1
Executives’ equity compensation and its relationship to acquisition and divestitures /Master thesis /
Lars Stein (355273ls) /
13.12.2012 /
Supervisor: Prof.dr.E.A. de Groot /
Table of Contents
Introduction
Literature review
Size & compensation
The link between executive compensation and firm performance
Conceptual development
Alignment theory
Differentiations and asymmetric risk
Equity compensation effects
In which way options do the job
Acquisitions
Divestitures
The context of incentives
CEO position tenure
Firm performance
Hubris
Methods
Sample and Data
Dependent variables
Independent variables
Control variables
Empirical Method
Analysis & Results
Deficiencies & Discussion
References
Appendix
Figures
Tables
Executives’ equity compensation and its relationship to acquisitions and divestitures
An exogenous study(Sanders, 2001) was pivotal to this thesis and thus was widely replicated. Particularly this underlying study analyzed how either CEO stock ownership or stock option pay affectsCEOs risk propensity. In this case, the willingness to engage into acquisition or divestitures was used as a proxy for risk. Further contextual circumstances like position tenure and firm performance were expected to have moderating characteristics. This thesis tried to provide additional assurance towards the results by carrying out the empirical study in two distinct economicalperiods and conditions i.e. high growth- versus a more volatile period, further including four different executive categories, questioning the effects of the moderating variables, and including theidea that hubris may be an additional driver. Surprisingly the results were widely insignificant and so naturally led to a troubleshooting approach. Indications are found that the initial findings of Sanders could have been somewhat created. At least, by adopting such an approach, significant results could finally be obtained in this thesis. Ultimately, the only consistent support that could be found is that stock option pay tends to be highly correlated to firms’ acquisition activity.
Introduction
The research stream pivoting around the link between executive compensation and its subsequent effects on firm performance is extensive and inconclusive. Basically this area of research can be divided into two main categories. First, research examines the links between firm size, performance, mergers and executive compensation, and second, research tried to shed light on equity based-compensation and its relationship to firm risk(Williams, et al., 2008). Combiningboth of these categories Sanders (2001) found that within CEO compensation packages, stock option pay and stock ownership have diametrically opposite effects on companies’ acquisition and divestiture activity. Particularly it is posited that CEO stock ownership, while creating a bond between CEO and shareholder wealth and so provoking downward as well as upward CEO wealth potential, generally leads to a risk-averse attitude facing possible acquisitions and divestitures. Thus the downside threat is deemed to dominate the risk perception. On the other hand, CEO stock options are found to have a risk-seeking propensity since stock option pay is considered to be a mere option on potential gains. Yet the backbone scientific economic literature often did not incorporate these opposing characteristics of executive equity pay. More over related scientific economic literature examining different risk incentives triggered by equity compensation as well as scientific economic literature scrutinizingthe underlying principal-agent alignment is generally discordant (as will be demonstrated later on). The findings of Sanders are in this sense ground-breakingand may be pointing towards unequivocal evidence concerning possible different effects of equity compensation. However the rather short time-frame employed by Sanders (1991-1995) may not provide enough assurance in order to maketight conclusions. Further his examination period was situated during times of moderate but steady economic growth and was investigated for CEOs only. It so seems important to remove ambiguity and to provide additional guidance and guaranteein which way executives’ common stock compensation and stock option pay affect firm performance.To strengthen the conclusions this thesis draws heavily upon Sanders work and so widely replicatesitsmethodology using a lengthier period of analysis, distinct economic conditions and incorporates the equity compensation schemes of the top three executives.
Obviously it could be anticipated that the here pursued outcomes will be very similar to those found by Sanders. The methodological proximity between this thesis and the paper of Sanders should allow a credible comparison of the results. Hence the findings have again been obtained by means of a negative binomial regression. Surprisingly the first output of results was nowhere close to what could have been expected. Such discrepancies seemed astonishing.Consequently a trouble-shooting approach was adopted in order to grope about possible explanations. Ultimately, the most important finding of this thesis is the suspicion that Sanders may have assembled his results using an econometrical ruse, which is in fact not sound. This thesis so contributes to management accounting research by demonstrating that even studies of prime academic journals, with their respective conclusions, must not be taken as being universally true or being usually assembled on high qualitative grounds. Accordingly, the premise that “if something is supported by data it must be true” should be encountered with more skepticism.
The remaining paper is structured as follows: The Literature Review provides a small digest of the vast background information and coherences that surround the topic of executive compensation. The Conceptual Development deepens the analysis of equity compensation in the context of the alignment theory, distinct risk propensities, acquisition and divestitures, different contextual circumstances and so simultaneously develops the hypotheses that need to be investigated. The Methods part explains the composition of the data-set and presents further details on the chosen variables and methods. The Analysis and Results part provides statistical and graphical interpretations about individual variables, reads the several regression results and eventually takes conclusions. Finally, in the Deficiencies and Discussion part, shortcomings of the employed method are raised, discussed and suggestions for future research are presented.
