Printer Friendly

An empirical examination of the stated purposes of issuer tender offers to purchase common stock.

This paper documents the purposes of issuer tender offers to repurchase stock, as stated in Securities' and Exchange Commission (SEC) disclosures, over the period 1994-2006. We explore whether stated purposes relate to announcement period returns and find returns are significantly lower when repurchases replace dividends, distribute cash from unspecified sources, or occur" subsequent to third-party tender offers. Announcement period returns are significantly higher when repurchases are viewed by management as the best investment opportunity available or when they occur subsequent to previous repurchase programs. Finally, we find evidence in support of signaling theory and Jensen's (1986) agency cost o f free cash flow theory.

The Personal Investing Section of the May 1964 issue of Fortune magazine begins with an article on the "aging" of stock in the United States. (1) In 1963, for the first time since World War II, more US corporate stock was retired than was issued. The article bemoans the declining importance of the New York Stock Exchange (NYSE) as a conduit for capital flows into markets and suggests the retirement of stock is probably related to mergers, treasurers that consider undervalued stocks attractive buys, stock option plans, and buyouts. Since then, a litany of academic research has sought to explain the motives and consequences of common stock repurchase transactions.

Early academic literature largely attributes the increase in repurchase activity during the 1960s to cash surpluses resulting from extended periods of profitability and declining investment opportunity sets. Repurchases were seen as a tax-advantaged alternative to dividends when distributing cash to equity holders (Brigham, 1964; Ellis, 1965; Guthart, 1965; Bierman and West, 1966). Indeed, as Elton and Gruber (1968) formalize a model of the tax advantage of repurchases over dividends, Chirelstein (1969) argues that the tax advantage of repurchases over dividends created by Internal Revenue Code Section 302 was unintended by Congress, but exploited by managers nonetheless.

While repurchases during the early 1960s appear driven largely by cash surpluses, they also affect the right-hand side of the balance sheet. Repurchases increase leverage and do so to a greater extent when they are debt financed. Accordingly, academics suggest both capital structure (Brigham, 1964; Ellis, 1965; Woods and Brigham, 1966; Bierman and West, 1966) and corporate control considerations (Woods and Brigham, 1966) contribute to the explanation of the prevalence of repurchases. Early research also considers the effect of repurchases on the price to earnings ratio. While Brigham (1964) predicts market indifference, Ellis (1965) suggests repurchases are used to avoid dilution when repurchased shares are not retired by the corporation but rather are issued upon option exercises, for bonuses as stock dividends, or in exchange for other convertible securities. Even so, Bierman and West (1966) suggest these types of repurchases account for only a fraction of total repurchases. Share repurchases may also be used instead of newly issued shares to avoid dilution when effecting mergers consummated with stock-for-stock swaps (Ellis, 1965).

Signaling theory, long attributed to Dann (1981 ) and Vermaelen (1981), also has roots dating back to the 1964 article in Fortune magazine, which suggests a repurchase can occur " ... simply because it looked like a good buy to corporate treasurers--i.e., when they thought the market was seriously undervaluing their stocks.... "Brigham (1964) presents a case study of American Standard that also foreshadows signaling theory by surmising, "Though it cannot be proved conclusively, we conclude that the stock purchase had a significant bearing on the relatively strong performance of American Standard's common stock." However, academics and practitioners of the day do not always agree. For example, Southworth (1965) argues that repurchases are negative events as they signal managements' inability to find positive net present value projects in which to invest.

As far back as Ellis (1965) and Stevenson (1966), academics have called for adequate disclosure such that no conflicts of interest exist between management, debt holders, tendering shareholders, and nontendering shareholders. Even so, during periods in which the Securities and Exchange Commission (SEC) essentially left tender offers unregulated, evidence suggests repurchases did not result in significant wealth transfers across security classes (Masulis, 1980a) or between tendering and nontendering shareholders (Dann, 1981). Further, Dann (1981) and Vermaelen (1981) suggest higher insider ownership is associated with stronger positive announcement period returns around repurchase announcements.

Continuing research largely expands on the hypotheses and evidence first set forth in the 1960s. Although open market repurchases account for more than 90% of both the number and dollar value of all repurchases (Grullon and Ikenberry, 2000), this paper seeks to contribute to this substantial body of literature by re-examining the motivation behind the relatively infrequent issuer tender offer method of repurchasing common stock. (2) We do so because, due to stringent disclosure requirements and fundamental implementation differences, tender offers have the potential to convey relatively more information to market participants (Ofer and Thakor, 1987). For instance, the reporting requirements for open market repurchase programs are lax. Companies can repurchase shares without announcing they are doing so ex ante (Brockman, Khurana, and Martin, 2008). In stark contrast, the SEC requires timely public disclosures of both tender offer program initiations and program specifics. In addition to strict disclosure rules, tender offers appear to be undertaken with a greater sense of urgency than do open market repurchases. Firms generally repurchase a larger fraction of shares over a shorter time period through tender offers relative to open market repurchases. Additionally, shares bought through tender offers, unlike those repurchased via open market transactions, are repurchased at a premium above prevailing market prices. Not surprisingly, abnormal stock returns around tender offers are generally found to be higher than around open market program initiation announcements. For example, for tender offers over the period 1970-1999, Anderson and Dyl (2004) find the median percentage of shares sought is 17.32, the median tender offer premium is 20.55%, and the median three-day market-adjusted announcement period return is 10.73%. In contrast, for open market repurchases over the period 1980-1997, Grullon and Michaely (2004) find the median percentage of shares sought is 5.00 and the median three-day market-adjusted announcement period return is 1.82%. By definition, the premium paid in open market repurchase transactions is zero.

We document the stated purposes of a sample of tender offers over a 13-year period ending December 31, 2006 using publicly available information from SEC filings. Further, we empirically test whether these proclamations relate to tendering firms' announcement period returns. Within our sample of 226 issuer tender offers, we identify 16 unique reasons given by management as to why firms engage in repurchase activity. The reasons range from quite specific to extremely vague. Our regressions indicate that some of these stated purposes do, in fact, relate to announcement period returns. Specifically, if a repurchase is the best investment opportunity or is used to continue a history of repurchase activity, announcement period returns are significantly higher. If, however, a tender offer is used to replace dividends or is used to distribute cash from an unspecified source, announcement period returns are significantly lower. Further, we find evidence that supports both Jensen's (1986) agency cost of free cash flow theory and signaling theory.

The remainder of the paper is organized as follows. Section I extends the literature review presented in the introduction to include more recent developments related to theories of common stock repurchase by tender offer. Section II describes the sample of Dutch auction and fixed-price tender offers, their stated purposes, and tendering firm characteristics. Section III details the methodologies used in our empirical analyses, while Section IV presents the results of those analyses. Section V presents our conclusions.

I. Literature Review

Jensen (1986) defines free cash flow as "... cash flow in excess of that required to fund all projects that have positive net present values ..." and suggests cash payouts to shareholders can reduce agency conflicts between owners and managers. Researchers have long acknowledged tender offers are but one mechanism through which cash distributions can be effected and have compared and contrasted tender offers versus open market share repurchases (Baker, Gallagher, and Morgan, 1981; Vermaelen, 1981; Asquith and Mullins, 1986; Gay, Kale, and Noe, 1991), specially designated dividends (Asquith and Mullins, 1986; Howe, He, and Kao, 1992; Chhachhi and Davidson, 1997; Lie, 2000), and regular dividend increases (Guay and Harford, 2000; Lie, 2000). Further, issuer tender offers take two forms, fixed price or Dutch auction, and the differences between the two forms have been analyzed (Comment and Jarrell, 1991 ; Gay et al., 1991 ; Kamma, Kanatas, and Raymar, 1992; Lee, Mikkelson, and Partch, 1992; Persons, 1994; Lie and McConnell, 1998). (3)

Fama and French (2001) demonstrate that the proportion of dividend-paying firms declines dramatically throughout the 1980s and 1990s. (4) Indeed, Grullon and Michaely (2002, 2004), Skinner (2008), and Hsieh and Wang (2009a) present evidence that repurchases surpassed dividends as the predominant form of payout at some point in the late 1990s to early 2000s. At least part of this increased tendency to substitute repurchases for dividends appears related to regulatory change. Specifically, Rule 10b-18, adopted by the SEC in 1982, effectively nullifies the stock price antimanipulation provisions of the Securities and Exchange Act of 1934 that relate to share repurchases (Grullon and Michaely, 2002; Hsieh and Wang, 2009a). Although the Tax Reform Act of 1986 raised the capital gains tax rate (Barclay and Smith, 1988; Lie and Lie, 1999) and the Jobs and Growth Tax Relief Reconciliation Act of 2003 reduced the tax rate on dividend income such that the marginal tax rates on dividends and capital gains are equal (Morck and Yeung, 2005), nontendering shareholders may still prefer repurchases to dividends because taxes on capital gains are not paid until shares are sold and gains are realized. (5) In support of the tax clientele hypothesis, Lie and Lie (1999) find tender offers are preferred to dividends when shareholders' dividend tax rates exceed capital gains tax rates. Additionally, Hsieh and Wang (2008) determine that firms with higher insider ownership (and higher implied tax liabilities) are associated with a greater propensity to distribute cash payouts via repurchase rather than through dividends.

The nature of the underlying cash "flow" also appears to play at least some part in the choice of distribution mechanism. Operating cash flow, cash windfalls, and cash stockpiles accumulated from continuing operations can all be considered free cash flows under Jensen's (1986) agency theory. Not surprisingly, researchers have measured the relationship of each type of cash flow to the choice of distribution method. For example, Hausch and Seward (1993), Guay and Harford (2000), and Lee and Rui (2007) find repurchases are used to distribute transient (i.e., stochastic) rather than permanent (i.e., deterministic) cash flow shocks. (6) Brennan and Thakor (1990) and Lucas and McDonald (1998) suggest tender offers are the preferred distribution method for the largest cash distributions. Not surprisingly, Lie (2000) finds self-tendering firms have funds in excess of industry norms prior to announcements. Further, Lie (2002) finds both stores of cash and operating cash flows are positively related to the probability of undertaking a nondefensive self-tender offer. Confirming Southworth's (1965) contention that repurchases signal a lack of positive net present value projects in which to invest, Grullon and Michaely (2004) find tendering firms subsequently reduce not only their cash reserves but also their capital and research and development expenditures. Similarly, Nohel and Tarhan (1998) find support for the free cash flow hypothesis and suggest tender offers are used to shrink the size of the firm. More recently, Wang et al. (2009) use Tobin's Q to proxy underinvestment and find share repurchases can reduce the agency costs associated with free cash flows.

In addition to distributing cash, issuer share repurchases also concentrate the ownership of nontendering shareholders and thereby have the potential to signal inside information to markets. Signaling theory, as it relates to share repurchase program announcements, largely begins when Masulis (1980a) finds positive announcement period returns around tender offers, Dann (1981) finds the positive returns do not come at the expense of other security classes, and Vermaelen (1981) finds the positive returns increase with increases in tender offer premiums, the fraction of shares sought, and the fraction of shares held by insiders. Despite the aforementioned tax-related benefits of repurchases over dividends, Brennan and Thakor (1990) suggest less informed investors may prefer dividends. More recent evidence supports this notion and, in contrast to Dann (198 I) and Masulis (1980a), suggests nontendering shareholders may expropriate value from both tendering shareholders (Brockman et al., 2008) and bond holders (Maxwell and Stephens, 2003).

