Printer Friendly

Bidder earnings management, cynical targets and acquisition premia.

Introduction

Like any other corporate investment decision, a merger and acquisition (M&A) proposal must pass through extensive ex-ante analyses for the costs and benefits of the deal from the perspective of the acquiring firm as well as the target. In particular, an acquiring firm must determine the maximum price it can feasibly offer to the shareholders of a potential target without destroying the wealth of bidder shareholders. In particular, bidders in non-cash acquisitions have an incentive to make their shares as attractive as possible to target managers and shareholders to incent them to take the bid rather than negotiating for a higher price. One technique which bidders employ is to manage earnings immediately prior to an acquisition in order to obtain a higher value for their shares.

The literature has documented that earnings data are considered value relevant and informative from the investors' perspective in general (Dechow 1994). The high value-relevance of earnings data for predicting future returns underlies the motivation of managers to manipulate their firms' earnings preceding specific events, such as seasoned equity offerings (SEOs), initial public offering (IPOs) or M&A deals. Evidence of pre-merger earnings management has been repeatedly documented in the literature (Erickson and Wang 1999; Louis 2004; Botsari and Meeks 2008; Gong, Louis et al. 2008; Yung, Sun and Rahman 2013; Higgins, 2013). They argue that managing the acquirer's earnings upwards does usually inflate its share price, therefore lowering the cost of the deal from the acquirer's perspective. Such empirical evidence indicates that managers often succeed in their attempts to influence the reported numbers on paper, with the aim of reducing the firm's cost of capital. However, this evidence per se does not necessarily indicate that the users' perceptions, especially those of the well-informed, are also successfully manipulated.

Surprisingly, the validity of the naive investor model has been challenged in IPO and SEO contexts but not within an M&A context. The earnings management research surrounding M&A events implicitly adopts the naive investor model to explain the reportedly positive evidence of pre-merger abnormal accruals in order to back the proposition that reducing the cost of capital is the managerial motivation (Erickson and Wang 1999; Asano, Ishii et al. 2007; Botsari and Meeks 2008).

This paper addresses the question of how earnings management undertaken by a firm is perceived and processed by relevant parties and whether the naive investor hypothesis can be extended to target managers. We investigate both cash and stock acquisition premia and find that bidders who manipulate earnings are unable to obtain a lower acquisition price using overvalued shares in stock deals. Our results support the "cynical investor hypothesis," which posits that target management and their financial advisors are suspicious of bidders with low earnings quality and are able to detect and thwart bidder earnings management schemes designed to obtain a lower acquisition price in stock transactions.

The paper proceeds as follows. In section 2, we review the relevant literature and develop our hypotheses. In section 3, we describe our data and empirical methods. In section 4, we present our results, and section 5 concludes.

Literature Review and Hypotheses

INFORMATION ASYMMETRY AND M&A PREMIA

An M&A deal is initiated in a market where buyers (i.e., acquirers) seek to buy the assets of sellers (i.e., targets). On each side of the transaction (e.g., the buyer's side and the seller's side) there is one agent (e.g., the management) acting on behalf of passive owners with the motivation of maximizing the owners' wealth (Myers and Majluf 1984). However, the efficiency of managers in decision making creates an agency problem between the current and the new shareholders every time new equity issues are decided upon by a given firm (Jensen and Meckling 1976). This means that management's decisions would favor current shareholders' interests over those of the new.

The lemons' problem theory of Akerlof (1970) suggests that in the M&A market, a potential acquirer assumes that a target firm will accept a cash offer only if it finds, according to its proprietary information, that its real value is lower than the offer received. Due to information uncertainty, the acquirer is likely to discount the target's value to hedge itself against adverse selection risk.

However, the scenario is bi-directional (i.e., dual direction of information asymmetry) when an acquiring firm offers equity shares to its target since equity has a contingent-pricing effect. The target firm must also deal with information uncertainty regarding the true value of the shares offered by an acquirer (Eckbo, Giammarino et al. 1990). Since the target is exposed to the risk of adverse selection (e.g., receiving overvalued shares), it is likely to discount the acquirer share's value. This situation is described by Hansen (1987) as a double lemons' problem. Assuming the lemons and double lemons problems in the M&A market, Hansen's (1987) theory for the choice of method of payment in M&A suggests that the acquirer offers cash when its shares are undervalued while it offers equity shares only when they are overvalued. According to this theory, a target firm can use the method of payment in M&A as a signal to learn about the fairness of the acquirer's share value.

From the acquirer's perspective, the awareness of the value-discounting strategy adopted by targets receiving offers that include equity shares presents a motivation for the acquirer's management to inflate premerger earnings and influence its share price before offering shares that are going to be eventually discounted during the M&A transacting process.

Therefore, information asymmetry is perceived by both sides of an M&A deal as each party possesses proprietary information on its own firm's value. The amount of information known about a firm and its earnings limits the extent of its ability to engage in earnings management. Therefore, the motivation, as well as the effectiveness of earnings

management, cannot be expected when information is perfectly symmetric between the acquiring firm and other stakeholders (Richardson 2000). Although agents bargain under imperfect informational conditions, they are well-informed users whose resources and access to information is superior to average users'.

EARNINGS MANAGEMENT, ACCRUALS MISVALUATION AND M&A PREMIA.

Evidence suggests that one reason bidder earnings management may mislead targets is that investors may be prone to a behavioral phenomenon known as accruals mispricing (Sloan 1996; Xie 2001; Fairfield, Whisenant and Yohn, 2003; Kraft, Leone and Wasley 2006). There appears to be a systematic bias in the ability of investors to comprehend the persistence of accruals and accrual components relative to cash flows, and thus, they tend to underreact (e.g., overvalue) to accruals. Sloan (1996), in the naive investor model, suggests "that investors 'fixate' on earnings, failing to distinguish fully between the different properties of the accrual and cash flow components of earnings." This view predicts that investors respond to the managed earnings, which contains abnormal accruals, as if this level of earnings would persist in the future. However, in reality, accruals have a transient effect on earnings among accounting periods, and abnormal accruals are likely to reverse in a future period (Dechow, Sloan et al. 1995; Chung, Firth et al. 2002).

Earnings management and investor sophistication has been shown to impact valuation in a number of contexts. From a sample of initial public offerings (IPO), DuCharme et al. (2001) examine how pre-offering earnings management impacts the offering pricing and post-offering performance. They find that investors are unable to distinguish between managed and unmanaged accruals, which allows managers to affect IPO pricing through managing accounting accruals. Ashiq-Ali et al. (2000) adopt a methodology that focuses on the use of proxies to measure the level of investor sophistication in contrast to the approach of investor naivety. They argue that given that naive investors fixate on earnings, which leads to an inverse relationship between the accrual component and future stock performance, the magnitude of such a relationship should be negatively associated with the proportion of sophisticated investors' interest in the firm, where the investor's sophistication is proxied by the institutional ownership. The results of Ashiq-Ali et al.'s (2000) study are counter to the naive investor proposition and its predictions that the inverse relationship between accruals and future stock performance is stronger for stocks with greater institutional ownership (i.e., investor sophistication.) Likewise, Balsam et al. (2002) use institutional ownership in order to proxy for the investor's sophistication and report evidence of a negative association between the abnormal accruals in the quarterly earnings and the cumulative abnormal returns around the statements' filing date. This indicates that the accrual component of earnings has valuation consequences for sophisticated investors.

In the M&A context, several studies document the tactic of bidders reporting abnormally high accruals, which boost earnings before announcing takeover attempts (Erickson and Wang 1999; Louis 2004; Botsari and Meeks 2008; Gong, Louis et al. 2008; Fisher and Louis 2011). The capital markets incentive approach posits that acquiring firms have incentives to manage accruals upward before offering shares in a M&A deal, with the intention of inflating the share price in order to issue less shares and, therefore, reducing the cost of capital used to discount cash flows from the acquisition. Prior empirical studies indicate significant evidence of premerger abnormal accruals, which are positively associated with the relative size of the deal, implying the economic incentives of earnings management (Erickson and Wang 1999; Louis 2004; Gong et al. 2008). It is intuitive that there are "capital market incentives" to inflate share prices in light of the acquirer-target information asymmetry and the agency problems between current and new equity-holders (Jensen and Meckling 1976; Jensen 1986). However, the argument does not fully describe how or whether new equity-holders (i.e., the target's shareholders or their agents) respond to this agency problem or how to deal with the over-reporting of accruals by the acquirer, which implicitly assumes naivety on the part of the target's management.

Louis (2004) provides evidence of post-merger underperformance in his examination of the impact of earnings management on the acquirers' performance. In explaining this anomaly, he concludes that the reversal effect of accruals is an important determinant of the post-merger market reaction and the stock's performance. Beyond the stock market's financial reaction, Gong et al. (2008) warn that potential post-merger litigation costs are likely to be a consequence for those acquiring firms that have adopted income-increasing accounting procedures prior to the M&A activity. They report a positive relationship between pre-merger earnings management and the probability of post-merger litigation.