Literature review
Starting with the literature focusing on executive compensation and its links to firm size, performance and merger activity, a brief summary highlights the divided findings and the unclear conclusion that can be drawn. Note that the following deliberations consider executive equity compensation as well as cash compensation or both. The storyline keeps track to a chronological order connected to the empirical data sets used in the relevant studies. This way a temporal trend may become apparent.
Size & compensation
The basic idea relating size and compensation is that increases in firm size, assumed tolead to a wider scope of operations, enables the top-management to restructure its compensation level with the board. Thus, increases in firm size should be related to inflating executive remuneration. This wayJensen (1986) already assesses that managerial compensation is positively related to size and hencemanagers have incentives to grow their firms beyond optimal size; hinting already at possible agency problems which will be seized later on. Schmidt and Fowler (1990) find this pattern for acquiring or bidding companies. Here, on average, post-acquisition periods reveal poor financial performance.However top management gets it right to experience a significant increase in cash compensation (salary and bonus).Later Hambrick and Finkelstein (1995) restore the findings claiming that only within management-controlled firms the same links between size and compensation (here: cash pay and stock options) can be found. Conversely, within externally-controlled firms’, they assess thatchanges in CEO pay are stronger related to economic returns.More recently,Grinstein and Hribar (2004) found that 39% of acquiring firms cite M&A deals as a reason to increase CEO cash bonuses. They state that deal size plays a larger part in bonus compensationlevels compared to effort and skill, and is independent with deal performance. Finally, managerial power is likely to explain a large part in bonus variations, emphasizing the cliché about self-serving CEO’s.Bliss & Rosen (2001) as well find that size adds to CEO compensation (here: cash payments, restricted shares and stock options). They find that, within bidding banks, acquisitions-related drops in stock prices, resulting in manager’s wealth losses, will be offset by the effects of size. Likewise and most interestingly Harford and Li (2007) find that executives of bidding companies are exuberantly compensated for growth through acquisitions with new stock and option grants. This leads to an insensitivity of CEO wealth towards poor post-acquisition stock performance and thus effectively limits the downside risk of acquiring CEOs. The made evidence is important as they exhibit in which way initially crafted compensation schemes in favor of a principal-agent alignment (use of stock and options) are prone to be neutralized within their actual range of action. However, these findings are found to be mitigated by the presence of a strong board.Finally,Williams, Michael & Waller (2008) analyze a wide range of corresponding literature andconclude that size appears to be positively related to managerial compensation levels. Further, they infer that acquiring managers are able to increase their pay with merger activity even so causing negative shareholder returns.
The link between executive compensation and firm performance
Inconsistent with principal-agent alignment ideals,that is inducing the agent (e.g. the executive) to work in line with the interests of the hiring principal (e.g. majority shareholder) i.e. creating a bond between executive compensation and firm performance, Kerr & Bettis (1987) find no relation between shareholder returns and executive salary and bonus variations. Baker, Jensen & Murphy (1988) come to a similar conclusion, particularly they state that “the absence of incentives is pervasive, and it’s not surprising that large organizations typically evolve into bureaucracies” (1988: 614). They reason that equity incentives are much more effective tools (among others) in order to align managerial stimulus with shareholders interests. Interestingly they found that low growth or declining industries tend to shift towards enhanced performance-pay structures. In this light Mehran (1995) found that firm performance (Tobins Q & ROA) is indeed positively related to the percentage of executive compensation that is equity based (here: salary & bonus, options, restricted stock, shares etc.). The same link is found for build up managerial equity holdings. In turn, Lambert et al. (1993) relatives the prior research and extends the understanding of organizational incentives design by finding explanatory evidence between tournament, managerial power and agency theories.NextJensen and Murphy (1990a) detect significant positive links between CEO pay-related wealth changes (here: salary & bonus, stock options, common shares) and shareholder wealth. However, they must conclude that this relationship is too weak in order to be consistent with formal agency models. Note that they differentiated between the effects of stock ownership (largest effect) and stock options. Similar Jensen and Murphy (1990b) find that “corporate America pays its most important leaders like bureaucrats” (1990b: 1). They find that on average CEO wealth changes by $2.59 for every $1000 change in shareholder wealth. The largest effect arisesagain from stock ownership. They conclude that changes in executive compensation are not associated with corporate performance and advocate the use of more equity-based compensation to increase the company value.
Consequently, it seems natural that equity-based compensation may be the right tool to create a link between managerial compensation and the firm performance. However,WestphalZajac (1994) find that the adoption of Long Term Incentive Plans (LTIP), consisting of stock options, stock appreciation rights (SARs), restricted stock and performance plans, may only be symbolic and that the actual usage is rather limited especially with a powerful CEO and poor prior performance. ZajacWestphal (1995) expressly underline these findings by indentifying different explanatory factors justifying the adoption of LTIPs which can be predicted on structural and political grounds. Particularlythey show that CEO compensation mirror both “substance and symbolism” (1995: 283). A general framework of understanding executive compensation has been developed by Barkema & Gomez-Mejia (1998). They summarize that research has found weak pay to performance sensitivities and reason that further research focusing in this area may stay inconclusive as long as it does not integrate additional variables e.g. peer compensation, firm governance structure, tax systems etc. Most recently, Hall and Liebmann (1998) were able to spot a strong relation between firm performance (stock return & accounting data) and the CEO compensation (here: salary & bonus, stock & options). Moreover they find that the bulk of this link is driven by CEO stock and stock options.