The models of Ofer and Thakor (1987) and Chowdhry and Nanda (1994) suggest firms choose repurchases over dividends when their stock is sufficiently undervalued, that is, when the intrinsic value of their common stock exceeds the market value by a significant margin. D'Mello and Shroff (2000) present empirical evidence supporting this notion. Dann, Masulis, and Mayers (1991) find tender offers are a source of positive earnings related information, which is evidenced by decreases in both subsequent unanticipated earnings announcements and equity market risk. Reflecting the transitory nature of the earnings surprise (and cash distribution), Hertzel and Jain (1991) find tender offers affect predominantly short-term forecast revisions. Since the low bid premiums in Dutch auctions are generally lower than the bid premiums in fixed-price tender offers, Comment and Jarrell (1991) suggest fixed-price tender offers are stronger signals of firm undervaluation. Not surprisingly, Lee et al. (1992) find insiders increase personal ownership prior to fixed-price tender offers but not prior to Dutch auctions. However, Lie and McConnell (1998) find no difference between the subsequent operating performance of fixed-price tender offers and Dutch auctions and, as such, no empirical justification for differential announcement period returns.

While free cash flow theory focuses on stores and flows of cash and signaling theory focuses on the release of inside information, researchers have also considered the effect of tender offer share repurchases on the composition of the right-hand side of the balance sheet. Indeed, Koch and Shenoy (1999) and DeAngelo and DeAngelo (2006) suggest payout policy and capital structure decisions are simultaneously determined. Masulis (1980b) finds evidence consistent with corporate debt tax shield effects around pure capital structure changes. Masulis (1980a) finds announcement period returns are higher for tender offers that are more than 50% debt financed. Hovakimian, Opler, and Titman (2001) suggest firms have a tendency to make capital structure decisions that result in a movement toward a target debt ratio and that equity repurchases can be used to effect such changes. Lie (2002) specifically tests whether tender offers are used to optimize capital structure and provides evidence that firms initiate self-tender offers when ex ante debt levels are below predicted levels. Specifically, nondefensive tender offers increase debt ratios to predicted levels while defensive tender offers increase debt ratios beyond predicted levels. Note that capital considerations may play a part in defensive tender offers. For example, Harris and Raviv (1988) suggest leverage is one of many "resistance techniques" used to eschew hostile takeover threats while Stulz (1988) suggests managers may make capital structure decisions that increase managerial controlled voting rights toward the same end. Further, Sinha (1991) reports that debt-financed repurchases can serve a dual purpose in deterring takeovers by 1) making the firm more expensive to potential acquirers and 2) reducing agency conflicts as in Jensen (1986).

In more recent years, researchers have considered the effect of repurchase activity on earnings per share (EPS) dilution, but these analyses have focused on open market rather than tender offer stock repurchases (Jolls, 1998; Weisbenner, 2000; Kahle, 2002). At the center of the dilution hypothesis is the denominator of diluted EPS, which is adjusted to reflect unconverted convertible debt, unexercised warrants, unexercised options, and other potential common share issuances. By repurchasing shares, managers can undo dilution caused by employee stock option programs. Notably, Bens et al. (2003) find repurchases increase when employee stock options increase and when earnings are insufficient to achieve desired EPS growth. Interestingly, only the number of unexercised top management options is positively and significantly related to the number of shares repurchased.

Clearly, issuer tender offers can be proactive measures taken to distribute cash, signal inside information, effect capital structure changes, manage dilution, or some combination of the like. However, as noted above, tender offer repurchases can also be reactive measures undertaken to prevent third-party takeovers. For example, Denis (1990) finds share repurchases are effective at thwarting hostile corporate control activity, although the cash flow ownership of management does not appear to be concentrated. Dittmar (2000) suggests repurchases are effective at deterring takeovers but are only used during periods of active corporate control activity. Further, Billett and Xue (2007) note that open market repurchases can reduce the probability of future takeover attempts, but tender offers are more likely to be used to thwart actual takeover attempts. The theoretical models of Gay et al. (1991) and Persons (1994) suggest Dutch auctions are more effective takeover deterrents than fixed-price tender offers as Dutch auctions make targets more expensive to outside bidders. Lee et al. (1992) analyze managerial trading around stock repurchases and find managers increase buying and decrease selling during the six months preceding fixed-price tender offers but not preceding Dutch auctions. However, when repurchase offers are responses to takeover threats, only preoffer decreases in selling are found.

II. Sample

Issuer tender offer filings dating back to 1994 are publicly available through the SEC's website at http://idea.sec.gov. We use LexisNexis to search for disclosures of issuer tender offer announcements to purchase common stock among Williams Act related schedules filed with the SEC and then access the documents through the SEC website. From the beginning of 1994 through the end of 1999, issuer tender offer statements are reported on Schedule 13E4. Beginning January 1, 2000, issuer tender offers are reported on Schedule TO-I. Accordingly, we search through 2006 for documents using the string "form type: sc 13e4" and "common stock" or "form type: sc to i" and "common stock" as appropriate for the given year. The initial search yields 6,148 filings by 1,369 firms. We delete redundant filings and those related to subsequent announcements (i.e., amendments to initial offers, etc.), offers to purchase or exchange securities other than common stock, and offers to purchase investment companies or trusts, American depository shares, financial firms, and utilities. The result is a sample of 353 announcements of issuer tender offers to purchase shares of common stock.

To perform our empirical analysis, we require sample firms have data available through Standard & Poor's Research Insight and through the Center for Research in Security Prices (CRSP). We are able to match 321 firms to Research Insight using the Central Index Key (CIK) identifier from the tender offer announcements but drop 41 of these firms due to incomplete information. Of the remaining 280 firms, 270 are listed in CRSR Fifteen of these firms delist prior to the initial tender offer announcement. We remove 16 offers through which multiple classes of common stock are targeted for purchase. From the remaining 239 observations, we remove three related to third-party tender offers incorrectly filed as issuer tender offers, two of which the source cash is from a new shareholder (i.e., third-party tender offers in disguise), and two that we are unable to locate the original tender offer announcement for. Finally, we delete six firms for which proxy statements are not available for the fiscal year-end immediately preceding the original tender offer announcement.

Table I lists the final sample of 226 issuer tender offers to purchase common stock by year and type. The sample consists of 141 Dutch auctions and 85 fixed-price tender offers. The fewest number of tender offers in a single year is nine, in both 1994 and 1995, and the most in a single year is 24, in 1998. There is little difference between the number of Dutch auctions and fixed-price tender offers most years, although considerably more Dutch auctions occur during the 1996-1998 period and in years 2005 and 2006.

A. Deal Characteristics

The SC 13E4 and SC TO-I filings that detail issuer tender offers to purchase common shares of equity explicitly state the number of shares sought and the price the issuer is willing to pay. A single price is stated for fixed-price offers, while a range of prices is stated for Dutch auctions. We follow Vermaelen (1981) and calculate the tender offer premium as the difference between the tender offer price and the closing stock price five days prior to the initial announcement of the tender offer standardized by the closing stock price five days prior to the offer for fixed-price offers. As seen in Panel A of Table II, the average and median premiums offered in fixedprice tender offers are 18.59% and 12.61%, respectively. We follow Lie (2000) to calculate the tender offer premium for Dutch auctions as the difference between the maximum tender offer price and the closing stock price five days prior to the initial announcement of the tender offer standardized by the closing stock price five days prior to the offer. The average and median premiums offered in Dutch auctions are 16.81% and 13.63%, respectively. Unlike previous studies, such as Comment and Jarrell (1991) and Lie and McConnell (1998), we do not find significant differences between premiums offered in Dutch auctions and premiums offered in fixed-price tender offers. However, our finding of "no difference" in bid premiums is consistent with evidence that earnings improvements subsequent to fixed-price and Dutch auction tender offers do not differ (Lie and McConnell, 1998).

The number of shares sought to total common shares outstanding at the time of the offer averages 35.40% for fixed-price tender offers. The median is 25.76%. For Dutch auctions, the average (median) shares sought is 16.64% (13.40%). As demonstrated in Panel A of Table II, fixed-price offers target a significantly larger percentage of shares for purchase than do Dutch auctions. The difference in the average shares sought, 18.76%, is significantly different from zero at the 0.01 level. Further, the difference in the averages does not appear driven by outliers as the (Wilcoxon) signed test statistic is significant at the 0.01 level. Again, our findings differ from previous studies, such as Comment and Jarrell (1991), Kamma et al. (1992), and Lie and McConnell (1998), who find little or no differences between shares sought in fixed-price and Dutch auction tender offers. (7) Taken with our finding of no significant difference between bid premiums in Dutch auction and fixed-price tender offers, it appears that, over time, fixed-price tendering firms have become more adept at projecting shareholder participation and are now more inclined to seek a larger proportion of shares outstanding for repurchase rather than "overpaying" for those shares sought (i.e., relative to Dutch auctions and holding total cash payouts constant). This apparent shift in policy implies fixed-price tendering firms consciously consider whether to buy more shares or pay more for the repurchased shares. (8)

Equally important to the tender premium and the percentage of shares sought, from a signaling perspective, is the percentage of insider ownership. If the premium offered over prevailing stock prices exceeds the true value of the stock, tendering shareholders can effectively expropriate funds from nontendering shareholders (Vermaelen, 1981). As greater (nontendering) inside ownership forces insiders to carry a large portion of the burden of expropriation, greater inside ownership can verify the signal sent via tender offer. Additionally, Babenko (2009) indicates that share repurchases can force employees to bear more risk by increasing the pay-performance sensitivity of their compensation packages. To measure the percentage ownership of insiders, we collect stock ownership information for the top five executives for each tendering firm from the proxy statement immediately preceding the tender offer event. Interestingly, inside ownership at firms in our sample, which spans the period 1994-2006, does not differ substantially from the tendering firms in Vermaelen's (1981) study that spans the period 1962-1977. The average (median) values for our sample firms are 16.94% (7.43%), which are only slightly lower than Vermaelen's (1981) sample firms whose average (median) inside ownership is 17.5% (8.8%). Consistent with Comment and Jarrell (1991), we find significant differences between fixed-price and Dutch auction tender offers. The average ownership of the top five insiders at firms that engage in fixed-price repurchases, 20.20%, is significantly higher than the average ownership at firms with Dutch auction tender offers, 14.98%, at the 0.05 level. Similarly, the signed test statistic is also significant at the 0.05 level.

B. Third-Party Tender Offers and Going Private Transactions

As noted, an active market for corporate control can precipitate tender offers. We search tender offer announcements for signs the offers are reactions to third-party tender offers. The number reported as defensive, however, is considerably lower than expected, only seven in all. We expand our search to include third-party tender offers (Schedules 14D9, 14D1E and TO-T) and communications thereof (Schedules 14D9C and TO-C) filed during the three months preceding the initial issuer tender offer announcements. Panel A of Table III presents the number of thirdparty tender offers for the full sample and for the subsamples of Dutch auctions and fixed-price tender offers. As shown, 85 of the 226 issuer tender offers occur after third-party tender offers. Interestingly, and counter to the prediction of the model of Persons (1994), the number of Dutch auctions that occur after overt third-party tender offers is only marginally more than that of fixedprice offers, 44 versus 41. However, recall that we do not find differences between the maximum bid premium in Dutch auction tender offers and the bid premium in fixed-price tender offers (see above). Thus, the Dutch auctions in our sample do not make targets more expensive to outside bidders as the models of both Gay et al. (1991) and Persons (1994) suggest.

The ultimate signal by insiders is the decision to take the firm private. Tender offers used to take firms private may or may not be in response to third-party tender offer activity. We search the tender offer announcements and find 20 firms explicitly state at least one of the purposes of the tender offer is to take the firm private or to get the stock delisted. A search of SEC disclosures related to going private transactions, filed on Schedule 13E3, reveals 16 sample firms went private within three months after the original tender offer announcement. Nine of the 16 firms file the tender offer statement and the going private schedule on the same day. Sample firms that go private choose fixed-price tender offers to do so almost exclusively. Interestingly, only 11 firms that went private indicate going private as a purpose in the initial tender offer announcement. Nineteen of the 20 firms that list getting the stock delisted or going private as a purpose for the tender offer use fixed-price tender offers, while 15 of 16 firms that actually go private after tender offers use fixed-price tender offers to effect the transactions (see below). These findings, at least in part, help explain our finding that significantly more shares are sought through fixed-price tender offers than through Dutch auction tender offers.

