Variable Construction

Market adjusted stock returns around M&A announcement day are used to measure how institutional ownership is associated with firm decision-making and thus able to create higher shareholder value. In the market-model adjusted return the expected return is computed based on a single factor market model. The parameters of the market model, i.e. [alpha] and [beta], are estimated using Ordinary Least Square (OLS) regression over the estimation period. This method allows controlling the relation between stock returns and market returns; it takes into account variation of risk associated with different stocks. The market-model-adjusted return is commonly found as an expected return in previous event studies (Bonnier and Bruner, 1989; Lummer and McConnell, 1989; Schipper and Thompson, 1983; Homan, 2006; Small et al., 2007).

Following prior studies such as Brown and Warner (1985), Jeng, Metrick, and Zeckhauser (2003) and Moeller, Schlingemann, and Stulz (2004), this paper uses CRSP value-weighted return as a proxy for the market return and estimate alpha and beta for each firm over a period of 200 days preceding announcement, from 210^th to 11^th day prior announcement. Cumulative Abnormal Returns (CAR) is measured over period of 5 trading days surrounding the announcement, from 2 trading days preceding announcement to 2 trading days after announcement, with event day 0 being announcement day. If the announcement day falls on a non-trading day (such as a

9

weekend, or when stock is withhold from trading due to announcement), then CAR is computed for a total of 4 trading days, 2 trading days before announcement and first 2 trading days the stock has been traded after the announcement.

The average 5-day CAR for whole sample is equal to 1.159% and the mean is significantly different from 0 at the 1% level. For transactions where all cash consideration was offered average CAR is equal to 1.446%, significantly different from 0 at the 1% level. For transactions where stock or combination of stock and other type of payment was offered average CAR is equal to 0.717%, significantly different from zero at the 10% level¹. Mean CARs for all cash and stock or partial stock payment are significantly different from each other at the 1% level. Those results are consistent with previous studies of Moeller, Schlingemann, and Stulz (2004), Masulis, Wang, and Xie (2007), and Chen, Harford, and Li (2007).

Institutional Ownership

To investigate the influence of institutional ownership presence on M&A announcement returns, this paper uses aggregated total institutional ownership, measured as number of a firm’s stocks in possession of all types of institutional investors in relation to all shares outstanding at the end of the quarter directly preceding acquisition announcement². Nonetheless, this paper expects that the relationship between institutional ownership and acquirer returns is not linear. Therefore, in the

1 In case of acquisitions financed entirely by stock (287 cases) mean CAR equals to -1.226%, significantly different from all cash acquisition mean CAR at the 1% level.

2 In data set provided by Thomson Reuters’ SDC Ownership Datafeed, some cases show number of shares held by institutional investors that outnumbers shares outstanding. This is due to the fact that the latest number of shares outstanding is not available, therefore the number from last reported period is shown instead. All those cases where the issue persists were excluded from the sample.

10

second approach, the institutional ownership variable is replaced with high institutional ownership indicator. Two models are built, one in which high institutional ownership indicator equals to one when level of ownership is higher than median of the whole sample, and second in which the indicator equals to one when level of ownership is greater than 80^th percentile of the whole sample.

In regard to different types of institutional investors that are examined to test Hypothesis 2, aggregated institutional ownership for each of the type (as provided by Thomson Reuters’ Global Equity Ownership Datafeed) is used. Chen, Harford, and Li (2007) and Ferreira and Matos (2008) classify institutions according to the potential for business ties to a corporation as independent and versus grey institutions. It should be noted however, that, as Ferreira and Matos (2008) warn, relying on institutional categories to classify institutions on their activism is not perfect. This is due to the facts that (1) investor is assigned to an institutional category in which it holds the largest part of its assets under management, yet it may manage several investment vehicles simultaneously – such as bank trusts and mutual funds; (2) there are differences across different countries with respect to the definition of institutional categories; (3) and there are different degrees of potential business ties among institutional investors such as bank trusts and mutual funds. Therefore, this paper draws upon Fung and Tsai (2012) methodology, which examines four different categories of institutional investors, separated by level of independence. Those categories are as follows:

a) Investment advisors, (together with mutual funds) which are thought to be better equipped, more active and more independent thus better suited to

11

perform effective management monitoring than other institutional investors (Brickley, Lease, and Smith (1988), Almazan, Hartzell, and Starks (2005) and Chen, Harford, and Li, (2007)).

b) Hedge funds, which also can be classified as independent investors. Brav, Jiang, Partnoy, and Thomas (2008) as well as Clifford (2008) have found that activist hedge funds are better informed and propose strategic, operational, and financial remedies to firms whose equity they own. Nevertheless, hedge funds enjoy less regulation and are more likely to be less effective monitors than independent advisors due to their active trading.

c) Pension funds, which are pursuing an active monitoring role. However their effectiveness and motivations are being questioned. For example, findings by Qiu (2004) suggest that public pension funds play important roles in corporate governance and are associated with a firm’s lower M&A activity. Del Guercio and Hawkins (1999) showed heterogeneity across funds in activism objectives, tactics, and impact on target firms; while some funds act as independent investors, others can be classified as grey investors.

d) Banks and insurance companies, which are often considered as grey institutions;

they are more “pressure-sensitive” or loyal to corporate management (Ferreira, Matos (2008)). For instance, Brickley, Lease, and Smith (1988) have found that banks and insurance companies tend to be more supportive for firm management than other types of institutional investors in antitakeover amendment proposals. Therefore, they are less likely to perform effective

12 monitoring and governance functions.

Other control variables

Similarly to Masulis, Wang, and Xie (2007), control for two types of factors that might have influence on an acquirer’s return are used in the models. The first one, which is connected to an acquirer’s characteristics, includes firm size, Tobin’s Q, free cash flow and leverage; all of which are measured at the fiscal-year end preceding the acquisition announcement. The second one, related to deal characteristics, consist of relative deal size and consideration used.

Bidder characteristics

Firm Size. According to Moeller, Schlingemann, and Stulz’s (2004) findings, acquirer’s size is negatively correlated with its announcement period abnormal returns, with the announcement return for acquiring firm stock roughly two percentage points higher for small acquirers. This might be evidence in support of Roll’s (1986) managerial hubris hypothesis, as it shows that, on average, larger acquirers pay higher premiums and are more likely to make acquisitions that generate economically significant negative synergies. A large size of a firm might also be an effective takeover defense, as more resources are needed to acquire such a company, hence providing managers with more freedom to indulge in shareholders-value-destroying activities, such as questionable acquisitions. In the model logarithm of an acquirer’s total assets is used as firm size variable.

Tobin’s Q. Existing literature shows diverse results regarding bidder’s Tobin’s Q influence on its CAR around acquisition announcement. On one hand, Lang, Stulz, and

13

Walking (1991) showed a positive impact of Tobin’s Q on tender offers acquisitions’ CAR, and Servaes (1991) documented such a positive relationship in the case of public firm acquisitions. On the other hand, using a comprehensive sample of acquisitions, Moeller, Schlingemann, and Stulz (2004) found a negative relation between acquirer’s Tobin’s Q and CAR. Tobin’s Q is computed by dividing the acquirer’s market value of assets by its book value of assets; market value of assets is computed as the book value of assets minus the book value of common equity plus the market value of common equity (total shares outstanding are multiplied by its price).

Free Cash Flows and Financial Leverage. In his theoretical paper, Jensen (1986) explains the role that debt plays in motivating organizational efficiency. In firms with large free cash flows, which mean more cash than profitable investment opportunities, conflict between managers and shareholders might become even more severe. Payout to shareholder’s means less resources under management control thus reduced managerial power. Managers of firms with large free cash flows have incentive to grow firms beyond their optimal size, which includes questionable acquisition activities. On other hand, higher leverage is incentive for managers to be more cautious not to make the company fall into financial distress, as that might also mean them losing their jobs (Gilson (1989 and 1990)). Therefore, financial leverage plays an important governance role and motivates firm management to improve firm performance. This suggests that higher financial leverage will be positively correlated to an acquirer’s CAR. However, free cash flows might also be a sign of better performing managers, which would imply that they can also make better acquisition decisions. Therefore the free cash flows

14

correlation with an acquirer’s CAR remains unclear and might be either negative or positive. Leverage is computed by dividing the sum of a firm’s market value of assets and short-term debt by market value of its total assets. Free cash flows are computed as operating income before taxes minus interest expense, income taxes, and capital expenditures, divided by the book value of total assets.

Deal characteristics

Relative Size. Asquith, Bruner, and Mullins (1983) documented that an acquirer’s returns are significantly related to the relative size of the target firm. Moeller, Schlingemann, and Stulz (2004) found the same relation, however for their subsample of large acquirer’s an opposite relation was found, too. Relative size is computed as a ratio of deal value to acquirer’s market value four weeks prior to announcement.

Consideration. Existing literature extensively documents method of payment influence on stock market’s reaction to acquisition announcement. Myers and Majluf (1984) developed “information content hypothesis” which predicts that stock-financed acquisitions are signaling to investors that firm stock is overvalued. According to Jensen’s (1986) “free cash flows” hypothesis, acquisitions paid by cash reduce the agency costs of free cash flows. A recent paper by Fung, Jo, and Tsai (2009) has found that value-destroying acquisitions are more likely to be financed with stock and during periods of high stock market valuation. Consequently, stock financed acquisitions can be expected to be associated with lower abnormal returns, while cash financed transactions can be related to higher abnormal returns. In the model, indicator variable for all-cash-financed acquisitions and stock-financed acquisitions is used and it equals to

15

zero when transaction is financed with cash only and equals to one if transaction is wholly or partially financed with stock.

All of the M&As used in the models are eventually completed ones. Asquith, Bruner, and Mullins (1983) have found that deals that are completed are associated with higher returns.

Table II presents summary statistics for each of the above variables. The average total institutional ownership of the sample is 47%, compared to 48% in Ferreira and Matos (2008) and 43% in Fung and Tsai (2012). Averages and medians for equity held by investment advisors are equal to 20% and 17% respectively, and are lower than those in Fung and Tsai’s (2012) sample – 30% and 34% respectively. This difference might be due to the lower investment advisor ownership level in companies that are pursuing M&A activities. For other investor categories, average levels of ownership resemble those of Fung and Tsai (2012).

Year dummies. Jarrell and Bradley (1980) and Schipper and Thompson (1981) suggested necessity of partition of merger analysis by time periods. This is due to the fact that there might be change in legal restrictions as well as general level of merger activity through time that might have influence on bidder’s returns.