As a counter argument to the naive target proposition, this study argues that pre-merger earnings management may not be an effective tool to mislead a well-informed participant. Assuming that a target firm is sophisticated enough to recognize the transience of acquirer's abnormal accruals, the non-cash acquisition premium, which is the final product of the bargaining process in an M&A deal, should reflect the target's adjustment to the acquirer's share value. Thus, low bidder earnings quality provides a signal that the cynical investor would detect regarding the price offered, and the target would negotiate a higher premium in the presence of bidder earnings management in stock deals.

This leads us to our first hypothesis:

H1: THERE IS A SIGNIFICANT POSITIVE ASSOCIATION BETWEEN THE ACQUISITION PREMIUM AND THE MAGNITUDE OF EARNINGS MANAGEMENT IN NON-CASH M&A.

If Sloan's (1996) naive investor perspective is correct and adequately describes target managers as well as institutional investors, we would observe that bidders in stock deals are able to use earnings management tactics to make their shares appear overvalued in order to achieve a lower bid price and that the impact of earnings management on premia for stock deals will be significantly more pronounced than in cash deals. We would observe significantly lower acquisition premia for stock deals (or more stock intensive deals) than cash deals. If managers are cynical investors, wary of potential bidders and their schemes and cognizant of appropriate accruals valuation, then we would expect to find that either bidders who manage earnings in stock deals will either pay a premium relative to cash (or less stock intensive) deals or that there is no significant, incremental difference. Therefore, our second hypothesis is as follows:

H2: THE INCREMENTAL IMPACT OF THE MAGNITUDE OF BIDDER EARNINGS MANAGEMENT AND THE ACQUISITION PREMIUM FOR STOCK TRANSACTIONS RELATIVE TO CASH TRANSACTIONS IS GREATER THAN OR EQUAL TO ZERO.

Data and Methods

DATA

Sampling Procedure: We obtain sample data from several sources. A sample of US acquiring firms is taken from Thomson One Banker according to the following criteria:

1. The M&A deals are announced between 01/01/1999 and 12/31/2008.1

2. All merger and acquisition deals included are completed transactions.

3. The acquiring firms and their respective targets are publicly listed companies in order to mitigate variation in information asymmetry and earnings management motives (Baik, et al. 2007).

4. All deals including either acquiring and/or target firms from the financial sector which have SIC codes between 6000 and 6999 are excluded from the sample. This is a common practice in the literature as this sector is subject to special regulations (Erickson and Wang 1999; Gong, Louis et al. 2008; Meisel 2010).

5. The deal value is at least $1 million to exclude all deals of negligible size. The purpose of this procedure is to discard deals where firms have a lower motivation to manage earnings due to the insignificant economic motivation to do so (Erickson and Wang 1999).

6. The deal should result in allowing the acquiring firm to obtain a controlling ownership interest in the deal (i.e., the acquirer owned less than 50% before the transaction and greater than 50% by completing the deal) so that the acquirer is more likely to offer an acquisition premium to the target (Arzac 2004).

7. The deals have acquisition premium data available on Thomson One Banker and earnings management data on Compustat.

Applying these criteria results in a total of 420 M&A deals where 263 of them represent a sample of non-cash deals, which are later used for assessing the research hypothesis, and the remaining 157 are cash deals used in a robustness check.

For acquisition-related data and firm characteristics, we use Thomson ONE Banker and "S&P Research Insight." In order to calculate the vertical relatedness of the deals (VRLTDi), we used the "Input-Output Accounts" from the industry data from Survey of Current Business on the website of US Department of Commerce in order to find "The Use of Commodities by Industries" tables. (2) Since industry classification used in the tables follow the North American Classification System (NAICS), the codes are converted into the Standard Industrial Classification (SIC) system using the correspondence tables between NAICS and SIC as provided by the US Census Bureau. (3)

VARIABLE DESCRIPTIONS

The premium (P[R.sub.i]) is calculated as:

[PR.sub.i] = Offer Price/Base Price - 1 (1)

Ideally, the base prices used should be those that are the closest to the deal announcement date but those that are not affected by the potential event-specific informational leakage. Schwert (1996) provides empirical evidence suggesting that the market on average anticipates a M&A deal 21 days before its official announcement. In accordance with Schwert's (1996) findings and following Porrini (2006), acquisition premium (PRi) data in this study is obtained from the Thomson ONE Banker database which calculates the premium using base prices four weeks prior to the announcement date. Relevant databases, including Thomson One Banker, do not distinguish between the acquisition premium of cash deals and those of non-cash deals. The formula in Equation (1) is normally used for calculating both types of acquisition premia. Therefore, PRi in a non-cash deal indicates the nominal acquisition premium as Thomson One Banker uses the pre-merger, publically available prices of equity shares of the acquirer for calculating the non-cash M&A offer value (4).

ACQUIRER EARNINGS MANAGEMENT

The accrual-based models of measuring earnings management have been exceedingly adopted in the literature, up to the point where the term "earnings management" per se, unless otherwise indicated, has become an implicit reference to the accrual-based type (Schipper 1989; Teoh, Welch et al. 1998; Dechow, Richardson et al. 2002; Lee, Kim et al. 2008; Grasso, Tilley et al. 2009; Liu, Ning et al. 2010). The acquirer's earnings management is measured using the signed value of the cumulative abnormal accruals calculated from quarterly data within a three-quarter window prior to the M&A announcement date (Alsharairi and Salama 2012). (5)

Similar to the work of Kothari et al.'s (2005), our accruals estimation approach considers the performance effect by calculating the expected accruals on Dechow et al's (1995) modified Jones (1991) model in a cross-sectional estimation for the industry-performance matched portfolios in each quarter, following the recommendation of Louis (2004) and Gong et al. (2008). More specifically, the current accruals are computed using the changes in the non-cash working capital following the balance sheet method (6) (Pungaliya and Vijh 2008) as follows:

CA[C.sub.i] = [DELTA]C[A.sub.i] = ([DELTA][CL.sub.i] - [DELTA][STD.sub.i])-[DELTA][CASH.sub.i]. (2)

Where:

CAC: denotes the current accruals,

[DELTA]CA: is the quarterly change in current assets (Compustat XPF mnemonic (7) code ACTQ),

[C.sub.i] CL: is the quarterly change in current liabilities (mnemonic code LCTQ),

[C.sub.i] CASH: is the quarterly change in cash (mnemonic code CHEQ), [DELTA]STD: is the quarterly change in current maturities of long term debt and other short term liabilities included in current liabilities (mnemonic code DLCQ and

i: denotes the acquiring firm's index.

A cross-sectional industry-performance-matched accruals model is used in this study, similar to the research design of Louis (2004) and Gong et al. (2008). The core of this model emanates from the work of Dechow and Sloan's (1995) modified Jones' (1991) model after considering Kothari et al.'s (2005) non-linear control for performance. (8)

The industry-performance matching procedure is achieved in this model by building matching portfolios using the universe of firms in each quarter. More specifically, data of all firms available on Compustat are clustered by calendar years and quarters. In each quarter, all firms are categorized into industry sectors based on their 2-digit SIC. In each industry, all firms are ranked according to their performance--defined as ROA of same quarter last year--to form five quintiles.

Before ranking firms' portfolios into quintiles, three procedures are followed for stronger robustness and to reduce measurement error at this stage (Gong, Louis et al. 2008): discarding the universe outliers represented by observations that have the highest and the lowest 0.1% ROA, dismissing each observation with the absolute value of current accruals divided by lagged total assets greater than one ([absolute value CA[C.sub.j/[TA.sub.j-4]]>1) to reduce the likelihood of including observations with extreme values due to improper data entry in the database and, finally, excluding portfolios with less than 20 observations. The matching procedure leads to the creation of five performance-matched portfolios per industry per quarter for each year of data. Each portfolio of peer firms is used as a firm's control in order to estimate the parameters that are used in calculating the expected current accruals for each firm in the same portfolio.

Therefore, the following cross-sectional model is estimated for each portfolio constructed by the aforementioned procedure:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Where:

[Q.sub.q]: is a dummy variable to control for seasonality, takes 1 if the deal is announced in quarter q prior to merger announcement and 0 if the otherwise;

[DELTA]REV: is the quarterly change in sales (code REVTQ);

[DELTA]AR: is the quarterly change in accounts receivables (RECTQ);

PPE: is the gross amount of property, plant and equipment in a quarter (PPENTQ);

[TA.sub.j-4]: denotes the total assets in the same quarter last year (ATQ); and

[euro]: denotes the residual term of the regression model.

To reduce potential heteroskedasticity in residuals, all variables are scaled by the total assets in the same quarter last year as a deflating procedure, following the recommendation of Kothari et al. (2005).

This study controls for a number of variables that may exert an impact over the level of the acquisition premium while assessing the relationship between [PR.sub.i] and [EM.sub.ACQi]. These variables include the pre-merger performance of both the acquirer ([ROE.sub.ACQ]) and the target ([ROE.sub.TGT]) as calculated by dividing their respective EBIT by total equity for the last 12 months prior to the M&A announcement date. Bugeja and Walter (1995) argue that acquirers with a better performance prior to a M&A deal are more likely to offer higher acquisition premia since this may indicate that the firm's management has a greater ability to successfully combine the two firms into one, creating higher post-merger value for the shareholders involved. [ROE.sub.TGT], however, is expected to be negatively associated with the acquisition premium because it is argued that the acquirer is less likely to be able to create greater value through acquiring a previously well-managed target firm (Haw, Pastena et al. 1987; Bugeja and Walter 1995).