C. Stated Purposes of Tender Offers

While many researchers have hypothesized, modeled, and tested competing theories of stock repurchases, few seek direct input from tendering firms' management. Baker et al. (1981), Wansley, Lane, and Sarkar (1989), Tsetsekos, Kaufman, and Gitman (1991), and Baker, Powell, and Veit (2003) fill this gap by surveying managers for their reasons for repurchasing stock. However, the vast majority of responses from these surveys involve open market repurchasing firms. Baker et al. (1981) include six, Wansley et al. (1989) include 22, and Tsetsekos et al. (1991) include six tender offers in their respective studies. Further, because Baker et al. (2003) receive survey responses from only 22 tendering firms, they analyze only open market repurchases. Rather than survey managers, we search the initial tender offer announcements for the stated purposes of the repurchase programs. Specifically, we search within each Schedule 13E4 and TO-I announcement, which includes a section titled Purpose of the Offer or something of the like. To date, we know of no other study that collects purposes of tender offers from this publicly available source. While Lie (2002) characterizes these stated purposes as vague and, as such, uninteresting and uninformative, we find these statements range from exceedingly vague to quite specific. Further, because the stated purposes are embedded in announcements long characterized as informative by finance researchers, we seek to determine whether they also signal meaningful information to market participants. (9) Beyond the intentions of thwarting hostile takeovers and taking firms private, we find the purpose sections of the sample tender offer statements contain at least 14 additional reasons for the tender offers. Panel B of Table III lists the purposes in descending order of frequency of their occurrence. The reasons are not mutually exclusive; the median number of listed purposes is between five and six.

As illustrated in Panel B of Table III, six of the 16 listed purposes occur in more than half of the sample firms. The most common purpose, cited in 180 of 226 instances, is to allow stockholders to avoid transaction costs and/or commissions associated with normal share dispositions. Providing liquidity to shareholders is often cited in close proximity to comments related to transaction cost avoidance, although this stated purpose occurs in only 120 instances. (10) The second most commonly cited reason, to increase EPS or to concentrate ownership of future earnings, also seems vague and uninformative, yet it is cited in 174 of the 226 tender offer announcements. The third most stated purpose, the repurchase is the best investment opportunity or is a prudent use of available resources, may seem vague on the surface but does imply management has considered alternative investments and believes returns from repurchasing shares dominate other endeavors. This purpose is listed in 166 of the 226 tender offers. Maximizing shareholder value is a particularly vague reason for repurchasing stock, yet over two-thirds of the tender offer announcements include this as a purpose for repurchasing stock, 155 firms in all. We find evidence congruent with signaling in almost 60% of the tender offers; 134 of the 226 announcements explicitly state the reason for the tender offer is the stock is undervalued by the market.

The remaining 10 purposes for repurchasing common stock are cited much less frequently than the first six reasons discussed above. Among these purposes, we find support for optimal capital structure theory. As demonstrated in Panel B of Table III, 79 firms cite capital restructuring as a purpose for the tender offer. Dutch auctions are used to effect the capital structure changes more often than fixed-price tender offers by a five-to-two margin. Sixty-two firms, 27.4% of the sample, state the tender offer is a continuation of a history of repurchase activity. Since the history of repurchase activity is known ex ante, this purpose may seem uninformative. Even so, it conveys information related to the reluctance of shareholders to tender shares in previous tender offers. (11)

Confirming theory first set forth in the 1960s, five of the remaining eight purposes relate to cash distributions. Interestingly, the more vague the stated purpose, the more often it is cited and vice versa. For example, 42 tender offers state the reason for the repurchase is to distribute cash without specifying the source of the cash. Less commonly, tender offers cite stores of cash accumulated from normal operations, 36 firms representing 15.9% of the sample, and windfalls of cash from asset dispositions, 30 firms representing 13.3% of the sample. When the source of the cash is cited, Dutch auctions are the predorninate form of repurchase. Thirty-two of the 36 firms that use repurchases to distribute cash from operations use Dutch auctions as do 21 of the 30 firms that distribute cash from asset dispositions.

Two purposes related to cash distributions are among the most specific reasons cited but are also among the most infrequently cited. Repurchases are cited as tax-preferred cash distributions eight times and are announced in conjunction with dividend terminations six times. While similar on the surface, the two purposes appear to be motivated differently and are likely to relate differently to announcement period returns. That is, repurchases as a tax-preferred distribution appear demand side driven as they relate to a preference of certain shareholders for capital gains tax treatment. As such, we expect a positive effect on announcement period returns. However, when repurchases replace dividends, a steady stream of cash distributions is replaced by a transient cash distribution, and we expect a negative effect on announcement period returns.

As noted above, tender offers are used to deter hostile takeovers and to take firms private or to get the stock delisted. We find that 20 firms in our sample explicitly state that the purpose for the tender offer is to get the stock delisted or to take the firm private. Fixed-price tender offers are used almost exclusively toward this end. in 19 of the 20 cases. Surprisingly, given our findings that 85 of our sample firms are in play when the issuer tender offer is announced, only seven firms list takeover deterrence as a purpose for the tender offer. Finally, we find 17 firms use repurchased shares to effect business combinations, with fixed-price tender offers the preferred method of repurchase used toward this end.

III. Methodologies

A. Announcement Period Returns

Researchers dating back to Masulis (1980a) have documented positive stock returns around issuer tender offer announcements. Additionally, theories of stock repurchases are often tested by attempting to explain variations in announcement period returns. We follow this research design by first estimating announcement period returns and then using cumulative announcement period returns as the dependent variable in regression analyses. We use standard event study methodology to measure abnormal announcement period returns by first estimating the market model over the 250 day period ending 10 days prior to the initial tender offer announcement for each firm using the equally weighted CRSP index as the market. We then use the coefficient estimates and actual market returns to calculate the expected return for each firm on the day before, the day of, and the day after the initial tender offer announcement. Next, we calculate each daily abnormal return as the difference between the actual return and the expected return. Finally, the three-day cumulative announcement period abnormal return (CAR) is the product of one plus each daily abnormal return, minus one. As demonstrated in Panel B of Table II, the average three-day announcement period return is 6.44% and the median is 2.79%. We find no significant differences between average CARs around fixed-price and Dutch auction tender offers. However, signed test statistics, significant at the 0.10 level, suggest Dutch auction tender offers are associated with only marginally higher CARs. These findings, which differ from evidence set forth by Comment and Jarrell (1991) and Lie and McConnell (1998) but are congruous with Kamma et al. (1992), echo our findings of no significant difference between the bid premiums associated with Dutch auctions and fixed-price tender offers.

B. Firm Characteristics

We next compile firm-specific information. Jensen's (1986) agency cost of free cash flow theory suggests tendering firms have higher ex ante cash flows. Accordingly, we measure both stockpiles of cash and operating cash flow at the most recent fiscal year-end preceding the tender offer announcement. We use cash and equivalents (Compustat Item #1) to proxy stockpiles of cash and the Lehn and Poulsen (1989) measure of free cash flow to proxy operating cash flow. Specifically, free cash flow is measured as operating income before depreciation (Compustat Item #13) minus total income taxes (Compustat Item #16) plus the change in deferred taxes from the previous year (the one-year change in Compustat Item #35) minus gross interest expense (Compustat Item #15) minus dividends on preferred and common stock (Compustat Items #19 and #21, respectively). We standardize both measures by the market value of assets, which we calculate as the market value of common equity (Compustat Item #199 times Item #25) plus the book value of preferred stock (Compustat Item #130) plus the book value of liabilities (Compustat Item #181) minus cash (Compustat Item #162).

As reported in Panel B of Table II, the average stockpile of cash at the full sample of tendering firm is 23.35% of the market value of the firm. Not surprisingly, this seemingly large value appears inflated by outliers; the median stockpile of cash is 5.18% of the market value of the firm. Firms that initiate fixed-price repurchases appear to have larger stockpiles of cash than firms that initiate Dutch auction repurchases. The average (median) stockpile at fixed-price tendering firms, 32.15% (7.47%), is significantly higher than the average (median) stockpile at Dutch auction tendering finns, 18.05% (3.97%), at the 0.10 (0.05) level. For the full sample of tendering firms, operating cash flows average 3.60% of the market value of the firm. The median is somewhat higher at 6.81%. As indicated in Panel B of Table III, no significant differences exist between operating cash flow at fixed-price tendering versus Dutch auction tendering firms.

We measure the debt ratio at the most recent fiscal year-end preceding the tender offer announcement to assess ex ante leverage at our sample of tendering firms. The debt ratio is defined as the book value of total liabilities (Compustat Item #181) as a percentage of the book value of total liabilities plus the book value of preferred stock (Compustat Item #130) plus the market value of equity (Compustat Item #199 times Item #25). As demonstrated in Panel B of Table II, the average debt ratio at sample firms is 36.39% and the median is 35.04%. The average debt ratio at fixed-price tendering firms is no different than the debt ratio of Dutch auction tendering firms, and we observe only marginal differences in median values.

To assess how the management of dilutive shares relates to CARs, we calculate the ratio of dilutive common shares outstanding to basic common shares outstanding using values from the most recent fiscal year-end preceding the tender offer. Specifically, we use the difference between common shares used to calculate diluted EPS (Compustat Item #171) and common shares used to calculate basic EPS (Compustat Item #54) standardized by common shares used to calculate basic EPS. Compustat Item #171 is not reported for a significant number of firms prior to 1997. We approximate Item #171 in these instances by multiplying common shares used to calculate basic EPS times the ratio of basic EPS (Compustat Item #58) to diluted EPS (Compustat Item #57). As shown in Panel B of Table II, dilutive shares average 2.65% of basic shares outstanding. The median is 1.07%. We find no significant differences between dilutive shares at fixed-price versus Dutch auction tendering firms.

We next calculate the market-to-book value of equity to proxy the ex ante growth prospects of the firm. Again, we use information from the most recent fiscal year-end preceding the tender offer announcement. Specifically, the market-to-book value of equity is calculated as the market value of common equity (Compustat Item #199 times Item #25) divided by the book value of common equity (Compustat Item #60). The average and median values of the market-to-book value of equity for the full sample of firms are nearly identical, 1.55 times and 1.56 times, respectively. While we find no significant differences between the average market-to-book ratios between fixed-price and Dutch auction tendering firms, the signed test statistic, significant at the 0.05 level, suggests market-to-book ratios are indeed higher at Dutch auction tendering firms. Taken together with our finding that fixed-price tendering firms have larger ex ante stockpiles of cash, the results imply fixed-price tendering firms are in the later stages of the firm life cycle. That is, as firms approach maturity, they have fewer positive net present value projects and, as such, larger stockpiles of cash. To test this issue further, we calculate a rough measure of firm age as the number of years each firm's stock trades publicly prior to each tender offer announcement and test for significant differences between the subsets of tendering firms. However, in contrast to our prediction, we find the common stocks of fixed-price tendering firms trade publicly for a significantly shorter amount of time than those of Dutch auction tendering firms. Specifically, the average (median) number of years fixed-price tendering firms trade prior to the tender offer is 13.01 (9.64), while the average (median) for Dutch auction tendering firms is 19.02 (14.04) years. The t-test and signed test statistic for differences between the two groups are both significant at the 0.01 level.

Panel B of Table II presents the average sample firm's book value of assets, measured at the most recent fiscal year-end preceding the tender offer announcement, which is $1,504.93 million. The median firm size is $313.65 million. Similar to previous research (Comment and Jarrell, 1991; Kamma et al., 1992; Lie and McConnell, 1998), we find firms that engage in Dutch auction repurchases are significantly larger than firms that engage in fixed-price repurchases. The average (median) book value of total assets of the average Dutch auction tendering firm is $1,953.67 ($480.78) million versus $760.55 ($134.17) million at fixed-price tendering firms. The t-test for differences in the means and the signed test statistic are both significant at the 0.01 level. (12)

C. Determinants of Announcement Period Returns

We next regress CARs on the aforementioned firm and tender offer characteristics to test competing theories of tender offers. We then add indicator variables to the regression to test whether or not the stated purposes included in the initial tender offer statements relate to market participants' reactions to the tender offer announcements. Specifically, we estimate variations of the following full form equation: (13)

[CAR.sub.i] = [a.sub.1] + [b.sub.1][PREMIUM.sub.i] + [b.sub.2] [SHRSGHT.sub.i] + [b.sub.3] [INSIDE.sub.i] + [b.sub.4] ln[(BVA).sub.i] + [b.sub.5][DUTCH.sub.i] + [b.sub.6] [DEFENSE.sub.i] + [b.sub.7] [CASH.sub.i] + [b.sub.8] [FREE.sub.i] + [b.sub.9] [DEBT.sub.i] + [b.sub.10] [DIL.sub.i] + [b.sub.11] [MKBK.sub.i] + [summation over 15 (n=1)] [c.sub.n] [PURPOSEn,i + [e.sub.i],

where

i = company identifier,

CAR = the three-day cumulative abnormal returns surrounding the initial tender offer announcement,

PREMIUM = the bid price minus the closing stock price five days prior to the announcement standardized by the closing stock price five days prior to the announcement,

SHRSGHT = shares sought as a percentage of shares outstanding,

INSIDE = percentage of shares owned by the top five executives,

ln(BVA) = the one-year lag of the natural log of total year-end assets,

DUTCH = an indicator variable equal to one for Dutch auction tender offers, and zero otherwise,

DEFENSE = an indicator variable equal to one for takeover deterrent tender offers, and zero otherwise,

CASH = the one-year lag of cash and equivalents divided by the market value of total assets,

FREE = the one-year lag of the Lehn and Poulsen (1989) free cash flow measure standardized by the market value of total assets,

DEBT = the one-year lag of the debt ratio = the book value of debt/(the book value of debt + the market value of equity),

DIL = the one-year lag of EPS dilutive shares held for conversion into common stock as a percentage of basic common shares outstanding,

MKBK = the one-year lag of the market-to-book value of equity,

[PURPOSE.sub.n] = an indicator variable equal to one if purpose n is cited as a reason for the tender offer, and zero otherwise,

a, b, c = the coefficients to be estimated, and

e = the error term.

IV. Results

A. The Base Models

Table IV lists the results of our initial series of regressions. Column 2 displays results from the first regression, which establishes our base model in which in addition to year fixed effects and a constant, we include the tender offer premium, the percentage of shares sought, the percentage ownership of the top five executives, and firm size as explanatory variables. Similar to Vermaelen (1981), we find CARs positively relate to the tender offer premium (the coefficient on PREMIUM is 0.319, t-statistic = 5.34) and to the percentage of inside ownership (the coefficient on INSIDE is 0.068, t-statistic = 2.81). Contrary to the findings of Vermaelen (1981), we find CARs negatively relate to shares sought to total shares outstanding (the coefficient on SHRSGHT is -0.042, t-statistic = -2.13). However, when additional controls are added, the SHRSGHT coefficient is no longer significant. The base model explains a significant amount of the variation in CARs; the adjusted [R.sup.2] is 42.5%.

We next add indicator variables to capture CAR variations related to the choice of repurchase form (i.e., Dutch auction or fixed price) and whether or not the tender offer is a response to a third-party tender offer. Column 3 of Table IV lists the results. Unlike previous research, we do not find Dutch auction announcements are associated with more positive CARs than fixedprice tender offers. While positive, the associated coefficient is not significant at conventional levels (the coefficient is 0.015, t-statistic = 1.27). Defensive tender offers are significantly negatively related to CARs; the coefficient on DEFENSE, -0.037, is significant at the 0.01 level (t-statistic = -3.51). With the exception of the shares sought variable, the base model variable coefficients remain significant. Thus, we find support for the signaling hypothesis; that is, large tender offer premiums and high insider ownership appear to signal positive news to market participants. Further, when tender offers are defensive in nature, announcement period returns are lower. While we have no definitive proof, we surmise this is likely because third-party tender offers bid up stock prices, thereby diluting the signals of issuer tender offers.

We next expand the base set of explanatory variables to test alternate theories of issuer tender offers. Column 4 of Table IV lists the results. As reported, both the size and the significance of the coefficients from the previous models remain relatively unchanged. CARs are positively related to the tender offer premium (the coefficient on PREMIUM is 0.312, t-statistic = 5.31) and to insider ownership (the coefficient on INSIDE is 0.060, t-statistic = 2.64) and are negatively related to tender offers defensive in nature (the coefficient on DEFENSE is -0.041, t-statistic = -3.82). We find support for Jensen's (1986) agency cost of free cash flow theory. The coefficients on the variables that measure stores of cash and operating free cash flow are both positive and significant. Specifically, the coefficient on CASH is 0.038, t-statistic = 3.05, and the coefficient on FREE is 0.104, t-statistic = 2.38. The finding that higher levels of cash and equivalents as a percentage of the market value of the firm relate positively to announcement period stock returns most directly supports Jensen's (1986) prediction of agency cost resolution. However, the size of the coefficient associated with operating free cash flows is more than two-and-one-half times that of the cash store coefficient. (14) The literature of the 1960s and Jensen's (1986) agency cost of free cash flow theory rationalize repurchases as a mechanism through which excess cash is distributed. Guay and Harford (2000) find tender offers are used to distribute transient rather than recurring cash flows. Since operating cash flows are recurring by nature, the finding that operating cash flows are associated with more positive announcement period returns is somewhat surprising. We suggest the signal associated with the concentration of ownership (i.e., by non-tendering shareholders) of the high cash flows relative to the market value of the firm may best explain our results. That is, the significance of the coefficient on FREE appears to support both signaling and agency theory. We find no evidence that either capital structure theory or dilution theory relates to CARs as the coefficients associated with DEBT and DIL are insignificant at conventional levels. Finally, the markets' ex ante assessment of the growth prospects of the firm does not relate to variations in CARs as the coefficient on MKBK is not significant at traditional levels.

B. Stated Purposes of Tender Offers and Announcement Period Returns

Next, we test whether the stated purposes of issuer tender offers relate to CARs holding tender offer and firm characteristics constant. Tables V and VI present results of regressions that examine each stated purposes individually (i.e., when added to the extended baseline model of Table IV, column 4). Table VII presents the results of regressions that examine the relationships between the stated purposes and announcement period returns holding the other stated purposes (as well as firm and tender offer characteristics) constant.

Table V presents the results of regressions that include the most commonly stated purposes. Specifically, in separate regressions, we add each of the six stated purposes that are cited in more than 50% of the tender offers: 1) the purpose is to avoid transactions costs, 2) the purpose is to increase EPS, 3) the tender offer is the best investment opportunity, 4) the purpose is to maximize shareholder value, 5) the stock is undervalued, and 6) the tender offer provides liquidity to shareholders. Five out of the six indicator variables have insignificant regression coefficients. In three of the cases, this is not surprising. All tender offers provide tendering shareholders with liquidity and the opportunity to avoid transactions costs and concentrate ownership of nontendering shareholders, thus increasing EPS. Therefore, these stated purposes are uninformative to markets and have no discernible relationship with announcement period returns. Additionally, purposes related to signaling (i.e., maximizing shareholder value and market participants undervalue the stock) do not relate to announcement period returns. The only indicator variable coefficient significant at conventional levels is that associated with the tender offer being the best investment opportunity for the firm. As demonstrated in column 4 of Table V, the coefficient on the best investment opportunity purpose is 0.030, significant at the 0.01 level. Similar to the aforementioned interpretation of the operating cash flow variable, this significant coefficient can support both agency theory and signaling theory. That is, this stated purpose implies management has considered alternative investments and believes the rate of return from repurchasing undervalued shares dominates other endeavors and, as such, is the best use of available cash and/or available debt capacity. Finally, we note that the size and significance of the extended base model variable coefficients remain relatively unchanged.

Table VI presents the results of regressions that include the relatively infrequently stated purposes. Specifically, in separate regressions, we add the nine stated purposes that are cited in less than 50% of the tender offer announcements 1) to facilitate a capital restructuring, 2) to continue a history of repurchase activity, 3) to distribute cash from an unspecified source, 4) to distribute cash from normal operations, 5) to distribute cash from asset dispositions, 6) to get the stock delisted, 7) to facilitate a business combination, 8) the tender offer is a tax preferred cash distribution over dividends, and 9) the tender offer is used in lieu of dividends that are simultaneously canceled. As with the more commonly cited purposes, the majority of the less frequently stated reasons for repurchasing shares through tender offers do not significantly relate to CARs. Six of the nine indicator variable coefficients are not significant at traditional levels. Two of the coefficients are marginally significant and only one is significant at better than the 0.05 level. CARs are marginally significant and higher when tender offers are a continuation of a history of repurchase activity (the coefficient is 0.021, t-statistic = 1.80). Additionally, CARs are marginally significant and lower when the purpose of the tender offer is to distribute cash from an unspecified source (the coefficient is -0.027, t-statistic = -1.88). Only in the least frequently cited case is the indicator variable coefficient significant at better than the 0.05 level. CARs are significantly lower when the repurchase is announced in conjunction with the termination of dividends (the coefficient is -0.088, t-statistic = -2.00).

Until now, we have considered relationships between each purpose and announcement period returns in relative isolation (i.e., without regards to other cited purposes). However, as noted above, the purposes are not stated in isolation. The median number of purposes is between five and six. Consequently, we examine the relationships between CARs and the stated purposes holding other stated purposes constant. The results of our analyses are presented in Table VII. In column 2, we add the six more frequently cited purposes to the extended base regression of Table IV, column 4. In column 3, we add the nine less frequently cited purposes, while in column 4, we add all 15 purposes. As demonstrated in column 2, the best investment opportunity purpose coefficient remains significant and positive even when holding other frequently cited purposes constant (the coefficient is 0.031, t-statistic = 2.55). Further, the size and significance of the extended base model coefficients remain relatively unchanged. However, the improvement in [R.sup.2] values (over the extended base model of Table IV, column 4) is not significant when the six frequently cited variables are added (F-statistic = 1.563, p-value = 0.160), but is significant when only the best investment variable is added (F-statistic = 6.967, p-value = 0.009).

As indicated in column 3 of Table VII, when the nine less frequently cited purposes are added to the extended base regression, the two marginally significant variable coefficients become significant at better than the 0.05 level. Specifically, the coefficient on the indicator variable associated with tender offers that are a continuation of historical repurchase activity is 0.022, t-statistic = 2.03, and the coefficient on the indicator variable associated with cash distributions from unspecified sources is -0.031, t-statistic = -2.05. Further, the coefficient on the indicator variable associated with repurchases that replace dividends remains negative and significant (the coefficient is -0.100, t-statistic = -2.49). Until this point, the coefficients on PREMIUM, INSIDE, and DEFENSE have changed little when indicator variables related to the purpose of the tender offers are added to the extended base regression. However, when the nine less frequently cited purposes are all added to the regression, the coefficient on INSIDE is no longer significant at conventional levels. We find significant Pearson correlation coefficients between the ownership of the top five executives and purposes related to capital restructuring ([rho] = 0.220, significant at the 0.01 level) and delisting the stock ([rho] = -0.211, significant at the 0.01 level). Thus, it appears multicollinearity may be driving this result. Notably, adding the infrequently cited purposes to the extended base regression statistically improves the model. The change in [R.sup.2] values is significant when the nine variables are added to the extended base model of Table IV, column 4 (F-statistic = 2.710, p-value = 0.005).

Finally, we add all 15 stated purposes for issuer tender offers to the extended base regression and report the results in column 4 of Table VII. The results support both Jensen's (1986) agency cost of free cash flow and signaling theories. Even when holding firm characteristics and stated purposes constant, the coefficient associated with the tender offer premium remains positive and strongly significant (the coefficient is 0.348, t-statistic = 6.51) and the coefficients associated with both stores of cash and flows of cash remain positive and significant (the coefficient on CASH is 0.036, t-statistic = 3.05, the coefficient on FREE is 0.103, t-statistic = 2.23). Further, CARs are significantly lower when tender offers are defensive in nature (the coefficient on DEFENSE is -0.034, t-statistic = -3.14). As in the previous regression, the coefficient associated with inside ownership is not significant at conventional levels (the coefficient on INSIDE is 0.034, t-statistic = 1.52). However, for the first time, we find the coefficient on the ex ante growth proxy is marginally significant (the coefficient on MKBK is 0.006, t-statistic = 1.68). Although statistically significant, we note that since the absolute size of the MKBK coefficient is relatively small, the economic effect of the relationship is likely minimal. The coefficient on the indicator variable associated with repurchases that replace dividends is negative and significant at the 0.05 level (the coefficient is -0.095, t-statistic = -2.21). Finally, the coefficients on the remaining three previously significant variables are now significant only at the 0.10 level. Specifically, the coefficient on best investment opportunity is 0.023, t-statistic = 1.96; the coefficient on continuing histories of repurchases is 0.020, t-statistic = 1.87; and the coefficient on distributing cash from unspecified sources is -0.028, t-statistic = 1.86.

V. Conclusions

Previous research suggests signaling, cash surpluses, capital structure considerations, and the market for corporate control can relate to the decision to initiate issuer tender offers to purchase common stock. We add to this substantial body of literature by documenting the purposes of a sample of tender offers, as stated by management in SEC-mandated disclosures, over the period 1994-2006. We identify 16 unique reasons for repurchase activity in our sample of 226 tender offers and find the reasons range from quite specific to exceedingly vague. We empirically examine these proclamations by assessing their relationships with stock returns around the tender offer announcements.

Regression results suggest stated purposes of tender offers significantly relate to announcement period returns. Specifically, announcement period returns are significantly higher when the share repurchase is the best investment opportunity for the firm or when the tender offer is a continuation of a history of repurchase activity. If, however, the tender offer is used to replace dividends or is used to distribute cash from an unspecified source, announcement period returns are significantly lower. Similar to previous research, we find announcement period returns are significantly lower when tender offers are defensive in nature, although managers are generally reluctant to cite third-party tender offers as reasons for share repurchases.

Unlike previous research (Comment and Jarrell, 1991; Lie and McConnell, 1998), we find no differences between fixed-price and Dutch auction tender offer premiums. Also contrary to previous research, we find fixed-price tender offers target a significantly larger number of shares than do Dutch auction tender offers. These results suggest that, over time, fixed-price tendering firms have become more adept at projecting shareholder participation. That is, fixed-price tendering firms are now more inclined to seek a larger proportion of shares outstanding (and thus concentrate nontendering shareholder ownership even further) rather than overpaying for those shares sought. We provide additional evidence in support of both signaling theory and Jensen's (1986) agency cost of free cash flow theory. Specifically, we find inside ownership and tender offer premiums positively and significantly relate to announcement period returns. Announcement period returns are also significantly higher when tendering firms have larger stores of cash and larger operating cash flows prior to the tender offer. Finally, we find no evidence that tender offers are used to optimize capital structure or are used to undo EPS dilution caused by unexercised options, warrants, or unconverted convertible debt.
Appendix: Non-categorical Variable Definitions

This table lists the technical definitions of all noncategorical
variables. Compustat items in are in bold.

Announcement period returns (CARS) = Three-day cumulative abnormal
 announcement period returns,
 centered on the day of the
 initial issuer tender offer
 announcement, are calculated
 from daily abnormal returns
 estimated as differences
 between actual returns and
 market model estimates. For
 the market model, alpha
 and beta are estimated over the
 250-day period ending 10 days
 prior to the initial tender
 offer announcement using the
 CRSP equally weighted index as
 the market.

Tender offer premium (PREMIUM) = For Dutch auction (fixed-price)
 tender offers, the maximum
 tender offer premium is the
 maximum (stated) tender offer
 premium divided by the closing
 stock price five days prior to
 the initial tender offer
 announcement.

Shares sought to shares outstanding The number of common equity
(SHRSGT) = shares sought in the tender
 offer as a percentage of total
 common shares outstanding at
 the time of the offer.

Market value of assets (MVA) = The one-year lag of the market
 value of common equity (the
 fiscal year-end closing price
 [#199] times the number of
 common shares outstanding [#25])
 plus the book value of preferred
 stock (#130) plus the book value
 of liabilities (#181) minus cash
 (#162).

Cash and equivalents to market The one-year lag of cash and
value of assets (CASH) = cash equivalents (#1)
 standardized by MVA.

Operating cash flows to market The one-year lag of free cash
value of assets (FREE) = flow per Lehn and Poulsen (1989)
 (operating income before
 depreciation [#13] minus total
 income taxes [#16] plus the
 change in deferred taxes from
 the previous year [one-year
 change in #35] minus gross
 interest expense [#15] minus
 dividends on preferred and
 common stock [#19 + #21])
 standardized by MVA.

Debt ratio (DEBT) = The one-year lag of the debt
 ratio; the book value of total
 liabilities (#181) as a
 percentage of the book value of
 total liabilities (#181) plus
 the book value of preferred
 stock (#130) plus the market
 value of equity (#199 x #25).

Dilutive shares to total shares The one-year lag of earnings per
(DIL) = share (EPS) diluting shares; the
 number of common shares used to
 calculate diluted EPS (#171)
 minus the number of common
 shares used to calculate basic
 EPS (#54) standardized by the
 number of common shares used to
 calculate basic FPS (#54) (we
 set negative values to zero).
 To approximate #171 when missing
 (which is often prior to 1997),
 we multiply common shares used
 to calculate basic EPS (#54)
 times the ratio of basic EPS
 (#58) to diluted EPS (#57).

Market-to-book value of equity The one-year lag of the
(MKBK) = market-to-book ratio; the market
 value of common equity (#199 x
 #25) divided by the book value
 of common equity (#60).

ln(BVA) = The one-year lag of the natural
 log of total year-end assets
 (#6).


We thank Bill Christie (Editor) and an anonymous referee for helpful comments that substantially improved this paper.

References

Amromin, G., P Harrison, and S. Sharpe, 2008, "How Did the 2003 Dividend Tax Cut Affect Stock Prices?" Financial Management 37, 625-646.

Anderson, A. and E. Dyl, 2004, "Determinants of Premiums on Self-Tender Offers," Financial Management 33, 25-45.

Asquith, P. and D. Mullins, Jr., 1986, "Signaling with Dividends, Stock Repurchases, and Equity Issues," Financial Management 15, 27-44.

Babenko, I., 2009, "Share Repurchases and Pay-Performance Sensitivity of Employee Compensation Contracts," Journal of Finance 64, 117-150.

Baker, H., R Gallagher, and K. Morgan, 1981, "Management's View of Stock Repurchase Programs," Journal of Financial Research 4, 233-247.

Baker, H., G. Powell, and E. Veit, 2003, "Why Companies Use Open-Market Repurchases: A Managerial Perspective," Quarterly Review of Economics and Finance 46, 483-504.

Barclay, M. and C. Smith, Jr., 1988, "Corporate Payout Policy: Cash Dividends versus Open-Market Repurchases," Journal of Financial Economics 22, 61-82.

Bens, D., V. Nagar, E Skinner, and M. Wong, 2003, "Employee Stock Options, EPS Dilution, and Stock Repurchases," Journal of Accounting and Economics 36, 51-90.

Bierman, H., Jr. and R. West, 1966, "The Acquisition of Common Stock by the Corporate Issuer," Journal of Finance 21, 687-696.

Billett, M. and H. Xue, 2007, "The Takeover Deterrent Effect of Open Market Share Repurchases," Journal of Finance 62, 1827-1850.

Brav, A., J.R. Graham, C.R. Harvey, and R. Michaely, 2008, "Managerial Response to the May 2003 Dividend Tax Cut," Financial Management 37, 611-624.

Brennan, M. and A. Thakor, 1990, "Shareholder Preferences and Dividend Policy," Journal of Finance 45, 993-1018.

Brigham, E., 1964, "The Profitability of a Firm's Purchase of Its Own Common Stock," California Management Review 7, 69-76.

Brockman, E, I. Khurana, and X. Martin, 2008, "Voluntary Disclosures around Share Repurchases," Journal of Financial Economics 89, 175-191.

Chhachhi, I. and W. Davidson III, 1997, "A Comparison of the Market Reaction to Specially Designated Dividends and Tender Offer Stock Repurchases," Financial Management 26, 89-96.

Chirelstein, M., 1969, "Optional Redemptions and Optional Dividends: Taxing the Repurchase of Common Shares," Yale Law Journal 78, 739-756.

Chowdhry, B. and V. Nanda, 1994, "Repurchase Premia as a Reason for Dividends: A Dynamic Model of Corporate Payout Policies," Review of Financial Studies 7, 321-350.

Comment, R. and G. Jarrell, 1991, "The Relative Signaling Power of Dutch Auction and Fixed-Price Self-Tender Offers and Open-Market Share Repurchases," Journal of Finance 46, 1243-1271.

Dann, L., 1981, "Common Stock Repurchases: An Analysis of Returns to Bondholders and Stockholders," Journal of Financial Economics 9, 113-138.

Dann, L., R. Masulis, and D. Mayers, 199 I, "Repurchase Tender Offers and Earnings Information," Journal of Accounting and Economics 14, 217-251.

DeAngelo, H. and L. DeAngelo, 2006, "The Irrelevance of the MM Dividend Irrelevance Theorem," Journal of Financial Economics 79, 293-315.

DeAngelo, H., L. DeAngelo, and D. Skinner, 2004, "Are Dividends Disappearing? Dividend Concentration and the Consolidation of Earnings," Journal of Financial Economics 72, 425-456.

Denis, D., 1990, "Defensive Changes in Corporate Payout Policy: Share Repurchases and Special Dividends," Journal of Finance 45, 1433-1456.

Dittmar, A., 2000, "Why Do Firms Repurchase Stock?" Journal of Business 73, 331-355.

D'Mello, R. and P. Shroff, 2000, "Equity Undervaluation and Decisions Related to Repurchase Tender Offers: An Empirical Investigation," Journal of Finance 55, 2399-2424.

Ellis, C., 1965, "Repurchase Stock to Revitalize Equity," Harvard Business Review 43, 119-128.

Elton, E. and M. Gruber, 1968, "The Effect of Share Repurchase on the Value of the Firm," Journal of Finance 23, 135-149.

Fama, E. and K. French, 2001, "Disappearing Dividends: Changing Firm Characteristics or Lower Propensity to Pay?" Journal of Financial Economics 60, 3-43.

Fortune, 1964, "Personal Investing: Why Is There a Stock Market? Aging Stock," May, 75-76.

Gay, G., J. Kale, and T. Noe, 1991, "Share Repurchase Mechanisms: A Comparative Analysis of Efficacy, Shareholder Wealth, and Corporate Control Effects," Financial Management 20, 44-59.

Grullon, G. and D. Ikenberry, 2000, "What Do We Know about Share Repurchases?" Journal of Applied Corporate Finance 13, 31-51.

Grullon, G. and R. Michaely, 2002, "Dividends, Share Repurchases, and the Substitution Hypothesis," Journal of Finance 57, 1649-1684.

Grullon, G. and R. Michaely, 2004, "The Information Content of Share Repurchase Programs," Journal of Finance 59, 651-680.

Guay, W. and J. Harford, 2000, "The Cash-Flow Permanence and Information Content of Dividend Increases versus Repurchases," Journal of Financial Economics 57, 385-415.

Guthart, L., 1965, "More Companies Are Buying Back Their Stock," Harvard Business Review 43, 40-53.

Harris, M. and A. Raviv, 1988, "Corporate Control Contests and Capital Structure," Journal of Financial Economics 20, 55-86.

Hausch, D. and J. Seward, 1993, "Signaling with Dividends and Share Repurchases: A Choice between Deterministic and Stochastic Cash Disbursements," Review of Financial Studies 6, 121-154.

Hertzel, M. and P Jain, 1991, "Earnings and Risk Changes around Stock Repurchase Tender Offers," Journal of Accounting and Economics 14, 253-274.

Hovakimian, A., T. Opler, and S. Titman, 2001, "The Debt-Equity Choice," Journal of Financial and Quantitative Analysis 36, 1-24.

Howe, K., J. He, and G. Kao, 1992, "One-Time Cash Flow Announcements and Free Cash-Flow Theory: Share Repurchases and Special Dividends," Journal of Finance 47, 1963-1975.

Hsieh, J. and Q. Wang, 2008, "Insiders' Tax Preferences and Firms' Choices Between Dividends and Share Repurchases," Journal of Financial and Quantitative Analysis 43, 213-244.

Hsieh, J. and Q. Wang, 2009a, "Stock Repurchases: Theory and Evidence, Part 1," in H. Baker, Ed., Dividends and Dividend Policy, Hoboken, N J, John Wiley and Sons, Inc.

Hsieh, J. and Q. Wang, 2009b, "Stock Repurchases: Theory and Evidence, Part 2," in H. Baker, Ed., Dividends and Dividend Policy, Hoboken, N J, John Wiley and Sons, Inc.

Jagannathan, M. and C. Stephens, 2003, "Motives for Multiple Open-Market Repurchase Programs," Financial Management 32, 71-91.

Jensen, M., 1986, "Agency Cost of Free Cash Flow, Corporate Finance and Takeover," American Economic Review 76, 323-329.

Jolls, C., 1998, "Stock Repurchase and Incentive Compensation," NBER Working Paper No. 6467.

Kahle, K., 2002, "When a Buyback Isn't a Buyback: Open Market Repurchases and Employee Options," Journal of Financial Economics 63, 235-261.

Kamma, S., G. Kanatas, and S. Raymar, 1992, "Dutch Auction versus Fixed-Price Self-Tender Offers for Common Stock," Journal of Financial Intermediation 2, 277-307.

Koch, ED. and C. Shenoy, 1999, "The Information Content of Dividend and Capital Structure Policies," Financial Management 28, 16-35.

Kumar, E and B.S. Lee., 2001, "Discrete Dividend Policy with Permanent Earnings," Financial Management 30, 55-76.

Lee, B. and O. Rui, 2007, "Time-Series Behavior of Share Repurchases and Dividends," Journal of Financial and Quantitative Analysis 42, 119-142.

Lee, D., W Mikkelson, and M. Partch, 1992, "Managers' Trading Around Stock Repurchases," Journal of Finance 47, 1947-196l.

Lehn, I. and A. Poulsen, 1989, "Free Cash Flow and Stockholder Gains in Going Private Transactions," Journal of Finance 44, 771-787.

Lie, E., 2000, "Excess Funds and Agency Problems: An Empirical Study of Incremental Cash Disbursements," Review of Financial Studies 13, 219-248.

Lie, E., 2002, "Do Firms Undertake Self-Tender Offers to Optimize Capital Structure?" Journal of Business 75, 609-639.

Lie, E. and H. Lie, 1999, "The Role of Personal Taxes in Corporate Decisions: An Empirical Analysis of Share Repurchases and Dividends," Journal of Financial and Quantitative Analysis 34, 533-552.

Lie, E. and J. McConnell, 1998, "Earnings Signals in Fixed-Price and Dutch Auction Self-Tender Offers," Journal of Financial Economics 49, 161-186.

Lucas, D. and R. McDonald, 1998, "Shareholder Heterogeneity, Adverse Selection, and Payout Policy," Journal of Financial and Quantitative Analysis" 33, 233-253.

Masulis, R., 1980a, "Stock Repurchase by Tender Offer: An Analysis of the Causes of Common Stock Price Changes," Journal of Finance 35, 305-319.

Masulis, R., 1980b, "The Effects of Capital Structure Changes on Security Prices: A Study of Exchange Offers," Journal of Financial Economics 8, 139-178.

Maxwell, W. and C. Stephens, 2003, "The Wealth Effects of Repurchases on Bondholders," Journal of Finance 58, 895-920.

McNally, W., 1999, "Multi-Dimensional Signaling with Fixed-Price Repurchase Offers," Managerial and Decision Economics 20, 131-150.

Morck, R. and B. Yeung, 2005, "Dividend Taxation and Corporate Governance," Journal of Economic Perspectives 19, 163-180.

Nayar, N., A. Singh, and A. Zebedee, 2008, "Share Repurchase Offers and Liquidity: An Examination of Temporary and Permanent Effects," Financial Management 37, 251-270.

Nohel, T. and V. Tarhan, 1998, "Share Repurchases and Firm Performance: New Evidence on the Agency Costs of Free Cash Flow," Journal of Financial Economics 49, 187-222.

Ofer, A. and A. Thakor, 1987, "A Theory of Stock Price Responses to Alternative Corporate Cash Disbursement Methods: Stock Repurchases and Dividends," Journal of Finance 42, 365-394.

Persons, J., 1994, "Signaling and Takeover Deterrence with Stock Repurchases: Dutch Auctions versus Fixed Price Tender Offers," Journal of Finance 49, 1373-1402.

Sinha, S., 1991, "Share Repurchase as a Takeover Defense," Journal of Financial and Quantitative Analysis 26, 233-244.

Skinner, D., 2008, "The Evolving Relation between Earnings, Dividends, and Stock Repurchases," Journal of Financial Economics 87, 582-609.

Southworth, A., Jr., 1965, "From the Thoughtful Businessman," Harvard Business Review 43, 40-42.

Stevenson, R., 1966, "Corporate Stock Reacquisitions," Accounting Review 41, 312-317.

Stulz, R., 1988, "Managerial Control and Voting Rights: Financing Policies and the Market for Corporate Control," Journal of Financial Economics 20, 25-54.

Tsetsekos, G., D. Kaufman, and L. Gitman, 1991, "A Survey of Stock Repurchases Motivations and Practices of Major U.S. Corporations," Journal of Applied Business Research 7, 15-21.

Vermaelen, T., 1981, "Common Stock Repurchases and Market Signaling: An Empirical Study," Journal of Financial Economics 9, 139-183.

Wang, C.S., N. Strong, S. Tung, and S. Lin, 2009, "Share Repurchases, the Clustering Problem, and the Free Cash Flow Hypothesis," Financial Management 38, 487-505.

Wansley, J., W. Lane, and S. Sarkar, 1989, "Managements' View on Share Repurchase and Tender Offer Premiums," Financial Management l8, 97-110.

Weisbenner, S., 2000, "Corporate Share Repurchases in the 1990s: What Role Do Stock Options Play?" Federal Reserve Board Working Paper.

Woods, D. and E. Brigham, 1966, "Stockholder Distribution Decisions: Share Repurchases or Dividends?" Journal of Financial and Quantitative Analysis l, 15-26.

(1) The article, referenced in footnote 1 of Brigham (1964), appears to be the first reference of the existence of a prevalence of share repurchases in the modern academic literature.

(2) See Hsieh and Wang (2009a, 2009b) for a comprehensive overview of theory and empirical studies of all types of common stock repurchases.

(3) In addition to open market repurchases and tender offers, Hsieh and Wang (2009a) note two alternate methods of repurchasing shares of common stock: targeted stock repurchases and transferable put-rights distributions.

(4) In contrast, DeAngelo, DeAngelo, and Skinner (2004) find aggregate real (and nominal) dividends actually increase over the same period. Note, however, that higher levels of aggregate dividend payouts are reflective of increasing dividend concentration at the largest and most profitable firms.

(5) Amromin, Harrison, and Sharpe (2008) do not find the 2003 dividend tax cut affected the value of the overall US stock market. However, they do find some evidence of asset reallocation as high-dividend-paying stocks outperformed low-dividend-paying stocks during their sample period.

(6) Kumar and Lee (2001) suggest firms increase dividends only when increases in earnings are permanent. Further, in Bray et al. (2008), the stability of future earnings is a more important determinant of dividend initiations than the 2003 dividend tax cuts.

(7) Specifically, Kamma et al. (1992) and Lie and McConnell (1998) find significant differences between the means, but not the medians while Comment and Jarrell (1991) find a significant difference between the standard deviation, but not the mean.

(8) See McNally (1999) for an analysis of multidimensional signaling inherent in fixed-price tender offers (i.e., signaling via shares sought and premiums paid).

(9) We thank the reviewer for noting an important caveat to our analysis. Simply requiring managers to disclose the purposes of the tender offers does not guarantee managers disclose their true motives. Take, for example, our finding that only seven of 85 firms "in play" proclaim the tender offer is a response to a third-party tender offer.

(10) Not surprisingly, Nayar, Singh, and Zebedee (2008) find improved market liquidity for targeted shares during the tender offer period for a sample that encompasses a similar time frame as our study.

(11) Jagannathan and Stephens (2003) find stronger positive announcement-period returns around the open market program initiations of infrequent repurchasers than frequent repurchasers.

(12) We find fixed-price tendering firms are smaller, younger, and have greater stores of cash. Although we present no definitive proof, we conjecture that fixed-price tendering firms could have shorter firm life cycles.

(13) We estimate year fixed-effects regressions, calculate t-values using robust standard errors, and winsorize variables at the top and bottom 1% level.

(14) The median value of CASH, 5.81%, is less than the median value of FREE, 6.81%. Thus, the level of the variable does not appear to exacerbate differences in the estimated coefficients.

Jessica Kay Dunn, Alex Fayman, and Jamie John McNutt *

* Jessica Kay Dunn is a Doctoral Student in the College of Business at Southern Illinois University, Carbondale, IL. Alex Fayman is an Assistant Professor of Finance in the College of Business at the University of Central Arkansas, Conway, AR. Jamie John McNutt is an Assistant Professor of Finance in the College of Business at Southern Illinois University. Carbondale. IL.
Table I. Issuer Tender Offers to Purchase Common Stock over the Period
1994-2006

This table lists a sample of issuer tender offer announcements to
purchase common stock. Tender offers are identified using LexisNexis
to search Williams Act filings. Documents are accessed through the
SEC's IDEA website (http://idea.sec.gov).

 Full Sample Fixed Price Dutch Auction
Year (N = 226) (N = 85) (N = 141)

1994 9 5 4
1995 9 3 6
1996 10 1 9
1997 20 6 14
1998 24 5 19
1999 23 12 11
2000 23 13 10
2001 18 11 7
2002 13 7 6
2003 21 9 12
2004 13 5 8
2005 23 5 18
2006 20 3 17

Table II. Descriptive Statistics for a Sample of Firms that Issue
Tender Offers to Purchase Their Own Common Stock over the Period
1994-2006

This table lists descriptive statistics of sample firms and their
issuer tender offers to purchase common stock. Tender offers are
identified using LexisNexis to search SEC Williams Act filings.
Documents are accessed through the SEC's IDEA website
(http://idea.see.gov). Financial data are obtained from Standard
& Poor's Research Insight. Stock price data are derived from the CRSP
database. Column 1 lists short descriptions of the variables;
technical definitions are located in the Appendix. Tests for
significant differences in means are based on unpaired t-tests.
Tests for significant differences in medians are based on
Wilcoxon-Mann-Whitney test statistics.

 Panel A. Deal Characteristics

 Full Sample Fixed Price
 (N = 226) (N = 85)

 Average Median Average Median

Maximum premium 17.48 13.54 18.59 12.61
(percent of stock
price five days
prior to tender
offer)

Shares sought to 23.70 16.30 35.40 25.76
shares outstanding
(%)

Insider ownership 16.94 7.43 20.20 10.48
(%)

 Panel A. Deal Characteristics

 Dutch Auctions Difference (Fixed-
 (N = 141) Dutch Auctions)

 Average Median Average Median

Maximum premium 16.81 13.63 1.78 -1.02
(percent of stock
price five days
prior to tender
offer)

Shares sought to 16.64 13.40 18.76 *** 12.36 ***
shares outstanding
(%)

Insider ownership 14.98 6.65 5.22 ** 3.83 **
(%)

 Panel B. Three-Day Cumulative Abnormal
 Announcement Period Returns and
 Preannouncement Characteristics of
 Repurchasing Firms

 Full Sample Fixed Price
 (N = 226) (N = 85)

 Average Median Average Median

Announcement 6.44 2.79 6.97 1.34
period returns (%)

Cash and 23.35 5.18 32.15 7.47
equivalents
to market value
of assets (%)

Operating cash 3.60 6.81 0.14 6.78
flow market value
of assets (%)

Debt ratio (%) 36.39 35.04 36.39 35.33

Dilutive shares to 2.65 1.07 2.24 0.83
total shares (%)

Market-to-book 1.55 1.56 1.69 1.32
value of equity
(times)

Book value of 1,504.93 313.65 760.55 134.17
assets ($MM)

 Panel B. Three-Day Cumulative Abnormal
 Announcement Period Returns and
 Preannouncement Characteristics of
 Repurchasing Firms

 Dutch Auctions Difference (Fixed-
 (N = 141) Dutch Auctions)

 Average Median Average Median

Announcement 6.13 4.43 0.84 -3.09 *
period returns (%)

Cash and 18.05 3.97 14.09 * 3.50 **
equivalents
to market value
of assets (%)

Operating cash 5.69 6.93 -5.56 -0.15
flow market value
of assets (%)

Debt ratio (%) 36.39 34.92 0.00 0.41

Dilutive shares to 2.90 1.25 -0.66 -0.42
total shares (%)

Market-to-book 1.47 1.75 0.22
value of equity
(times)

Book value of 1,953.67 480.78 -1,193.12 *** -346.61 ***
assets ($MM)

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

* Significant at the 0.10 level.

Table III. Characteristics of Issuer Tender Offers to Purchase Common
Stock over the Period 1994-2006

Tender offers are identified using LexisNexis to search Williams Act
filings. All documents are accessed through the SEC's IDEA website
(http://idea.sec.gov).

 Full Sample Fixed Price Dutch Auction
 (N = 226) (N = 85) (N = 141)

 Count % Count % Count %

 Panel A. Defensive and Going Private Issuer Tender Offers

Third-party tender 85 37.6 41 48.2 44 31.2
offer filed in
90 days preceding
initial issuer
tender offer

Going private 16 7.1 15 17.6 1 0.7
transaction filed
within 90 days
after initial
issuer tender
offer

Panel B. Purpose(s) as Stated in Initial Issuer Tender Offer (Schedule
13E4 or TO-1)

1. Avoid transactions 180 79.6 51 60.0 129 91.5
 costs/commissions

2. Increase earnings 174 77.0 55 64.7 119 84.4
 per share

3. Best investment 166 73.5 49 57.6 117 83.0
 opportunity

4. Maximize 155 68.6 51 60.0 104 73.8
 shareholder value
 in the long term

5. Stock undervalued 134 59.3 40 47.1 94 66.7
 by market

6. Provide liquidity 120 53.1 47 55.3 73 51.8
 to stockholders

7. Capital 79 35.0 22 25.9 57 40.4
 restructuring

8. Continue a history 62 27.4 18 21.2 44 31.2
 of repurchase
 transactions

9. Distribute cash 42 18.6 16 18.8 26 18.4
 from an
 unspecified source

10. Distribute cash 36 15.9 4 4.7 32 22.7
 from normal
 operations

11. Distribute cash 30 13.3 9 10.6 21 14.9
 from asset
 dispositions

12. Get the stock 20 8.8 19 22.4 1 0.7
 delisted/go
 private

13. Facilitate a 17 7.5 15 17.6 2 1.4
 business
 combination

14. Tax preferred 8 3.5 3 3.5 5 3.5
 cash
 distribution

15. Defensive/ 7 3.1 4 4.7 3 2.1
 takeover
 deterrence

16. Terminating 6 2.7 1 1.2 5 3.5
 certain
 dividends

Table IV. Regression Analysis of Announcement Period Returns around 226
Issuer Tender Offers to Purchase Common Stock over the Period 1994-2006

The dependent variable, the three-day cumulative abnormal announcement
period return, is calculated using the market model estimated over the
250 day period ending 10 days prior to the initial tender offer
announcement. Column 1 list short descriptions of the independent
variables while columns 2-4 list regression coefficient estimates from
the five regressions. Technical definitions for noncategorical
variables are located in the Appendix. t-values, shown in parentheses,
are calculated using robust standard errors. We winsorize extreme
observations of each variable.

(1) (2) (3)

Intercept 0.096 0.082
 (1.37) (1.20)
Premium as a percent of stock price 0.319 *** 0.310 ***
five days prior to tender offer (5.34) (5.37)
Shares sought to total shares -0.042 ** -0.013
outstanding (-2.13) (-0.60)
Insider ownership 0.068 *** 0.064 ***
 (2.81) (2.83)
Natural log of book value of total -0.005 -0.004
assets (-1.37) (-1.19)
Dutch auction (1 if Dutch auction 0.015
tender offer, otherwise 0) (1.27)
Defensive (1 if yes, otherwise 0) -0.037 ***
 (-3.51)
Cash and equivalents to market
value of assets
Free cash flow to market value of
assets
Debt ratio

Dilutive to total common shares
outstanding
Market-to-book value of common
equity
Year F-value 2.276 3.096
Adjusted [R.sup.2] 0.425 0.454

(1) (4)

Intercept 0.052
 (0.75)
Premium as a percent of stock price 0.312 ***
five days prior to tender offer (5.31)
Shares sought to total shares -0.024
outstanding (-0.97)
Insider ownership 0.060 ***
 (2.64)
Natural log of book value of total -0.004
assets (-1.17)
Dutch auction (1 if Dutch auction 0.015
tender offer, otherwise 0) (1.21)
Defensive (1 if yes, otherwise 0) -0.041 ***
 (-3.82)
Cash and equivalents to market 0.038
value of assets (3.05)
Free cash flow to market value of 0.104 ***
assets (2.38)
Debt ratio 0.044
 (1.30)
Dilutive to total common shares 0.002
outstanding (0.02)
Market-to-book value of common 0.005
equity (1.25)
Year F-value 2.844
Adjusted [R.sup.2] 0.469

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

Table V. Regression Analysis of Announcement Period Returns around 226
Issuer Tender Offers to Purchase Common Stock over the Period 1994-2006

The dependent variable, the three-day cumulative abnormal announcement
period return, is calculated using the market model estimated over the
250-day period ending 10 days prior to the initial tender offer
announcement. Column I lists short descriptions of the independent
variables while columns 2-7 list regression coefficient estimates from
the six regressions. Technical definitions for noncategorical variables
are located in the Appendix. The t-values, shown in parentheses,
are calculated using robust standard errors. We winsorize extreme
observations of each variable.

(1) (2) (3) (4)

Intercept 0.041 0.054 0.033
 (0.58) (0.79) (0.48)
Premium as a percent of 0.318 *** 0.312 *** 0.320 ***
stock price five tender (5.39) (5.31) (5.54)
offer
Shares sought to total -0.014 -0.025 -0.003
shares outstanding (-0.57) (-0.99) (-0.14)
Insider ownership 0.060 *** 0.060 *** 0.058 ***
 (2.64) (2.62) (2.59)
Natural log of book -0.004 -0.004 -0.005
value of total assets (-1.23) (-1.14) (-1.36)
Dutch auction (1 if 0.013 0.015 0.012
Dutch auction tender
otherwise 0) (1.07) (1.21) (0.98)
Defensive (1 if yes, -0.038 *** -0.041 *** -0.036 ***
otherwise 0) (-3.44) (-3.83) (-3.18)
Cash and equivalents 0.038 *** 0.038 *** 0.036 ***
to market value of (2.99) (3.07) (2.72)
assets
Free cash flow to 0.105 ** 0.104 ** 0.104 **
market value of assets (2.41) (2.37) (2.34)
Debt ratio 0.047 0.043 0.052
 (1.36) (1.22) (1.50)
Dilutive to total -0.012 0.004 0.003
common shares (-0.12) (0.04) (0.03)
outstanding
Market-to-book value 0.005 0.005 0.006
of common equity (1.28) (1.25) (1.32)
Purpose is to avoid 0.016
transactions costs/
commissions
(1 if yes, otherwise 0) (1.28)
Purpose is to increase -0.003
EPS (1 if yes, (-0.22)
otherwise 0)
Purpose is best 0.030 ***
investment opportunity (2.70)
(1 if yes,
otherwise 0)
Purpose is to maximize
shareholder value (1 if
yes, otherwise 0)
Purpose is stock is
undervalued by market
(I if yes, otherwise 0)
Purpose is to provide
liquidity to
shareholders (1 if yes,
otherwise 0)
Year F-value 2.777 2.804 2.816
Adjusted [R.sup.2] 0.470 0.467 0.484

(1) (5) (6) (7)

Intercept 0.057 0.034 0.046
 (0.82) (0.49) (0.64)
Premium as a percent of 0.312 *** 0.312 *** 0.311 ***
stock price five tender (5.30) (5.33) (5.31)
offer
Shares sought to total -0.025 -0.021 -0.023
shares outstanding (-1.03) (-0.82) (-0.94)
Insider ownership 0.059 *** 0.059 *** 0.059 ***
 (2.63) (2.61) (2.63)
Natural log of book -0.004 -0.004 -0.004
value of total assets (-1.19) (-1.00) (-1.13)
Dutch auction (1 if 0.015 0.012 0.015
Dutch auction tender
otherwise 0) (1.22) (1.01) (1.22)
Defensive (1 if yes, -0.041 *** -0.040 *** -0.039 ***
otherwise 0) (-3.81) (-3.80) (-3.70)
Cash and equivalents 0.038 *** 0.038 *** 0.037 ***
to market value of (3.13) (2.99) (2.93)
assets
Free cash flow to 0.105 ** 0.105 ** 0.100 **
market value of assets (2.41) (2.38) (2.23)
Debt ratio 0.044 0.043 0.042
 (1.31) (1.28) (1.23)
Dilutive to total 0.002 -0.004 0.005
common shares (0.02) (-0.04) (0.05)
outstanding
Market-to-book value 0.005 0.005 0.006
of common equity (1.27) (1.18) (1.29)
Purpose is to avoid
transactions costs/
commissions
(1 if yes, otherwise 0)
Purpose is to increase
EPS (1 if yes,
otherwise 0)
Purpose is best
investment opportunity
(1 if yes,
otherwise 0)
Purpose is to maximize -0.005
shareholder value (1 if (-0.39)
yes, otherwise 0)
Purpose is stock is 0.012
undervalued by market (1.23)
(I if yes, otherwise 0)
Purpose is to provide 0.008
liquidity to (0.72)
shareholders (1 if yes,
otherwise 0)
Year F-value 2.794 2.680 2.749
Adjusted [R.sup.2] 0.467 0.470 0.468

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

Table VI. Regression Analysis of Announcement Period Returns around 226
Issuer Tender Offers to Purchase Common Stock over the Period 1994-2006

The dependent variable, three-day cumulative abnormal announcement
period returns, is calculated using the market model estimated over the
250-day period ending 10 days prior to the initial tender offer
announcement. Column 1 lists short descriptions of the independent
variables while columns 2-10 list regression coefficient estimates from
the nine regressions. Technical definitions for noncategorical
variables are located in the Appendix. The t-values, shown in
parentheses, are calculated using robust standard errors. We winsorize
extreme observations of each variable.

(1) (2) (3) (4)

Intercept 0.039 0.066 0.062
 (0.57) (0.96) (0.92)
Premium as a percent 0.318 *** 0.308 *** 0.316 ***
of stock price five (5.52) (5.24) (5.54)
days prior to tender
offer
Shares sought to -0.021 -0.018 -0.028
total shares (-0.87) (-0.75) (-1.19)
outstanding
Insider ownership 0.057 ** 0.054 ** 0.057 **
 (2.51) (2.38) (2.58)
Natural log of book -0.003 -0.005 -0.004
value of total (-0.89) (-1.46) (-1.24)
Dutch auction (1 if 0.016 0.014 0.015
Dutch auction (1.26) (1.20) (1.24)
tender offer
otherwise 0)
Defensive (1 if yes, -0.041 *** -0.039 *** -0.042 * **
otherwise 0) (-3.86) (-3.72) (-4.04)
Cash and equivalents 0.035 *** 0.038 *** 0.042 ***
to market value of (2.81) (2.96) (3.67)
assets
Free cash flow to 0.105 *** 0.105 *** 0.103 ***
market value of (2.35) (2.31) (2.34)
assets
Debt ratio 0.036 0.049 0.036
 (1.04) (1.44) (1.04)
Dilutive to total 0.001 -0.016 -0.023
common shares (0.02) (-0.16) (-0.23)
outstanding
Market-to-book value 0.005 0.006 0.006
of common equipment (1.15) (1.44) (1.42)
Purpose is capital -0.014
restructuring (1 if (-1.41)
yes, otherwise 0)
Purpose is to 0.021
continue a history
of repurchases (1 if
yes,otherwise 0) (1.80) *
Purpose is to -0.027
distribute cash from (-1.88) *
unspecified source
(1 if yes,
otherwise 0)
Purpose is to
distribute cash from
normal operations
(1 yes, otherwise 0)
Purpose is to
distribute cash from
asset dispositions (1
if yes, otherwise 0)
Purpose is to get
the stock delisted
(1 if yes, else 0)
Purpose is to
facilitate a business
combination (1 if
yes, otherwise 0)
Purpose is tax
preferred cash
distribution (1 if
yes, otherwise 0)
Purpose is replace
dividends (1 if yes,
else 0)

Year F-value 2.959 2.766 2.998
Adjusted [R.sup.2] 0.470 0.476 0.479

(1) (5) (6) (7)

Intercept 0.062 0.066 0.055
 (0.88) (0.93) (0.78)
Premium as a percent 0.311 *** 0.314 *** 0.318 ***
of stock price five (5.27) (5.35) (5.50)
days prior to tender
offer
Shares sought to -0.023 -0.029 -0.017
total shares (-0.96) (-1.19) (-0.66)
outstanding
Insider ownership 0.057 ** 0.061 *** 0.063 ***
 (2.52) (2.67) (2.74)
Natural log of book -0.005 -0.005 -0.004
value of total (-1.32) (-1.36) (-1.21)
Dutch auction (1 if 0.014 0.014 0.013
Dutch auction (1.10) (1.19) (1.09)
tender offer
otherwise 0)
Defensive (1 if yes, -0.040 *** -0.039 *** -0.040 ***
otherwise 0) (-3.75) (-3.69) (-3.63)
Cash and equivalents 0.037 *** 0.038 *** 0.038 ***
to market value of (3.04) (3.15) (2.96)
assets
Free cash flow to 0.103 *** 0.105 *** 0.103 ***
market value of (2.34) (2.41) (2.35)
assets
Debt ratio 0.047 0.045 0.047
 (1.38) (1.32) (1.36)
Dilutive to total -0.009 0.008 -0.005
common shares (-0.08) (0.07) (-0.05)
outstanding
Market-to-book value 0.005 0.005 0.005
of common equipment (1.30) (1.28) (1.26)
Purpose is capital
restructuring (1 if
yes, otherwise 0)
Purpose is to
continue a history
of repurchases (1 if
yes,otherwise 0)
Purpose is to
distribute cash from
unspecified source
(1 if yes,
otherwise 0)
Purpose is to 0.012
distribute cash from (1.03)
normal operations
(1 yes, otherwise 0)
Purpose is to 0.015
distribute cash from (1.07)
asset dispositions (1
if yes, otherwise 0)
Purpose is to get -0.019
the stock delisted (-0.88)
(1 if yes, else 0)
Purpose is to
facilitate a business
combination (1 if
yes, otherwise 0)
Purpose is tax
preferred cash
distribution (1 if
yes, otherwise 0)
Purpose is replace
dividends (1 if yes,
else 0)

Year F-value 2.894 2.857 2.838
Adjusted [R.sup.2] 0.469 0.469 0.469

(1) (8) (9) (10)

Intercept 0.052 0.049 0.046
 (0.75) (0.69) (0.67)
Premium as a percent 0.309 *** 0.312 *** 0.333 ***
of stock price five (5.27) (5.33) (6.25)
days prior to tender
offer
Shares sought to -0.016 -0.026 -0.024
total shares (-0.62) (-1.06) (-1.02)
outstanding
Insider ownership 0.059 *** 0.059 *** 0.053 **
 (2.64) (2.62) (2.34)
Natural log of book -0.004 -0.004 -0.004
value of total (-1.17) (-1.15) (-1.01)
Dutch auction (1 if 0.014 0.015 0.015
Dutch auction (1.12) (1.20) (1.23)
tender offer
otherwise 0)
Defensive (1 if yes, -0.039 *** -0.039 *** -0.041 ***
otherwise 0) (-3.70) (-3.70) (-3.93)
Cash and equivalents 0.036 *** 0.039 *** 0.036 ***
to market value of (2.74) (3.17) (3.00)
assets
Free cash flow to 0.099 *** 0.106 *** 0.104 **
market value of (2.20) (2.41) (2.47)
assets
Debt ratio 0.047 0.046 0.036
 (1.35) (1.36) (1.05)
Dilutive to total -0.003 0.006 -0.001
common shares (-0.03) (0.06) (-0.01)
outstanding
Market-to-book value 0.005 0.006 0.004
of common equipment (1.24) (1.35) (1.12)
Purpose is capital
restructuring (1 if
yes, otherwise 0)
Purpose is to
continue a history
of repurchases (1 if
yes,otherwise 0)
Purpose is to
distribute cash from
unspecified source
(1 if yes,
otherwise 0)
Purpose is to
distribute cash from
normal operations
(1 yes, otherwise 0)
Purpose is to
distribute cash from
asset dispositions (1
if yes, otherwise 0)
Purpose is to get
the stock delisted
(1 if yes, else 0)
Purpose is to -0.016
facilitate a business (-0.99)
combination (1 if
yes, otherwise 0)
Purpose is tax 0.028
preferred cash (0.65)
distribution (1 if
yes, otherwise 0)
Purpose is replace -0.088 **
dividends (1 if yes, (-2.00)
else 0)

Year F-value 2.776 2.745 2.879
Adjusted [R.sup.2] 0.468 0.470 0.488

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

* Significant at the 0.10 level.

Table VII. Regression Analysis of Announcement Period Returns around
226 Issuer Tender Offers to Purchase Common Stock over the Period
1994-2006

The dependent variable, the three-day cumulative abnormal
announcement period return, is calculated using the market model
estimated over the 250-day period ending 10 days prior to the
initial tender offer announcement. Column 1 lists short descriptions
of the independent variables while columns 2-4 list regression
coefficient estimates from the three regressions. Technical
definitions for noncategorical variables are located in the
Appendix. The t-values, shown in narentheses. are calculated using
robust standard errors. We winsorize extreme observations of each
variable.

(1) (2) (3) (4)

Intercept 0.033 0.068 0.054
 (0.45) (0.95) (0.66)
Premium as a percent of 0.321 *** 0.340 *** 0.348 ***
Shares sought to (5.51) (6.53) (6.51)
total shares -0.009 -0.015 -0.008
outstanding

Shares sought to (-0.32) (-0.57) (-0.29)
total shares 0.055 ** 0.037 0.034
outstanding

Insider ownership (2.50) (1.69) (1.52)
 -0.004 -0.004 -0.005

Natural log of book (-1.21) (-1.18) (-1.19)
value of total 0.010 0.012 0.009
assets

Dutch auction (1 if (0.84) (1.02) (0.76)
Dutch auction tender -0.035 *** -0.038 *** -0.034 ***
offer, otherwise 0)

Defensive (1 if yes, (-3.06) (-3.74) (-3.14)
otherwise 0) 0.036 *** 0.038 *** 0.036 ***

Cash and equivalents (2.78) (3.14) (3.05)
to market value of 0.102 ** 0.102 ** 0.103 **
assets

Free cash flow to (2.28) (2.23) (2.23)
market value of 0.046 0.030 0.028
assets

Debt ratio (1.27) (0.83) (0.76)
 0.007 -0.062 -0.051
Dilutive to total (0.07) (-0.63) (-0.49)
common shares 0.006 0.005 0.006 *
outstanding

Market-to-book value (1.37) (1.59) (1.68)
of common equity 0.007 0.013

Purpose is to avoid (0.51) (0.94)
transactions costs/
commissions (1 if
yes, otherwise 0)

Purpose is to -0.013 -0.014
increase EPS (1 if (-0.90) (-1.05)
yes, otherwise 0)

Purpose is best 0.031 ** 0.023 *
investment (2.55) (1.96)
opportunity (1 if
yes, otherwise 0)

Purpose is to -0.008 -0.006
maximize shareholder (-0.65) (-0.43)
value (1 if yes,
otherwise 0)

Purpose is stock is 0.006 0.001
undervalued by (0.58) (0.12)
market (1 if yes,
otherwise 0)

Purpose is to 0.008 0.007
provide liquidity to (0.77) (0.75)
shareholders (I if
yes, else 0)

Purpose is capital -0.014 -0.015
restructuring (1 if (-1.45) (-1.52)
yes, else 0)

Purpose is to 0.022 ** 0.020 *
continue a history (2.03) (1.87)
of repurchases (1 if
yes, otherwise 0)

Purpose is to -0.031 ** -0.028 *
distribute cash from (-2.05) (-1.86)
unspecified source
(1 if yes, otherwise
0)

Purpose is to 0.014 0.013
distribute cash from (1.11) (0.94)
normal operations (1
if yes, otherwise 0)

Purpose is to -0.001 0.002
distribute cash from (-0.08) (0.18)
asset dispositions
(1 if yes, otherwise
0)

Purpose is to get -0.009 -0.007
the stock delisted ( (-0.43) (-0.09)
I if yes, otherwise
0)

Purpose is to -0.012 -0.007
facilitate a (-0.61) (-0.34)
business combination
(1 if yes, otherwise
0)

Purpose is tax 0.025 0.020
preferred cash (0.58) (0.48)
distribution (1 if
yes, otherwise 0)

Purpose is replace -0.100 ** -0.095 **
dividends 1 if yes, (-2.49) (-2.21)
otherwise 0)

Year F-value 2.748 2.988 2.948
[R.sup.2] 0.478 0.507 0.508

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

* Significant at the 0.10 level.
COPYRIGHT 2011 Financial Management Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dunn, Jessica Kay; Fayman, Alex; McNutt, Jamie John
Publication:Financial Management
Date:Dec 22, 2011
Words:16428
Previous Article:Buy smart, time smart: are takeovers driven by growth opportunities or mispricing?
Next Article:Corporate repurchase decisions following mutual fund sales.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters