding family ownership and earnings quality. Journal of Accounting
Research, 44(3), 619-656. doi:10.1111/j.1475-679X.2006.00213.x
Zang, A. Y. (2012). Evidence on the trade-off between real activities manipulation and accrual-based earnings management. The Accounting Review : A Journal of the American Accounting