Certain target-related characteristics may lead some targets to be better informed than others and to detect bidder earnings management. Erickson and Wang (1999) advise the acquirer's management team to consider the M&A target firm as a well-informed user that may also hire advisors to encounter the existing agency problem and reduce information asymmetry. Therefore, the risk of being caught with pre-merger earnings management is a cost that should not be ignored, and acquirers are cautioned to choose to manage earnings unless the cost of undoing earnings management outweighs the cost of undertaking. The model also includes a variable denoted [DBT.sub.TGT], which represents the target's debt ratio as a proxy for its capital structure, following the recommendation of Crawford and Lechner (1996), who assert that firms with lower leverage are more attractive M&A targets

Moreover, the acquirer's pre-merger toehold ownership in the target firm ([TOE.sub.ACQ]) is also controlled for because it may be inversely related to the acquisition premium due to the fact that a bidder with a toehold position in the firm needs to acquire fewer shares in order to obtain a controlling interest. This increases the chance that challenging bidders will enter the competition (Stulz 1988; Bugeja and Walter 1995; Wickramanayake and Wood 2009).

If there is more than one potential acquirer competing for a certain target, then the acquisition premium is likely to be higher (Schwert 2000). Hence, the dummy variable ([CHLNG.sub.i]) is added to the model to capture the effects emanating from a multiple bidder contest. Similarly, the relative size of the M&A deal ([RSIZE.sub.i]) is an alternative important factor since prevailing empirical evidence finds that relatively large deals produce lower post-merger returns for the acquiring firms (Antoniou, Arbour et al. 2008). [RSIZE.sub.i] is a known control variable that has been used in the M&A literature and is usually calculated as the total assets of the target divided by the total assets of the acquirer (Moeller, Schlingemann et al. 2004; Wickramanayake and Wood 2009; Madura and Ngo 2010).

We also control for degree of relatedness between the acquirer and target--that is, whether or not they belong to the same industry. We include a variable ([VRLTD.sub.i]) that indicates the vertical relatedness of the merging firms. The vertical integration of production for two companies is likely to realize economies of scale for the merged firm and to create greater value from the cost reduction (Haunschild 1994). To calculate this variable, we follow Haunschild (1994). By comparing the 2-digit SIC of the acquirer with its respective target, the M&A deal is defined as being a vertical integration if 5% or more of the output of the target's industry is used as input by the acquirer's industry or if 5% or more of the output of the acquirer's industry is used as input by the target's industry.

Finally, the sampled M&A deals are obtained from a ten-year time period wherein the M&A market has witnessed considerable changes in the behavior and valuation of deals, especially since the occurrence of big US corporate scandals and the enactment of Sarbanes-Oxley Act of 2002 (Madura and Ngo 2010). Thus, this study includes a set of year dummies in the model to capture the macroeconomic differences over years for other factors that are not specified by the model (9).

SAMPLE CHARACTERICS

The descriptive statistics and the sample distribution by year are organized in three panels in Table 1 for the overall sample, non-cash deals and cash deals, respectively.

Panel A of Table 1 reveals that 39% (165 deals) of the total 420 deals were announced between 1999 and 2001. Interestingly, when the big corporate scandals engulfed the financial world and when Sarbanes-Oxley Act was enacted in 2002, the statistics show that deals within this year exhibited the highest mean acquistion premium of 114.94%, with the highest standard deviation of 365.32% (for 37 deals). In contrast, the 35 deals undertaken during 2007 reveal the lowest mean acquistion premium of 29.789%, as well as the lowest standard deviation of 28.84%. Overall, the full sample has a mean acquistion premium of 51.103%, which is consistent with previously reported average acquistion premium levels in the literature.

Panels B and C of Table 1 describe the acquistion premium by year for non-cash deals and cash deals, respectively. It can be seen that the mean (median) value of the acquisition premium in cash deals is 71.962% (41.170%), higher than that in non-cash deals, which is 38.651% (28.360%). The acquisition premium found in cash deals is generally attributed to compensation to the target shareholders for the personal tax that is immediately deducted from their capital gains from selling their shares for cash, unlike non-cash deals where the tax component is deferred (Huang and Walkling 1987).

The descriptive statistics of the variables used in the regression model are reported in Table 2 for the overall sample, as well as the relevant results once the sample has been stratified by the method of payment used.

In this table, the pre-merger abnormal accruals of acquirers ([EM.sub.ACQi]) in non-cash deals show a mean (median) value of 0.877% (0.612%), which is much higher than the mean (median) abnormal accruals value of cash acquirers 0.340% (-0.303%). This observed difference is consistent with the literature as cash acquirers lack the economic incentive to adopt income-increasing reporting methods unless they use equity shares to finance the M&A deal (Erickson and Wang 1999). The premium paid in cash deals is higher than in non-cash deals, with a mean (median) of 71.962% (41.170%) versus 38.651% (28.360%), respectively.

On average, acquirers that are able to offer cash are more profitable with a mean (median) pre-merger ROE of 6.50% (7.60%) than bidders in non-cash deals that have a mean (median) ROE of -0.131% (2.50%). This is preliminary evidence that bidders in non-cash deals are poor performing with an economic incentive to manage earnings to lead to overvaluation of their shares. Table 2 also reveals that cash acquirers have a greater pre-merger toehold in their target firms, with a mean (median) value of 1.44% (0.00%) versus acquirers in non-cash deals with a mean (median) value of 0.278% (0.00%). This is intuitive as obtaining a controlling interest in a target firm using cash is more feasible if the acquirer has a greater toehold in it (Bugeja and Walter 1995). We find that non-cash bidders are more levered (in terms of both mean and median debt ratios) than cash bidders and that non-cash bidders are more likely to make related acquisitions, which are challenged less frequently, with a smaller average transaction size. Investment banking fees are larger for cash transactions than non-cash deals in raw terms, although not relative to deal size.

Friedman (2006) finds that bidders in non-cash deals exhibit a greater propensity to manage premerger earnings and that targets in non-cash deals are less likely to "see through" bidder efforts to manage earnings. Therefore, we first examine whether bidders are more prone to manage earnings in equity deals than in cash deals. Accordingly, we test for significant mean differences in all dependent and independent variables (which includes our measure of earnings management) across the two groups (cash and non-cash deals).

Our findings presented in Table 3 indicate that earnings management is significantly higher for the non-cash sample (p-value < 0.05). This indicates that the potential for bidders to use earnings management to overstate the value of equity to target shareholders exists to a greater extent in equity deals. (10)

Table 3 also provides useful information supporting Table 2. We see that bidder ROE in cash deals is significantly greater than non-cash deals at the 1% level, although the difference between target ROE is not significant. Bidder debt in non-cash deals is significantly higher than in cash deals (at the 1% level). Therefore, bidders in non-cash deals, which are less profitable and more heavily leveraged, also tend to practice earnings management prior to bids.

LINEAR REGRESSION MODEL

To examine the relationship between the level of acquisition premium and acquirers' pre-merger earnings management, the following linear regression model is used:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)

Where:

PR: is the acquisition premium in an M&A deal over the four weeks prior to the announcement (described below);

[EM.sub.ACQ]: is the acquirer's earnings management as proxied by the abnormal accruals during the last three quarters prior to the deal's announcement date;

[ROE.sub.ACQ]: is the acquirer's return on equity ratio prior to the M&A announcement date. It is calculated based on the acquirer's performance in the 12 months preceding the deal;

[ROE.sub.TGT]: is the target's return on equity ratio prior to M&A announcement date. It is calculated based on the target's performance in the 12 months preceding the deal;

[IB.sub.TGT]: is the advisory fees paid for investment banking services used by the target firm during the M&A transaction;

[TO.sub.EACQ]: is the acquirer's toehold ownership interest in the target firm prior to the controlling M&A deal, and 0 otherwise;

RSIZE: is the size of the M&A deal relative to the acquirer's size;

[DBT.sub.TGT]: indicates the target's debt ratio before the M&A;

CHLNG: is a dummy variable which takes the value of 1 if the deal was challenged (i.e., by having multiple bidders), 0 otherwise;

VRLTD: is a dummy variable which takes the value of 1 if the deal was a synergistic vertical integration M&A, 0 otherwise;

[D.sub.y+1999]: is a dummy variable for the year in which the M&A was announced, where y=[1,m],m=9;

[gamma]: is the coefficients' index;

[euro]: is the error term and

i: is the sampled M&A deals' index.

We use a pooled ordinary least squares premium regression model and a panel data set consisting of firm-year observations for the period 12/31/1999 to 12/31/2008. We present all analyses with clustered standard errors at the firm level to address any potential cross-sectional correlation of error terms (Petersen 2009). Furthermore, all models include year controls to address macroeconomic and time period effects as well as industry controls (11).

Results

In Table 4, the results of the OLS regression are presented for noncash deals. We first examine the relationship between earnings management and acquisition premia in non-cash deals. The results of our regressions are shown in Table 4. We show three models for the sake of robustness.