Finally,Williams, Michael & Waller (2008) summarize the scientific economic literature and infer that firm performance seems to be positively related to managerial compensation. Again, they argue that equity-based compensation, especially stock options, seems appropriate to align the interest of managers with those of shareholders.
Note that the before summarized literaturevaries in its research framework. It is ambiguous what kind of performance measure seems right i.e. accounting-performance indicators or share-price returns. Both of them are somewhat noisy and inappropriate to capture the subsequent performances of executives. Nevertheless, equity-based compensation seems to be the main driver behind the pay-to-performance relationship. Yetthe dispute whether managers are indeed aligned to shareholder interests or not, is not settled. Therefore the next passage tries to shed some light on this topic while, at the same time, leading to the conceptual development of this paper.
Conceptual development
Alignment theory
The previous concerns whether executives are doing a good job or not i.e. whether they are paid according to their delivered performance or rather for unwarranted reasons, fostersthe classical principal-agent problem. In other words the issue if agents are effectively working on behalf of the principals interests is still prevailing.The often-cited agency theory haggles over the interest discrepancies between executives and shareholders (AgrawalMandelker, 1987;Mehran, 1995; Jensen & Murphy, 1990a; Jensen & Murphy, 1990b; Yermack, 1995; Guay, 1999; Hall & Liebmann, 1998; Wiseman & Gomez-Mejia, 1998; Williams & Rao, 2006; Bryan, Hwang & Lilien, 2000; Sanders & Hambrick, 2007; Hanlon, RajgopalShevlin, 2003; Low, 2009). More precisely, managers are generally characterized as being risk-averse and over-invested with their respective managerial wealth (human capital e.g. reputation / financial capital) into a single company.Consequently, increases in firm risk are posited to reduce executives’ expected utility.This is conceptualized as ofsubsequently leading managers to choose variance-decreasing projectse.g. diversifying acquisitions, reducing firm risk, or at least to decline variance-increasing opportunities in order to protect their own under-diversified wealth portfolio. While trying to bear a minimum of personal risk, executives are also sometimes characterized as preferring fixed income over equity-based compensation (Mehran, 1995).Generally the described and assigned risk-avoidance tendencies come along with a more conservative approach when facing uncertain decision outcomes e.g. acquisitions/divestitures.
As opposed to this, shareholders generally have diametrically different interests.Shareholders are considered to be risk-neutral since they should be able to diversify themselves on the market. Thus, shareholders are outlined as toprefer a manager-type who is supposed to select every Net Present Value project regardless of risk. Put another way, principals want their agents to maximize firm returns and so to grab every beneficial opportunity unrestrained of variance-increasing concerns. In order to align the managerial interests with those of the shareholders, equity-based compensation is again sketched as being the right tool to overcome the previous discrepancies. This way managerial compensation is tied to firm as well as to stock performance so as to motivate executives to select variance-increasing and value-maximizing projects. The logic herereasons from the simple cohesionthat enhanced firm-relatedvariancesraise the value of underlying managerial equity such that interests areeffectively aligned with those of shareholders.Finally, a substantial fraction of common stock and stock options within in managerial compensation schemes is supposed to alter executives’ investment and finance decisions so as to be within shareholders appetite.
The simple reasoning seems appealing; however the initial issue of managerial under-diversification is not dispersed at all. Moreover equity compensation and so built-up wealth is still threatened by losses. The desired effects balance on a fine line between penalties for losses on the one hand and rewards for gains on the other hand. In brief, it is unclear in which way the initial alleged risk-aversion has been overcome. To solve the latter Guay (1999) and Low (2009) depict a deeper picture and advocate that a sound compensation policy should manage both the slope and the convexity between the relation of managerial wealth and the stock price. To do this it is necessary to discern within equity compensation.
Differentiations and asymmetric risk
Previously,executives’ common stock holdings and option grants have been regarded as being somewhat substitutes.However, the individual risk- and resulting incentive characteristics of both tools remain foggy. Hence it may be important to differentiate between them. Within the relationship between firm performance/shareholder wealth and managerial wealth,common stock, or even equity compensation as a whole, is considered to contribute to a slope-like link between both interest groups. This slope relation, particular due to managerial stock holdings, induces managers and shareholders to share both gains and losses and thus discourages managers from taking harmful actions.Due to the option-like structure of stock options,permitting the collection of gains in favorable circumstances and alternatively allowing to phase out the option during tough times, stock options add to the convexity. Thus optionsarecharacterized as to encourage managerial risk-taking (Guay, 1999; Low, 2009).Surprisingly and contrary to the latter, Hall (1998) even attributes downside sensitivity to stock options. The ideal relative composition of both instruments within executives’ compensation packages, however, remains unclear (Guay, 1999). Ultimately,Sanders (2001) especially addressed this issue of unclear equity incentives and found that, due to downside risk, executive stock ownership leads to risk aversion,whereas stock option pay, due to mere gain potentials, leads to risk seeking behavior.