The coefficient of acquirer's pre-merger earnings management ([EM.sub.ACQi]) (i.e., the measurement variable) is positive and significant sign (P-value< 1 percent) for all three models. Thus, all three models indicate a significant positive relation between pre-merger earnings management by acquirers and the acquisition premium calculated based on share prices in non-cash M&A deals. This is consistent with our cynical investor hypothesis. Low bidder earnings quality is a signal to target managers and their financial advisors that the bidder is attempting to manipulate earnings in an effort to provide a lower bid; as informed capital market participants, they are aware that the impact of earnings management is transient.

The control variables [ROE.sub.ACQi], [IB.sub.TGTi], [TOE.sub.ACQi] and [DBT.sub.TGTi] are statistically significant and provide insights into the acquisition valuation process The coefficient estimate of [ROE.sub.ACQi] ([gamma]2) indicates a positive association between the pre-merger acquirer's profitability and the size of the acquisition premium offered in a M&A deal. Offering a higher premium in a share-swap deal implies that the bidder will issue more shares than is necessary to the shareholders of the target, which eventually dilutes the EPS for the original shareholders of the acquirer. Therefore, this positive relationship between [PR.sub.i] and [ROE.sub.ACQi] indicates that the acquirer's management team must hold a greater level of confidence over the future prospects of the firm. Thus it is more convincing to bidding firm shareholders to pay a relatively higher acquisition premium (i.e., issuing more shares) if the acquiring firm has shown better pre-merger performance (Roll 1986; Hayward and Hambrick 1997).

However, the coefficient of [ROE.sub.TGTi], which indicates the pre-merger target's performance, is negative and significant (at the 10% level) in models 2 and 3, as shown in Table 3. These results indicate evidence of an inverse relation between [ROE.sub.TGTi] and the premium paid in a M&A deal, which is consistent with the empirical findings of some previous studies arguing that there is higher potential of creating value by acquiring the poorly managed targets (Lang, Stulz et al. 1989; Servaes 1991; Bugeja and Walter 1995). Therefore, the acquirer's management can justify offering higher premia by the anticipated potential of post-merger value creation.

The control variable [TOE.sub.ACQi] is included in regressions 2 and 3. The OLS output provides negative, significant coefficient estimates at a 10% significance level. Consistent with the literature, the acquirer's pre-merger ownership interest in the target firm before a controlling position has been achieved has a reducing effect on the acquisition premium offered (Stulz 1988; Bugeja and Walter 1995; Wickramanayake and Wood 2009). Naturally, the acquirer's toehold in the target should be inversely related to the incremental percentage for obtaining a controlling interest, which means there will be a lower quantity demanded for the target's shares. Thus, there should be a lower equilibrium price and consequently a lower premium offered.

The model also controls for the capital structure of the target firm ([DBT.sub.TGT]).The results indicate that there is a negative relationship between the target's debt ratio and the acquisition premium as predicted by Crawford and Lechner (1996), who argued that the target firm's attractiveness decreases with the level of its leverage. However, the results produced by the OLS regression for the non-cash M&A deals in Table 3 do not indicate significance for the coefficients of the variables [IBT.sub.GTi], [RSIZE.sub.i], [CHLNG.sub.i] or [VRLTD.sub.i].

CASH DEALS

As argued in the hypothesis development section, pre-merger earnings management would have an impact on the acquisition premium only if the acquirer decided to offer its equity within the M&A payment structure. This hypothesis is directly tested in the previous section using a non-cash sample. Alternatively, it could be argued that the hypothesis itself must imply that the acquirer's pre-merger earning management should be shown to be irrelevant to the acquisition premium in the case of a 100% cash deal. Therefore, this section presents the results of analysis conducted using a sample of cash only deals.

In Table 5, the OLS regression results of the model are reported for a concurrent sample of cash deals for the same period.

In models 1, 2 and 3, the coefficient of [EM.sub.ACQi] is insignificantly different from zero, unlike the results reported for the non-cash deals. The overall findings of the analysis of pure cash deals fails to show any significant relation between acquirers' pre-merger earnings management and the acquisition premium. This evidence adds greater robust support for the earlier findings regarding the noted significant relationship in non-cash deals and supports our second hypothesis.

Taken together, the results in Tables 4 and 5 indicate that while bidders in stock deals attempt to influence target perception of acquirer share value so that they may offer a lower premium based on over-valued bidder shares, target managers are able to extract a higher premium regardless. This runs counter to the naive investor hypothesis and indicates that target managers (and their investment advisors) act as sophisticated fiduciary advocates for their principles (i.e., shareholders). (12) An explanation for why institutional investors have been shown to exhibit naivete in matters of accruals misevaluation but target managers and investment advisors may not is that target managers have only one firm (the bidder) to evaluate the characteristics for and a legal duty to do so carefully, whereas institutional investors have a large, diversified pool of firms to evaluate, often across industries. Investment advisors are often specialized, typically generate income as a percentage of deal value, deal with a relatively small number of clients and also have a reputational incentive to ensure they get subsequent business by "getting the best deal" for target shareholders.

THE INCREMENTAL IMPACT OF EM ON PREMIA FOR NON-CASH VERSUS CASH PAYMENT

The results in Tables 4 and 5 address whether earnings management impacts acquisition premia in stock only deals and in cash only deals and indicate that bidder earnings management effectively impacts acquisition premia exclusively in stock deals. We next explore the incremental impact of stock payment and of full versus partial acquisitions on acquisition premia, relative to all cash deals. (13) The results of this analysis are shown in Table 6. We present four models, where one represents cash deals with stock deals as a benchmark, and three with a dummy equal to 1 if the deal is a stock deal and 0 for all cash deals, 0 for hybrid deals, or 0 for cash or hybrid deals, respectively.

Panel A, Model 1 of Table 6 shows the results of regression specifications with a cash variable (CASH) equal to 1 if the deal is a cash deal, and 0 otherwise, along with the interaction of the cash variable and earnings management variable. We find that, consistent with the results in Table 5, bidder earnings management ([EM.sub.ACQ]) does not affect the premium in cash deals. We find that, consistent with Table 3, cash deals involve higher premia. The interaction of CASH and [EM.sub.ACQ] is insignificantly different from zero. This implies that when firms pay with cash, earnings management does not significantly impact acquisition premia versus non-cash deals. Conversely, relative to cash deals, earnings management in stock deals does not yield higher premia.

Panel B, Model 1 of Table 6 shows the results of regression specifications with a share variable called Share%, which is defined as percentage of the deal value consisting of stock payment to the transaction value (i.e., a share% of 100 would reflect a full acquisition, whereas a share% of 0 would reflect a cash acquisition). Model 1 shows that earnings management is insignificant in determining the premium in the acquisition. The negative and significant coefficient of Share% indicates that deals with greater percentage of transaction value consisting of stock pay significantly lower premia than those with lower (or no) levels of stock, consistent with prior results. The interaction of Share% and E[M.sub.ACQi] however, is insignificantly different from zero. This indicates that there is no incremental impact of bidder earnings management on premia for deals more heavily concentrated in stock payment than in deals with lower levels of stock (or cash deals).

Panel B, Model 2 of Table 6 reports the results of a model specification where the share variable is defined as Share100, which is a dummy =1 if the acquisition is a full acquisition, and 0 otherwise (including cash and hybrid deals). Therefore, the results express the incremental impact of bidder earnings management on acquisition premia for full acquisitions versus less equity concentrated acquisitions. We find that the coefficient of bidder earnings management is negatively related to premia, though insignificantly different from zero. The share variable is negative and significant, indicating that full acquisitions are characterized by lower premia than less equity intensive transactions or cash transactions. However, in this specification, the interaction of Share and [EM.sub.ACQ] is positive and significant, albeit at the 10% level. Therefore, in full acquisitions, the incremental impact of bidder earnings management on premia is more pronounced and leads to significantly higher premia. It may be the case that the effect observed in Table 4, that earnings management in share deals positively impacts acquisition premia, is noted primarily in greater equity intensive deals. The results are qualitatively similar when the share variable is defined as 90% equity or above versus lower than 90% equity (untabulated). These results are consistent with Hypothesis 2.

Therefore, our results again support the conjecture that target manager, shareholders and investment advisors are not naive. Incrementally, while the evidence is ambiguous that they may not earn a significantly higher premium in acquisitions than cash deals or lower equity intensive deals, we find no evidence that the bidder is able to "get away" with paying a lower price for the target using overvalued shares.

ROBUSTNESS TESTS

In order to ensure that our results are robust to alternative model specifications, we conduct a number of additional robustness tests. First, we examine the impact of earnings management on acquisition premia by target firms. The framework of this study considers the implication of the acquirer's pre-merger earnings management in particular, and we leave the issue of target earnings management to subsequent perusal. However, one could argue that the rationale of the study's theoretical model can be extended by including the potential pre-merger earnings management of the target firm as well. Thus, we expect to find a negative relationship between the target's pre-merger earnings management and the acquisition premium paid by the acquirer. On the other hand, Anilowski et al. 2009 find that target earnings management prior to an acquisition impacts the number of bidders and thus a higher purchase price. For this reason, a further investigation is conducted by including the pre-merger abnormal accruals of the target, [EM.sub.TGTi].

In untabulated results, we find that the coefficient of [EM.sub.ACQi] retains its sign and significance. On the other hand, the coefficient estimate of [EM.sub.TGTi] shows the expected negative sign in all regressions but is statistically insignificant. This weak evidence on the negative relationship between targets' earnings management and the acquisition premium can be explained initially by the inconsistent evidence of earnings management undertaken by target firms before a M&A, which may be attributed to the time constraint facing the target's management (Erickson and Wang 1999). Target's management is typically not aware about a potential M&A plan before it is approached by an acquirer. Therefore, it cannot plan and manage the target's accruals efficiently in advance before the deal's official announcement. (14)

It may also be possible that another important variable, auditor quality, might mitigate the relationship between the magnitude of the premium and higher abnormal accruals given that Big 4 auditors have been associated with higher financial reporting quality. (15) In additional model specifications, we include a Big 4 auditor dummy variable as well as an interaction with our EM variable. We find that the interaction variable is insignificantly different from zero and conclude that auditor quality does not reduce the ability of the impact of auditor earnings management on the premium paid to targets.

In addition, we control for the growth potential of the target firm. First, we conducted all regressions using R&D expense scaled by total assets (RD), given that Hamza (2011) and Laaamanen (2008) find that higher acquisition premium may reflect rational valuation for intangibility oriented targets. We find that the positive and significant association and magnitude between earnings management and premium still holds for the non-cash deal group and is still insignificant for the cash deal group. We find a significant and negative association between the premium and RD only for the cash deal group. We next repeat our analysis while including market to book (MTB) as a proxy for growth. The positive and significant association and magnitude between earnings management and premium still holds for the non-cash deal group and is still insignificant for the cash deal group. We find a marginal significant and negative association between the premium and MTB only for the non-cash deal group. Alternatively, we include a sales growth percentage measure of growth and find the positive and significant association and magnitude between earnings management and premium still holds for the non-cash deal group and is still insignificant for the cash deal group. We find a significant and positive association between the premium and sales growth (p = 0.05) only for the non-cash deal group.

Furthermore, we employ different time horizons in order to define the acquisition premium. Instead of using base prices four weeks ([PR.sub.t-4w]) before the deal announcement date, the acquisition premium is re-calculated using base prices one week ([PR.sub.t-1w]) before the announcement date, reflecting the market valuation for the acquiring and target firms closer to the official announcement date. (16) A comparison of the descriptive statistics per year for the two definitions of the acquisition premium (untabulated) reveals that [PR.sub.t-1w] is, on average, slightly lower than [PR.sub.t-4w] with a mean (median) value of 34.29 (26.19) compared to 36.23 (27.82), respectively. In untabulated results the regression model with robust standard errors is replicated using [PR.sub.t-1w'i] as the dependent variable. The results are similar to the results reported earlier here, as the coefficient estimate of [EM.sub.ACQi] remains highly significant (P-value < 1 percent) in all regressions, ranging between 1.18 (robust t-value = 2.99) to 1.34 (robust t-value = 3.37). Therefore, the overall results of the variable [PR.sub.t-1w] provide robust evidence between the positive relationship between the acquirer's pre-merger earnings management and the acquisition premium in non-cash M&A deals. This relationship holds under different approaches used for calculating the acquisition premium.

Additional tests control for: the pre-merger share performance impact on valuation, the impact of the post-2007 financial crisis and considering the effects of bear and bull markets. Our central finding that earnings management is ineffective in paying a lower premium in an acquisition in non-cash deals - is robust to alternative model specifications.

Conclusions

The existing literature argues that capital markets provide incentives for event-specific earnings management when firms issue equity shares to raise capital, such as SEO (Rangan 1998 and Teoh, Welch et al. 1998), IPO (DuCharme, Malatesta et al. 2001) and M&A (Louis 2004; Gong, Louis et al. 2008). The common justification of this argument is that firms attempt to reduce their cost of capital (equity) and try to mitigate the dilutive effect on the existing shareholders' interests (Hansen 1987 and Fields, Lys et al. 2001). Therefore, the specific research area of event-specific earnings management and accruals misvaluation implicitly consider Sloan's (1996) naive investor hypothesis--by assuming that earnings management cannot be uncovered by market participants--in order to validate the managerial motivation argument.

By considering the M&A transacting context, this paper addresses the question of how pre-merger earnings management is handled by targets. This study challenges the naive investors' hypothesis and its relevance to the theoretical rationalization of earnings management in an M&A context. It offers a counter-theory suggesting that M&A participants are relatively sophisticated--rather than naive--users, since there are informed agents (e.g., managements) acting on behalf of investors (e.g., the shareholders). Correspondingly, the cynical investor hypothesis posits that pre-merger earnings management exhibited by an acquiring firm can be detected, and the acquirer's share price will be discounted by a target firms in non-cash deals.

Our results document significant and robust empirical evidence showing that the acquirer's pre-merger earnings management and acquisition premium are positively related in non-cash M&A deals but not cash transactions. Non-cash acquirers that adopt income-increasing pre-merger earnings management pay higher acquisition premia in completing their M&A deals. Our results are consistent with inefficiency on behalf of the acquirer's management team such that their efforts are not beneficial due to the target's ability to detect and adjust the acquirer's earnings management. In terms of incremental impact of bidder earnings management on acquisition premia in stock (or full stock) versus cash (or lower equity intensive deals), the evidence supports the conjecture that bidders are not able to "get away with" paying less for the target using overvalued shares. It appears that, acting as fiduciary agents of their shareholders, target managers and investment advisors are able to uncover the cosmetic inflation of their acquirers' pre-merger earnings, which represents a bargaining advantage where targets in stock deals can demand a higher share swap-ratio (i.e., premium) in order to compensate for the potential overvaluation of the acquirer's share so as to avoid the adverse selection problem. This explanation directly challenges the naive investor hypothesis in the M&A context and further suggests that acquirers' efforts to manage earnings prior to non-cash M&A deal are wasteful.

References

Akerlof, G.A. "The Market for 'Lemons': Quality Uncertainty and the Market Mechanism." Quarterly Journal of Economics 84.3(1970): 488-500.

Alsharairi, M., and A. Salama. "Does High Leverage Impact Earnings Management?: Evidence from Non-cash Mergers and Acquisitions." Journal of Financial and Economic Practice 12.1(2012): 17-33.

Anilowski, C.L., A.J. Macias, and J.M. Sanchez. "Target Firm Earnings Management and the Method of Sale: Evidence from Auctions and Negotiations." Working paper. Purdue University and University of Arkansas, 2009.

Antoniou, A., P. Arbour, et al. "How Much Is Too Much: Are Merger Premiums Too High?" European Financial Management 14.2(2008): 268-287.

Arzac, E.R. Valuation for Mergers, Buyouts, and Restructuring. New York: John Wiley & Sons, Inc., 2004.

Asano, T., Y. Ishii, et al. "Earnings Management and the Market Performance of Acquiring Firms in Stock-for-Stock Mergers--Evidence from Japan." 30th Annual Congress of European Accounting Association Conference. Lisbonne. 25-27 April 2007.

Ashiq Ali, D.D., H. Lee-Seok, et al. "Accruals and Future Stock Returns: Tests of the Naive Investor Hypothesis." Journal of Accounting, Auditing & Finance 15.2(2000): 161-181.

Baik, B., J.-K. Kang, and R. Morton. "Earnings Management in Takeovers of Privately Held Targets." Working Paper, 2007. Available at SSRN: http:// ssrn.com/abstract=1013639.

Balsam, S., E. Bartov, et al. "Accruals Management, Investor Sophistication, and Equity Valuation: Evidence from 10-Q Filings." Journal of Accounting Research 40.4(2002): 987-1012.

Botsari, A. and G. Meeks. "Do Acquirers Manage Earnings Prior to a Share for Share Bid?" Journal of Business Finance & Accounting 35.5-6(2008): 633.

Bugeja, M. and T. Walter. "An Empirical Analysis of Some Determinants of the Target Shareholder Premium in Takeovers." Accounting & Finance 35.2(1995): 33-60.

Chung, R., M. Firth, et al. "Institutional Monitoring and Opportunistic Earnings Management." Journal of Corporate Finance 8.1(2002): 29-48.

Crawford, D., and T.A. Lechner. "Takeover Premiums and Anticipated Merger Gains in the US Market for Corporate Control." Journal of Business Finance & Accounting 23.5-6(1996): 807-829.

Dechow, P.M. "Accounting Earnings and Cash Flows as Measures of Firm Performance: The Role of Accounting Accruals." Journal of Accounting & Economics 18.1(1994): 3-42.

Dechow, P.M., R.G. Sloan, et al. "Detecting Earnings Management." Accounting Review 70.2(1995): 193-225.

Dechow, P.M., S.A. Richardson, et al. "Why Are Earnings Kinky? An Examination of the Earnings Management Explanation." Review of Accounting Studies 8.2-3(2003): 355-384.

DuCharme, L.L., PH. Malatesta, et al. "Earnings Management: IPO Valuation and Subsequent Performance." Journal of Accounting, Auditing & Finance 16.4(2001): 369-396.

Eckbo, B.E., R.M. Giammarino, et al. "Asymmetric Information and the Medium of Exchange in Takeovers: Theory and Tests." Review of Financial Studies 3.4(1990): 651-675.

Erickson, M. and S.-W. Wang. "Earnings Management by Acquiring Firms in Stock for Stock Mergers." Journal of Accounting & Economics 27.2(1999): 149.

Fairfield, R, J.S. Whisenant, and T.L. Yohn. "Accrued Earnings and Growth: Implication for Future Profitability and Market Mispricing." The Accounting Review 78.1(2003): 353-371.

Fields, T.D., T.Z. Lys, et al. "Empirical Research on Accounting Choice." Journal of Accounting & Economics 31.1-3(2001): 255-307.

Fisher, P.E., and H. Louis. "Financial Reporting and Conflicting Managerial Incentives: The Case of Management Buyouts." Management Science 54.10(2008): 1700-1714.

Friedman, John N. "Stock Market Driven Acquisitions: Theory and Evidence." Working paper, Harvard University, 2006.

Gong, G., H. Louis and A.X. Sun. "Earnings Management, Lawsuits, and Stock-for-Stock Acquirers' Market Performance." Journal of Accounting and Economics 46.1(2008): 62-77.

Grasso, L.P., P.A. Tilley, et al. "The Ethics of Earnings Management: Perceptions after Sarbanes-Oxley." Management Accounting Quarterly 11.1(2009): 45-69.

Hajbaba, A., and R. Donnelly. "Acquirers' Performance in Hot and Cold Merger Markets: Evidence of Mispricing." Review of Accounting & Finance 12.3(2013): 204-225

Hamza, T. "Determinants of Short-term Value Creation for the Bidder: Evidence from France." Journal of Management & Governance 15.2(2011): 157-186.

Hansen, R.G. "A Theory for the Choice of Exchange Medium in Mergers and Acquisitions." The Journal of Business 60.1(1987): 21.

Haunschild, PR. "How Much Is That Company Worth?: Interorganizational Relationships, Uncertainty, and Acquisition Premiums." Administrative Science Quarterly 39.3(1994): 391-411.

Haw, I.-M., V. Pastena, et al. "The Association between Market-Based Merger Premiums and Firms' Financial Position Prior to Merger." Journal of Accounting, Auditing & Finance 2.1(1987): 19.

Hayward, M.L.A. and D.C. Hambrick. "Explaining the Premiums Paid for Large Acquisitions: Evidence of CEO Hubris." Administrative Science Quarterly 42.1 (1997): 103-127.

Higgins, H. "Do Stock-for-Stock Merger Acquirers Manage Earnings? Evidence from Japan." Journal of Accounting and Public Policy 32.1(2013): 44.

Huang, Y-S., and R.A. Walkling. "Target Abnormal Returns Associated with Acquisition Announcements." Journal of Financial Economics 19.2(1987): 329-349.

Jensen, M.C. "Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers." American Economic Review 76.2(1986): 323.

Jensen, M.C. and W.H. Meckling. "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure." Journal of Financial Economics 3.4(1976): 305-360.

Jones, J.J.. "Earnings Management during Import Relief Investigations." Journal of Accounting Research 29.2(1991): 193-228.

Kothari, S.R, A.J. Leone, et al. "Performance Matched Discretionary Accrual Measures." Journal of Accounting & Economics 39.1(2005): 163-197.

Kraft, A., A. Leone, and C. Wasley. "An Analysis of the Theories and Explanations Offered for the Mispricing of Accruals and Accrual Components." Journal of Accounting Research 44.(2006): 297-339.

Laamanen, T. "On the Role of Acquisition Premium in Acquisition Research." Strategic Management Journal 28.13(2008): 1359-1369.

Lang, L.H.P., R.M. Stulz, et al. "Managerial Performance, Tobin's Q, and the Gains from Successful Tender Offers." Journal of Financial Economics 24.1(1989): 137-154.

Lee, H., S. Kim, et al. "Earnings Management of Acquiring Firms in Stock-for-Stock Takeovers in the Telecommunications Industry." Journal of Media Economics 21.4(2008): 17.

Liu, Y., Y. Ning, et al. "Earnings Management Surrounding New Debt Issues." Financial Review 45.3(2010): 659-681.

Louis, H. "Earnings Management and the Market Performance of Acquiring Firms." Journal of Financial Economics 74.1(2004): 121.

Madura, J., and T. Ngo. "Flow Accounting Fraud Has Changed Merger Valuation." Applied Financial Economics 20.12(2010): 923-940.

Meisel, S. "Detecting Earnings Management in Bank Merger Targets Using an Industry Specific Model." Southern Business Review 37.3(2013): 1.

Moeller, S.B., F.P. Schlingemann, et al. "Firm Size and the Gains from Acquisitions." Journal of Financial Economics 73.2(2004): 201-228.

Myers, S.C., and N.S. Majluf. "Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have." Journal of Financial Economics 13.2(1984): 187-221.

Porrini, P. "Are Investment Bankers Good for Acquisition Premiums?" Journal of Business Research 59.1(2006): 90-99.

Pungaliya, R.S., and A.M. Vijh. "Do Acquiring Firms Manage Earnings?" SSRN: http://ssrn.com/abstract=1273464. (2008).

Rangan, S. "Earnings Management and the Performance of Seasoned Equity Offerings." Journal of Financial Economics 50.1(1998): 101-122.

Richardson, V.J. "Information Asymmetry and Earnings Management: Some Evidence." Review of Quantitative Finance & Accounting 15.4(2000): 325.

Roll, R. "The Hubris Hypothesis of Corporate Takeovers." The Journal of Business 59.2(1986): 197-216.

Schipper, K. "Commentary on Earnings Management." Accounting Horizons 3.4(1989): 91-102.

Schwert, G.W. "Markup Pricing in Mergers and Acquisitions." Journal of Financial Economics 41.2(1996): 153-192.

Schwert, G.W. "Hostility in Takeovers: In the Eyes of the Beholder?" Journal of Finance 55.6(2000): 2599-2640.

Servaes, H. "Tobin's Q and the Gains from Takeovers." The Journal of Finance 46.1(1991): 409-419.

Sloan, R.G. "Do Stock Prices Fully Reflect Information in Accruals and Cash Flows about Future Earnings?" Accounting Review 71.3(1996): 289-315.

Stulz, R. "Managerial Control of Voting Rights: Financing Policies and the Market for Corporate Control." Journal of Financial Economics 20.1-2(1988): 25-54.

Teoh, S.H., I. Welch, et al. "Earnings Management and the Underperformance of Seasoned Equity Offerings." Journal of Financial Economics 50.1(1998): 63-99.

Wickramanayake, J., and A. Wood. "Determinants of Acquisition Premiums: Empirical Evidence from Mining Industry in Australia and Canada." EFMA Annual Meetings. Milan. 24-27 June 2009.

Xie, H. "The Mispricing of Abnormal Accruals." The Accounting Review 76.3(2001): 357-373.

Yung, K., Q. Sun, and H. Rahman. "Acquirer's Earnings Quality and the Choice of Payment Method in Mergers and Acquisitions." Managerial Finance 39.10(2013): 979-1000

KIMBERLY C. GLEASON

University of Pittsburgh

MALEK A. ALSHARAIRI

German Jordanian University (GJU)

YEZEN H. KANNAN

Duquesne University

(1) The sampling period cut-off point (12/31/2008) is determined once data collection for this study started in the beginning of 2009. A period of ten years is chosen to include both pre-and post-financial crisis periods.

(2) The Input-Output Accounts were obtained from http://www.bea.gov/industry/iotables/ on January 20, 2011.

(3) The NAICS-SIC conversion tables were found at the US Census Bureau website at http://www.census.gov/epcd/www/naicstab.htm on February 4, 2011.

(4) Unlike in cash deals, the acquisition premiums in non-cash M&As are calculated based on publicly available pre-merger share prices. We use the term "nominal" premium to describe this value because non-cash offers may contain mispriced assets (e.g., equity shares) that may be subject to revaluation by the well-informed beneficiary.

(5) Our EM variable incorporates accrual expenses. It considers both the accrual revenues as well as the accrual expenses when estimating the abnormal accruals. This definition of earnings management has been tested and modified several times (Healy 1985, DeAngelo 1986, Jones 1991, Dechow et al 1995, Kothari et al 2005) and has been widely adopted in the literature (Schipper 1989; Teoh, Welch et al. 1998; Dechow, Richardson et al. 2002; Louis 2004; Gong et al. 2008; Lee, Kim et al. 2008; Grasso, Tilley et al. 2009; Liu, Ning et al. 2010).

(6) As a check for the method used in calculating the current accruals, the abnormal accruals are also calculated using the cash flow method for comparison by which current accruals are calculated as CAC[C.sub.i,j] = IBC[Q.sub.i,j]]-OANCF[Q.sub.i,j] +DPC[Q.sub.i,j], where IBC[Q.sub.i,j] is income before extraordinary items appeared in the statement of cash flow of firm i at quarter j and calculated using the Compustat year-to-date item of a mnemonic code IBCY; OANCF[Q.sub.i,j]]. is net cash flow from operating activities of firms i on quarter j but calculated using the year-to-date item of a mnemonic code OANCFY and DPCQi,j is the depreciation and amortization reported in the statement of cash flow of firm i on quarter j but calculated using the year-to-date item of a mnemonic code DPCY. The correlation coefficients are examined when relating abnormal accruals calculated using balance sheet method and cash flow method. Pearson's coefficients range from 0.237 to 0.555 while Spearman's coefficients range from 0.453 to 0.628 for both acquirers and targets. The coefficients found positive and very significant (P<0.00001), indicating additional robustness of the findings.

(7) In late 2007 Compustat switched to Xpressfeed delivery mechanism (XPF) using mnemonic coding to data items.

(8) This study follows Kothari et al.'s (2005) recommendation of employing portfolio performance matching instead of adding a performance measure as a regressor to the accrual regression model for more reliable results.

(9) Since the Sarbanes-Oxley Act (SOX) has had a significant impact on the corporate environment in general in the US, a dummy variable, [SOX.sub.i], is included in the model instead of yearly dummies to indicate the post-SOX era to control for the differences in the corporate investment attitudes and the regulation setting between the two periods. When applying this procedure, the results do not significantly change.

(10) We thank an anonymous reviewer for suggesting this test.

(11) Hajbaba and Donnelley (2013) argue that premia vary across "hot" and "cold" markets. Alterative time controls, including those used in Hajbaba and Donnelley, yield qualitatively similar results.

(12) Advocacy may be present in extracting higher premia in cash deals as well, but as bidders in cash deals do not have an economic incentive to manage earnings, cash deals do not offer an incentive to evaluate the naive investor hypothesis from the same perspective. We thank a reviewer for pointing out this insight.

(13) We thank an anonymous reviewer for this suggestion.

(14) Target earnings management may also affect the bidder premium in cash deals. In further tests, we explore the impact of target EM on the premium. We find that there is a positive and significant relationship between target EM and premia, including in model specifications where target EM appears in regressions with bidder EM. The sign and significance of bidder EM remains the same, even when target EM is controlled for. This in and of itself has many interesting implications, but it is beyond the scope of this paper. Due to length considerations, we leave the subject to further exploration in future research. We thank three anonymous reviewers for their suggestions on additional tests.

(15) We thank an anonymous reviewer for this suggestion.

(16) Using base prices closer to the deal's announcement date may seem much more realistic, but also prices of closer dates are more likely to reflect an event-specific informational leakage.
TABLE 1
Sample Distribution and Descriptive Statistics of Acquisition
Premium by Year

The following table presents the distribution of the
acquisition premium, which is calculated using shares prices
four weeks prior to the M&A deal's announcement date. The
distribution by year is provided for the total 424 deals in
Panel A, for the 262 non-cash deals in Panel B and for the
162 cash deals in Panel C. The samples were initially taken
from Thomson One Banker for the period 1999-2008; then we
require that acquiring and target firms of the respective
deals have accounting data on Compustat from 1997 and 2008
inclusive. Descriptive statistics including mean, standard
deviation (STDV), median, minimum, maximum and the count of
deals (N) are provided on the acquisition premium of the
deals per year in each panel.

Panel A: Total Deals

Year         1999       2000       2001       2002       2003

Mean        69.855     49.686     44.760    114.950     60.060
STDV        56.266     53.471     48.054    365.317     68.781
Median      58.300     40.520     44.480     23.910     45.485
N             27         77         61         37         38

Panel B: Non-cash Deals

Year         1999       2000       2001       2002       2003
Mean        71.377     49.213     36.260     11.374     53.946
STDV        54.739     54.299     47.839     52.421     63.340
Median      58.410     37.760     38.735     11.440     32.700
N             21         58         40         20         24

Panel C: Cash Deals

Year         1999       2000       2001       2002       2003
Mean        64.530     51.128     60.950    236.805     70.541
STDV        66.586     52.273     45.223    517.996     78.610
Median      38.615     41.170     54.910     41.590     46.150
N             6          19         21         17         14

Panel A; Total Deals

2004         2005       2006       2007       2008      Total
34.988      36.462     39.884     29.789     40.640     51.103
39.188      35.481     31.977     22.356     49.434    118.228
28.895      28.240     32.190     28.840     30.290     32.645
40            47         36         35         22        420

Panel B: Non-cash Deals

2004         2005       2006       2007       2008      Total
28.684      27.173     33.978     22.736     32.483     38.651
43.209      27.148     35.980     16.964     29.863     48.689
20.070      21.320     25.760     24.510     25.655     28.360
24            31         15         18         12        263

Panel C: Cash Deals

2004         2005       2006       2007       2008      Total
44.443      54.460     44.103     37.258     50.427     71.962
31.163      43.160     28.954     25.320     66.460    181.280
38.445      40.485     37.780     30.000     32.215     41.170
16            16         21         17         10        157

TABLE 2
Descriptive Statistics

This table presents the descriptive statistics of the main
variables used in the regression for the total deals, the
non-cash deals and the cash deals, where PR indicates the
acquisition premium in a controlling M&A deal based on
shares price index four weeks prior to announcement date;
[EM.sub.ACQ], indicates acquirer earnings management,
proxied by the abnormal accruals during the last three
quarters prior to deal's announcement date; [ROE.sub.ACQ]
indicates the acquirer's return on equity ratio prior to the
deal's announcement date and calculated based on the
acquirer's performance in the preceding 12 months;
[ROE.sub.TGT] indicates the target's return on equity ratio
prior to the deal's announcement date and calculated based
on the target's performance in the preceding 12 months;
[IB.sub.TGT] indicates the advisory fees of investment
banking services used by the target firm for completing the
deal; [TOE.sub.ACQ] indicates the acquirer's toehold
ownership interest in the target firm prior to the
controlling M&A deal, and 0 otherwise; RSIZE. indicates the
size of the deal relative to the acquirer's size;
[DBT.sub.TGT] indicates the target's debt ratio before the
deal; CHLNG, is a dummy variable which takes 1 if the deal
was challenged (i.e., by having multiple bidders), and 0
otherwise; and VRLTD is a dummy variable which takes 1 if
the deal was a synergistic vertical integration M&A, and 0
otherwise.

Total N=420

                            Mean         STDV       Median

[PR.sub.i]                51.103      118.228       32.645
[EM.sub.ACQi]              0.676        6.928        0.171
[ROE.sub.ACQi]            -0.057        0.906        0.047
[ROE.sub.TGTi]            -0.154        0.556        0.004
[IB.sub.TGTi]            217.094      725.116        0.370
[TOE.sub.TGTi]             0.713        4.506        0.000
[RSIZE.sub.i]             38.224      120.586        5.039
[DBT.sub.TGTi]             0.470        0.428        0.396
[CHLNG.sub.i]              0.050        0.218        0.000
[VRLTD.sub.i]              0.802        0.399        1.000

Non-cash Deals N=263

                            Mean         STDV       Median

[PR.sub.i]                38.651       48.689       28.360
[EM.sub.ACQi]              0.877        7.376        0.612
[ROE.sub.ACQi]            -0.131        1.135        0.025
[ROE.sub.TGTi]            -0.182        0.565       -0.009
[IB.sub.TGTi]            133.180      695.124        0.300
[TOE.sub.TGTi]             0.278        2.314        0.000
[RSIZE.sub.i]             20.659       75.640        3.265
[DBT.sub.TGTi]             0.492        0.491        0.418
[CHLNG.sub.i]              0.046        0.209        0.000
[VRLTD.sub.i]              0.814        0.390        1.000

Cash Deals N=157

                            Mean         STDV       Median

[PR.sub.i]                71.962      181.280       41.170
[EM.sub.ACQi]              0.340        6.111       -0.303
[ROE.sub.ACQi]             0.065        0.123        0.076
[ROE.sub.TGTi]            -0.106        0.537        0.015
[IB.sub.TGTi]            357.665      754.284        1.000
[TOE.sub.TGTi]             1.441        6.685        0.000
[RSIZE.sub.i]             67.648      167.496       13.718
[DBT.sub.TGTi]             0.433        0.292        0.374
[CHLNG.sub.i]              0.057        0.233        0.000
[VRLTD.sub.i]              0.783        0.413        1.000

TABLE 3
Descriptive Statistics

The following table presents t-test across the two groups
(non-cash and cash deals) across the dependant variable
(acquisition premium) and all independent vaiables. PR
indicates the acquisition premium in a controlling M&A deal
based on shares price index four weeks prior to announcement
date; [EM.sub.AGQ] indicates the acquirer's earnings
management proxied by the abnormal accruals during the last
three quarters prior to deal's announcement date;
[ROE.sub.ACQi] indicates the acquirer's return on equity
ratio prior to the deal's announcement date and calculated
based on the acquirer's performance in the preceding 12
months; [ROE.sub.TGT] indicates the target's return on
equity ratio prior to the deal's announcement date and
calculated based on the target's performance in the
preceding 12 months; [IB.sub.TGTi] indicates the advisory
fees of investment banking services used by the target firm
for completing the deal; [TOE.sub.ACQi] indicates the
acquirer's toehold ownership interest in the target firm
prior to the controlling M&A deal, and 0 otherwise;
[RSIZE.sub.i] indicates the size of the deal relative to the
acquirer's size; [DBT.sub.TGT] indicates the target's debt
ratio before the deal; CHLNG is a dummy variable which takes
1 if the deal was challenged (i.e.. by having multiple
bidders), and 0 otherwise; and VRLTD is a dummy variable
which takes 1 if the deal was a synergistic vertical
integration M&A, and 0 otherwise. Numbers in parentheses
represent t-values in two-tailed tests. The symbols *, **
and *** denote confidence interval of 10, 5 and 1%,
respectively.

                   Non-Cash        Cash Deal       F-Statistic
                     Deal

                     Mean            Mean

[PR.sub.i]          38.651          71.962         (13.86) ***
[EM.sub.ACQi]        0.877           0.340          (1.46) **
[ROE.sub.ACQi]      -0.131           0.065         (85.40) ***
[ROE.sub.TGTi]      -0.182          -0.106           (1.11)
[IB.sub.TGTi]       133.200         357.300          (1.18)
[TOE.sub.ACQi]       0.028           1.441         (8.35) ***
[RSIZE.sub.i]       20.660          67.648         (4.90) ***
[DBT.sub.TGTi]       0.492           0.433         (2.83) ***
[CHLNG.sub.i]        0.046           0.057           (1.24)
[VRLTD.sub.i]        0.814           0.783           (1.12)
N                     263             157

TABLE 4
Results of the Ordinary Least Squares' Regression for
Non-cash Deals

The following table presents the results of the OLS
regression model [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE
IN ASCII] where PR indicates the acquisition premium in a
controlling M&A deal based on shares price index four weeks
prior to announcement date; [EM.sub.ACQi] indicates the
acquirer's earnings management proxied by the abnormal
accruals during the last three quarters prior to deal's
announcement date; [ROE.sub.ACQi] indicates the acquirer's
return on equity ratio prior to the deal's announcement date
and calculated based on the acquirer's performance in the
preceding 12 months; [ROE.sub.TGT] indicates the target's
return on equity ratio prior to the deal's announcement date
and calculated based on the target's performance in the
preceding 12 months; [IB.sub.TGT] indicates the advisory
fees of investment banking services used by the target firm
for completing the deal; [TOE.sub.ACQi] indicates the
acquirer's toehold ownership interest in the target firm
prior to the controlling M&A deal, and 0 otherwise;
[RSIZE.sub.i] indicates the size of the deal relative to the
acquirer's size; [DBT.sub.TGTi] indicates the target's debt
ratio before the deal; CHLNG, is a dummy variable which
takes 1 if the deal was challenged (i.e., by having multiple
bidders), and 0 otherwise; and [VRLTD.sub.i] is a dummy
variable which takes 1 if the deal was a synergistic
vertical integration M&A, and 0 otherwise. Numbers in
parentheses represent t-values in two-tailed tests based on
clustered standard errors (Petersen 2009). The symbols *, **
and *** denote confidence interval of 10, 5 and 1%,
respectively.

                       (1)            (2)            (3)

Constant            77,655 **      82.602 **      83.327 **
                      (2.00)         (1.99)         (2.02)
[EM.sub.ACQi]        0.983 **       0.979 **       0.985 **
                      (2.02)         (2.05)         (2.03)
[ROE.sub.ACQi]                      5.589 *        5.647 *
                                     (1.84)         (1.85)
[ROE.sub.TGTi]                       -2.085         -2.365
                                    (-0.23)        (-0.26)
[IB.sub.TGTi]                       0.005 *        0.005 *
                                     (1.74)         (1.68)
[TOE.sub.ACQi]                     -1.793 **      -1.814 **
                                    (-2.50)        (-2.52)
[RSIZE.sub.i]                        0.023          0.019
                                     (0.96)         (0.78)
[DBT.sub.TGTi]                     -13.357 **     -13.264 **
                                    (-2.39)        (-2.36)
[CHLNG.sub.i]                                       -4.986
                                                   (-0.63)
[VRLTD.sub.i]                                       -2.914
                                                   (-0.27)
[SIGMA] Year           Yes            Yes            Yes
[SIGMA] Industry       Yes            Yes            Yes
N                      263            263            263
F-statistic         13.90 ***      17.54 ***      14.66 ***
Adj [R.sup.2]         15.78%         19.66%         19.75%

TABLE 5
Results of the Ordinary Least Squares' Regression for Cash Deals

The following table presents the results of the OLS
regression model [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE
IN ASCII] where PR, indicates the acquisition premium in a
controlling M&A deal based on shares price index four
weeks prior to announcement date; [EM.sub.ACQi] indicates the
acquirer's earnings management proxied by the abnormal
accruals during the last three quarters prior to deal's
announcement date; [ROE.sup.ACQi] indicates the acquirer's
return on equity ratio prior to the deal's announcement date
and calculated based on the acquirer's performance in the
preceding 12 months; [ROE.sub.TGTi] indicates the target's
return on equity ratio prior to the deal's announcement date
and calculated based on the target's performance in the
preceding 12 months; [IB.sub.TGTi] indicates the advisory
fees of investment banking services used by the target firm
for completing the deal; [TOE.sub.ACQi] indicates the
acquirer's toehold ownership interest in the target firm
prior to the controlling M&A deal, and 0 otherwise;
[R.sub.SIZE], indicates the size of the deal relative to the
acquirer's size;[DBT.sub.TGTi] indicates the target's debt
ratio before the deal; [CHLNG.sub.i] is a dummy variable
which takes 1 if the deal was challenged (i.e., by having
multiple bidders), and 0 otherwise; and [VRLTD.sub.i] is a
dummy variable which takes 1 if the deal was a synergistic
vertical integration M&A, and 0otherwise. Numbers in
parentheses represent t-values in two-tailed tests based on
clustered standard errors(Petersen 2009). The symbols *, **
and *** denote confidence interval of 10, 5 and 1%,
respectively.

                        (1)              (2)              (3)

Constant              121.376           34.031           -6.778
                     (2.46) **          (0.34)          (-0.06)
[EM.sub.ACQi]          -1.980           -1.589           -1.165
                      (-0.93)          (-0.74)          (-0.60)
[ROE.sub.ACQi]                         110.355           75.055
                                        (1.31)           (0.98)
[ROE.sub.TGTi]                         -30.805          -35.262
                                       (-0.36)          (-0.43)
[IB.sub.TGTi]                           -0.034           -0.033
                                       (-1.39)          (-1.50)
[TOE.sub.ACQi]                          -0.263           -0.101
                                       (-0.30)          (-0.10)
[RSIZE.sub.i]                           -0.009           -0.015
                                       (-0.06)          (-0.10)
[DBT.sub.TGTi]                         120.415          118.375
                                        (0.92)           (0.97)
[CHLNG.sub.i]                                           132.377
                                                         (1.05)
[VRLTD.sub.i]                                            24.600
                                                         (0.94)
[SIGMA] Year            Yes              Yes              Yes
[SIGMA] Industry        Yes              Yes              Yes
N                       157              157              157
F-statistic           1.77 **            0.52             0.53
Adj [R.sup.2]          22.00%           28.13%           30.89%

TABLE 6
Results of Incremental Premium Charged in the Presence
of Cash and Non-cash Deals

CASH_1 a measure of cash payment according to the
PAYMENT_TEXT variable from  Thomson Banker (1 if method of
payment = 'CASH.' and 0 otherwise);  Share% = SHAREXVALUE-
DEAL_VALUE; SharelOO = 1 if share% = 1, and 0 otherwise.
Numbers in parentheses represent t-values in two-tailed
tests based on clustered standard errors (Petersen 2009).
The symbols *, ** and  *** denote confidence interval of 10.
5 and 1%, respectively.

Panel A:
                      CASH_1

Constant              83.364
                    (2.72)***
[EM.sub.ACQi] (1)     0.684
                      (1.05)
CASH (2)              33.067
                     (2.41)**
(1)*(2)               -2.369
                     (-1.45)
[SIGMA] Controls       Yes
[SIGMA] Year           Yes
[SIGMA] Industry       Yes
N                      420
F-statistic          2.18 ***
Adj. R square         14.01%

Panel B:

                        (1)                  (2)
                       Share%             Share100_2

Constant           107.507 (2.94) ***   106.218 (3.27) ***
[EM.sub.AVQi](1)    -1.183 (-0.97)       -1.188 (-1.10)
Share (2)           20.418 (-1.27)     -22.330 (-2.06) **
(1)*(2)              1.801 (1.33)        2.230 (1.82) *
[SIGMA] Controls        Yes                  Yes
[SIGMA] Year            Yes                  Yes
[SIGMA] Industry        Yes                  Yes
N                       420                  420
F-statistic          1.97 ***             1.98 ***
Adj. R square         12.86%               13.30%
COPYRIGHT 2014 Creighton University
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Gleason, Kimberly C.; Alsharairi, Malek A.; Kannan, Yezen H.
Publication:Quarterly Journal of Finance and Accounting
Article Type:Statistical data
Geographic Code:1USA
Date:Sep 22, 2014
Words:12849
Previous Article:Transition economy and equity home bias: the case of Vietnam.
Next Article:Gold and the value-growth differential in stock returns.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters