Martin et al AMP 2016

r Academy of Management Perspectives2016, Vol. 30, No. 4, 349–368.http://dx.doi.org/10.5465/amp.2012.0134S

A R T I C L E S

BRIDGING FINANCE AND BEHAVIORAL SCHOLARSHIP ONAGENT RISK SHARING AND RISK TAKING

GEOFFREY P. MARTINUniversity of Melbourne, Melbourne Business School

ROBERT M. WISEMANMichigan State University

LUIS R. GOMEZ-MEJIAArizona State University

A large volume of research has examined agent risk taking and the contracting problem ofrisk sharing—the sharing of performance risk across agent and principal—to advance ourknowledge of mechanisms that can align the assumed divergent interests and risk pref-erences of the managerial agent and shareholder principal. This research has been un-dertaken in two streams—financial economics and behavioral science—that appear tohave operated in silos, using different theoretical frameworks and methodological ap-proaches. To identify opportunities for cross-fertilization and to advance future researchin both streams,we review the theoretical paradigms and empirical findings deriving fromboth fields. We also make an assessment of how the combined research efforts of financeand behavioral scholars have progressed in developing our understanding of agent risktaking and mechanisms for achieving agent–principal incentive alignment. Finally, wediscuss how this research has influenced the corporate world for better or for worse.

For nearly four decades scholars have been inves-tigating the agency problem and equity-based pay asa means of aligning the incentives of a firm’s princi-pals and agents; this paper examines the progressmade in this area and maps a future research agenda.To do so, we examine theoretical development andselected empirical research emanating from two dif-ferent theoretical foundations thathavedrivenagencyresearch in the context of agent risk taking: financialeconomics and behavioral science. These perspec-tives have operated largely in research silos sepa-rated by different theoretical assumptions andterminology. The lack of interaction between thesestreams is puzzling because both areas of researchare effectively concerned with the same question:how to design a contractual relationship betweenthe principal and agent that reduces agency costs tothe principal.

Financial economistsoften refer to this arrangementin terms of risk sharing (e.g., Ai & Li, 2015; Gao, 2010;Holmstrom, 1979; Shavell, 1979), while behavioral

scientists focus on incentive alignment (Gomez-Mejia,Welbourne,&Wiseman, 2000;Kolev,Wiseman,&Gomez-Mejia, in press; Nyberg, Fulmer, Gerhart, &Carpenter 2010; Cuevas‐Rodrıguez, Gomez-Mejia, &Wiseman, 2012; Wiseman & Gomez-Mejia, 1998),though both risk sharing and incentive alignment re-flect the same phenomenon: making a portion of theagent’s compensation contingent on achieving out-comes important to the principal. For both groups ofscholars, the design of compensation is a criticalmechanism for reducing agency costs that arise whenagents shirk their obligation to enhance shareholderwealth.A typeof agent cost of great interest to agencyscholars exploring risk sharing and incentive align-ment involves the agent minimizing his or herconcentrated firm-specific risk—the agent is as-sumed to be less diversified than the principal—atthe expense of positive net present value (NPV) in-vestments. We argue that bridging these streams canenhance our understanding of how risk sharingthrough incentive alignment may influence agent

349

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download, or email articles for individual use only.

http://dx.doi.org/10.5465/amp.2012.0134S

risk taking and thus the returns—or residualclaims—accruing to the principal.

Financial economists have argued that agencycostsarise because agents are rationally opportunistic inpursuing self-interests that do not necessarily alignwith the interests of principals (e.g., Holmstrom,1979; Jensen & Meckling, 1976; Shavell, 1979). Thedifficulty and cost of monitoring agent behavior haveled positivist agency scholars in the finance field tofocus on incentive structures that reward agents forachieving goals important to the principal (Jensen &Murphy, 1990). In other words, due to the limitationsof directly monitoring agent effort, finance researchhas sought to design efficient incentive-alignmentsystems that allow principals to transfer a portion oftheir risk onto the agent by tying a portion of theagent’s compensation to the achievement of perfor-mance outcomes important to the principal (Jensen& Meckling, 1976). This transferring of performancerisk to the managerial agent is referred to as risksharing, given that the agent shares the principal’srisks associated with share price performance.

While positivist agency scholars have argued thatthe transfer of risk from the firm’s principal to themanagerial agent—through incentive alignment—may encourage wealth maximizing efforts by agents,normative agency scholars point out that this also canaggravate agent risk bearing (Holmstrom, 1979). Thatis, risk bearing created by the agent’s concentratedinvestment of human capital in his or her employeris aggravated if the agent must also share the perfor-mance risks of the firm. The agent’s overinvestmentproblem could encourage an attitude toward risk thatresults in suboptimal investments from the perspec-tive of the principal (Eisenhardt, 1989).

The contracting problem of risk sharing betweenprincipals andagentshas given rise to a large literaturewithin the finance field. This research has sought theoptimal contract for balancing the costs and benefitsof risk sharing between agents and principals(e.g., Bettis, Bizjak, & Lemmon, 2001; Gao, 2010). Inparticular, financial economists have focused con-siderable attention on equity forms of pay (suchas stock options) and how different characteristicsof equity-based pay may influence agent decisionsunder assumptions of rational self-interest.

Attracted to its essential prediction that incentivesmay reduce agency costs, organizational and behav-ioral scholars adopted positive agency theory as afoundation for a corresponding stream of researchexamining the role of executive compensation ininfluencing firmperformanceandexecutivebehaviors(e.g., Kish-Gephart & Campbell, 2015; Gomez-Mejia,

Berrone, & Franco-Santos, 2010a; Martin, Gomez-Mejia, & Wiseman, 2013). Inspired by Eisenhardt’s(1989) elegant explication of agency theory for thebenefit of management scholars, behavioral researchjudiciously attempted to enrich the managementfield’s understanding of how incentive alignmentproperties of equity-based pay may control agencycosts such as shirking and perquisite consumption.In adopting the positive agency view of Jensen andMeckling (1976), early applications of the theorywithin organizational sciences retained the assump-tions of rational self-interest that ignored the morenuanced view of human nature emerging from be-havioral sciences.

Using the rich empirical findings and theoreti-cal development of behavioral decision research,a behavioral approach to understanding the agencyproblem—in particular the problem of risk sharing—was subsequently developed in an effort to reconcilethe disparate views of positive and normative agencyscholars regarding the merits of risk sharing asa possible solution to the agency problem (Wiseman& Gomez-Mejia, 1998). The behavioral view usedthe guiding principles of prospect theory research(e.g., Kahneman & Tversky, 1979) along with otherseminal research emanating from behavioral scienceand psychology (e.g., Cyert & March 1963/1992;Lopes, 1984, 1987).

In particular, prospect theory suggested that (1)individuals are inherently loss averse rather thanrisk averse; (2) the value assigned by individuals towealth is referent dependent—for example, thenegative utility of financial losses outweighs thepositive utility attached to gains of the same dollarvalue; (3) risk preferences are context dependent,allowing for risk-seeking behaviorwhenanticipatinga loss; and (4) probability is nonlinear such thatprobabilities near certainty are givenmore weight indecisions than less certain outcomes. Individualsare risk averse if they prefer a certain outcome to agamble with the same expected value. Loss aversionsuggests that individuals weigh losses more heavilythan gains, meaning they will prefer the aforemen-tioned gamble over a certain outcome if the expectedvalue of both alternatives equates to a loss. Individ-uals are described as risk seeking when a gamble ispreferred over a certain outcome with the same posi-tive expected value.

Though retaining a weak form of the assumption ofself-interest, the behavioral approach discarded theassumption—long held by financial economists—ofrationality,which impliedunwavering and consistentpreference orderings; consistent preference orderings

350 NovemberAcademy of Management Perspectives

simply means that individuals always prefer morewealth to less and less risk over more (Jensen, 1994).This assumption provides the foundation on whichfinancial economistsmodel agents as consistently riskaverse and principals as risk neutral. The behavioralapproach replaces the assumption of consistent riskpreference orderings with empirically validated pat-terns of behavior.

As a result, behavioral agency scholars have come todifferent conclusions than financial economists aboutthe role and influence of incentive-alignment mecha-nisms on agent preferences regarding risk (e.g., Devers,McNamara,Wiseman,&Arrfelt,2008;Larraza-Kintana,Wiseman, Gomez-Mejia, & Welbourne, 2007; Martinet al., 2013; Sanders, 2001; Sanders &Hambrick, 2007).The behavioral approach challenged the finan-cial economists’ view that all forms of wealth (orforms of compensation) are valued equally by theagent and thus should have similar influences onbehavior (Devers et al., 2008; Martin et al., 2013).Recognizing these differences opened new lines ofinvestigation to behavioral agency research that areclosed to financial economists’ theorizing as a re-sult of the use of restrictive assumptions. For in-stance, relaxing the assumption that managerialagents will always prefer less risk to more (that is,are consistently risk averse) is likely to sub-stantially alter financial economists’ predictions ofagent risk behavior and their modeling of the opti-mal principal–agent contract and provide greaterinsight into the efficient use of stock and options inequity-based pay.

Given the different theoretical foundations andconclusions about individual preferences regardingrisk, perhaps it should not be surprising that therehas been limited cross-fertilization of ideas betweenfinancial economists and behavioral scholars whenconsidering agent responses to risk-sharing arrange-ments, especially as it applies to agent risk-takingbehavior. The omission of behavioral research fromstudies of agency by financial economists is curiousgiven the growing influence of the behavioral per-spective in other areas of finance, such as investorbehavior (e.g., Barberos & Thaler, 2003). Despite thedifferences between the two streams, we believe a re-view of the respective literatures will yield positivecontributions to our understanding of agent behaviorunder conditions of risk sharing.

Thus, we review and compare the theory and se-lected empirical findings of finance and behavioralviews of agency theory with a focus on the study ofrisk sharing, principal–agent incentive alignment,and the role of equity-based pay in incentive

alignment. Thoughwehope to capture the essence ofthe two primary research streams regarding risksharing between principals and agents, we focus onwhat we believe to be foundational contributions tothe understanding of risk sharing and its role in risktaking. Admittedly, this delimiting of our review issomewhat subjective, and every scholar is likely tohave a personal list of important contributions.Therefore, our task is complicated by the normalscholarly debates over merit, as well as by the volu-minous amount of research and theorizing that hasoccurred on these topics. Recognizing this limitation,we offer our brief review as a platform for outlininga future research agenda that integrates key elementsof each perspective in an attempt to focus and en-hance agency research. Finally, we develop a pro-gress report for this research and examine how ithas affected corporate practice in the past or has thepotential to affect it in the future.

RISK SHARING AND THE AGENCY PROBLEM

Most of the research and theorizing about theagency problem over the past 40 years can be tracedto Jensen and Meckling’s (1976) seminal article de-veloping a formal model of agency theory. Theirwork was prompted by the inability of financialeconomists to explain why firms failed to maximizeshareholder wealth. Agency theory was developedto explain this failure of neoclassical economic the-ory of firms by suggesting that self-interest on thepart of management may result in shirking, per-quisite consumption, and other behaviors thatfulfill the manager’s interests at the expense ofshareholders. Given the difficulty and costs of di-rectly monitoring agents, agency scholars advo-cated the use of incentive alignment as a means forcontrolling these agency costs (Eisenhardt, 1989).As noted previously, this resulted in risk sharing bythe principal and agent such that the agent boresome risk to personal wealth in pursuit of perfor-mance outcomes desired by the shareholder prin-cipal (Holmstrom, 1979; Shavell, 1979). This risksharing was thought to provide agents with an in-centive to maximize shareholder wealth by linkinga portion of agent compensation to increases inshareholder wealth.

However, sharing of risk exacerbates the exposureto harm (or risk bearing) of agents who are unable todiversify their human capital investment in the firm(Fama, 1980; Holmstrom, 1979; Shavell, 1979). Rec-ognizing the competing influences of risk sharing onagent behavior, financial economists devoted much

2016 351Martin, Wiseman, and Gomez-Mejia

of their attention to devising compensation con-tracts that attempt to balance the benefits of con-tingent compensation’s incentives with the costsdemanded by agents in return for the additional riskbearing it creates (e.g., Bettis, Bizjak, & Lemmon,2001; Gao, 2010). We next review the theoreticaldevelopments and empirical findings of both thefinance and behavioral scholars in examining risksharing and incentive alignment. We summarizethe theoretical assumptions of both streams inFigure 1.

The Economic Perspective on Risk Sharing

Theory. Financial economists’ theoretical ap-proach to agency research has been based on theassumptions that human beings are rational utilitymaximizers and that the risk preferences of princi-pals (risk neutral) and agents (risk averse) are deter-mined by their relative abilities to diversify personalrisk (e.g., Holmstrom, 1979; Jensen & Meckling,1976; Shavell, 1979). Shareholders (the principal)can diversify their financial investment in the firm,leading to the conclusion that they are generallyrisk neutral in their preferences regarding eachinvestment;managers (the agents) cannot diversifytheir human capital investment, leading to theconclusion they are risk averse (Holmstrom, 1979;Shavell, 1979).

The assumption of a risk differential betweenagents and principals has led finance researchers toexamine the risk incentives associated with stockoptions and restricted stock to examine how they

maybe used to incentivize the agent to takemore risk(e.g., Coffee, 1988; Jensen & Meckling, 1976). Firmsutilize a mix of options and restricted stock in anattempt to create optimal agent risk incentives andthus the optimal contract (Core & Guay, 1999). Theoptimal contract weighs the costs of compensatingthe agent for bearing that additional risk against thebenefits assumed from incentivizing the agent totake more risk (Holmstrom, 1979). According to thisline of research, the nature of the optimal contractwill depend on capital structure (Farmer & Winter,1986; Jensen & Meckling, 1976; John & John, 1993),firm size (Core & Guay, 1999; Smith & Watts, 1992),monitoring difficulty (Core & Guay, 1999; Demsetz &Lehn, 1985), the agent’s attitude toward risk (Hall& Murphy, 2002; Lambert, Larcker, & Verrecchia,1991; Lewellen, 2006; Ross, 2004; Shavell, 1979),and the agent’s ability to reduce (or hedge) his or herexposure to adverse changes in firm-specific wealth(Gao, 2010).

Stock option grants—a formof executive (or agent)compensation and equity-based pay—have been thefocus of a large body of finance research examininghowequity-basedpayaffects agent risk taking. Indeed,finance scholars have developed a strong empiricalliterature examining how rational utility-maximizingagents respond to the dynamic nature of stock options(see review below). Beginning with the assumptionthat stock options create no downside and unlimitedupside potential for agents, financial economists havelooked to stock options as a way to avoid aggravat-ing the agent’s risk burden while also incentivizingshareholder wealth maximization (Jensen & Murphy,

FIGURE 1Contrasting and Comparing—Different Theoretical Approaches to the Study of Agent Risk Taking

Financial economics Behavioral agency

Theory Positive agency theory Normative agency theory

Positive agency theory Prospect theory Behavioral theory of the firm Stakeholder theory Upper echelon theory

Theoretical assumptions

Managers as agent Shareholders as principal Individuals are rationally self-

interested Self-interest results in goal conflict Information asymmetry between

agents and principals Preeminence of efficiency

Multiple agents possible Multiple shareholders possible Enlightened self-interest possible Goal conflict possible, not inevitable Individuals are loss averse,

boundedly rational, and subject to a variety of decision heuristics

Information asymmetry between agents and principals

Preeminence of efficiency


1990). They justify this assumption by arguing thatbecause agents have no upfront investment in theoptions and no obligation to exercise them, they ex-perience no loss of wealth should the options proveworthless. Thus, stock options have been viewed ashaving great potential for providing an efficient solu-tion to balancing incentive alignment with the prob-lemof risksharing (Flor,Frimor,&Munk,2014; Jensen& Murphy, 1990; Smith & Stulz, 1985). That is, theasymmetry of option payoffs, creating larger propor-tionate gains for the agent than the principal if thestock price increases, could be justified based on thefinancial economists’ assumption that agents must becompensated for the increased risk they bear fromhaving a portion of their pay dependent on fluctua-tions in firm performance (Holmstrom, 1979). Com-bining options with stock provides an efficientsolution to avoiding agent risk aversion,more efficientthan using stock alone as the equity incentive (Floret al., 2014).

In sum, the literature from financial economicslargely views agent characteristics as a constant1 andthus devotes most of the effort toward developinganalytic predictions about the effects of awardingequity—stock and options—to the agent with anobjective of creating optimal agent risk incentives (orthe optimal principal–agent contract). Aswewill seein the next section, empirical findings have not beenentirely supportive of these predictions.

Empirical findings. Empirical research by finan-cial economists has examined the effect of equityownership on agent risk incentives and the design ofthe optimal agent–principal contract. For example,this research has explored the relationship betweenagent risk taking and the dynamic characteristics ofstock options agents hold as a result of past equitygrants. Convexity of stock options (i.e., gamma) heldby the agent reflects the rate of (exponential) increasein his or her option wealth as a result of increases inthe firm’s stock price. This option characteristic hasbeen shown to positively influence executive risktaking, which has been explained on the basis thatgreater convexity increases the monetary incentivesfor the agent to pursue growth opportunities throughinvestments that are associatedwith greater earningsvolatility (Chava&Purnanandam, 2010; Cohen et al.,2000; Cohen, Dey, & Lys, 2013; Guay, 1999; Rajgopal& Shevlin, 2002). That is, empirical findings supportthehypothesis that thepositive relationshipbetween

stock volatility and agent option wealth leads agentsto make higher risk decisions given that these de-cisions increase stock volatility and therefore theiroption wealth. Thus, empirical support has beenprovided for the idea that stock options encouragerisk taking.

Despite these findings, empirical finance researchhaschallengedfoundationalagencytheory—suggestingthatequityownership leads togreater agent risk takingand that equity is awarded by boards of directors withthe objective of purposefully creating optimal agentrisk incentives—on four issues. First, the aforemen-tionedconvexity fromoptionpaydoesnotnecessarilyalways succeed in increasing agent risk taking andtherefore does not unequivocally reduce the assumedrisk aversion of managerial agents (Carpenter, 2000;Coles et al., 2006; Hall & Murphy, 2002; Lambert,Larcker, & Verrecchia, 1991; Lewellen, 2006; Ross,2004). Second, the empirical task of examining therelationship between equity-based pay and agent risktaking is confounded by the risk premium required tohold stock options (Coles et al., 2006; Hall &Murphy,2002;Lambert,Larcker,&Verrecchia, 1991;Lewellen,2006). That is, as executives take more risk—possiblyto enhance the value of their stockoptionwealth—thehigher risk premium associated with stock and op-tionsmakes additional risk taking less attractive to theCEO. The difficulty of controlling for this risk pre-mium in studies of the relationship between stockoptions and risk behavior has thwarted this stream ofempirical research (Hayes, Lemmon, & Qiu, 2012).

Third, there is possible endogeneity of the riskincentives attached to stock optionswhenpredictingagent risk taking, given a causal effect of varianceincreasing investments or exogenous events thatincrease firm risk (or risk taking) on the use (or in-clusion) of executive stock options in compensationcontracts (Dai, Jin, & Zhang, 2014; Rajgopal &Shevlin, 2002). Fourth, a significant decrease in theawarding of option-based compensation after theintroduction of new accounting rules (FAS 123R)suggests that stock options are merely perceived asanother form of compensation (Hayes et al., 2012),challenging the idea that firms purposely awardstock options with the objective of incentivizingrisk taking, as suggested by earlier agency research(cf. Core & Guay, 1999).

Acorporate decisionproviding further insight intoagent risk taking and opportunistic agent risk man-agement that has also been explored by financescholars is corporate diversification. Similar to theeffect of stock option awards to the CEO, researchexamining corporate diversification has featured

1 Though the field of finance has recognized the impor-tance of individual differences, as reflected in the conceptof managerial style (Bertrand & Schoar, 2003).


conflicting theorizing and empirical findings. Somescholars have argued that diversification allows themanagerial agent to reduce the firm’s overall riskprofile and therefore the firm-specific risk of themanagerial agent (Amihud & Lev, 1981; May 1995).Thus, diversification has proved a useful decision toanalyze, given that it allows scholars to further ex-amine the relationship between firm-specific riskand actions that can reduce it.

However, the view that greater firm-specific riskleads to more diversification has been challenged byempirical research demonstrating that the primarymotive for undertaking corporate diversificationwasless powerfully driven by the agent’s desire to reduceidiosyncratic firm risk and more strongly driven bythe private benefits the managerial agent extractsfrom managing a diversified firm, such as greaterability to skim (throughperquisite consumption) andenhanced career prospects (Aggarwal & Samwick,2002; Denis, Denis, & Sarin, 1997). That is, it appearsfrom this latter research stream that the motive tofurther enrich oneself overrides the risk reductionmotive for managerial agents.

Alongwith examining the effect of stock options onagent risk taking in the form of strategic investments(or propensity for leverage), the finance literature hasadeptly identified other forms of opportunistic agentriskbehavior that imposecosts onshareholders.Theseinclude the misreporting of financial performanceor smoothing of income to minimize the risk of lossof equity wealth (Armstrong et al., 2013; Bergstresser& Philippon, 2006; Burns & Kedia, 2006; Grant,Markarian, & Parbonetti, 2009; Kim, Li & Li, 2015).These forms of opportunistic agent risk behavior—oropportunistic risk reduction—through the manip-ulation of the stock price create additional agencycosts; these costs arise given that smoothing ofearnings or earningsmanagement (ormanipulation)deceives shareholders and other firm stakeholdersregarding the agents’ performance and the value oftheir investment.2

Similarly, the sensitivity of option value to changesin the stock’s volatility (i.e., vega) has been found topositively influence firm leverage (a common riskproxy) and stock return volatility (Cohen et al., 2000;Guay, 1999). Extending the study of the effects of CEO

incentives to the employee (as opposed to the share-holder, as the stakeholder who bears the cost of op-portunisticCEObehavior),CEOshavealsobeen foundto increasingly underfund pensions as option deltaand vega increase (Anantharaman & Lee, 2014). Inexploring and detailing these forms of opportunisticagent behavior, the finance literature has successfullyoutlined limitations to the effectiveness of equity-based pay as a mechanism for creating optimal riskincentives and thus reducing agency costs.

In sum, the above brief reviewof empirical researchemanating from financial economics provides somesupport for the initial and seminal theorizing regard-ing thepositive risk incentivesassociatedwithequity-based pay. However, it has also provided evidence tothe contrary and underlined empirical shortcomings.

A Behavioral View of Risk Sharing

Organizational and behavioral scholars becameattracted to agency theory as away to understand therole of senior executives in firm performance.Agency theory’s accessibility to organizational sci-ence can be credited in large part to the work ofEisenhardt (1989, p. 57), who concluded that agencytheory “(a) offers unique insights into informationsystems, outcome uncertainty, incentives, and riskand (b) is an empirically valid perspective, particu-larly when coupled with complementary perspec-tives.” Eisenhardt noted that the theory elucidatedissues within the firm that are relevant to organiza-tional and behavioral scholars—such as the use ofinformation systems, incentives, andmanagerial riskpreferences—and would benefit from being com-bined with other theoretical perspectives.

Heradmonition to reconsideragency theory throughthe lens of organizational scholarship led to a growingagency literature rooted in sociology and psychol-ogy. This research produced an alternative per-spective on the causes and consequences of agencycosts, as well as the mechanisms for controllingagency costs (e.g., Gomez-Mejia & Balkin, 1992).While some responses to the financial economists’view of agency theory merely attacked it as beingoverlycriticalofhumannature (e.g.,Davis,Schoorman,& Donaldson, 1997; Perrow, 1986), others sought tosupplement agency theory’s contribution by intro-ducing alternative perspectives that replaced someof its restrictive assumptions to broaden its appeal(Gomez-Mejia,Welbourne,&Wiseman,2000;Wiseman& Gomez-Mejia, 1998). Collectively referred to as be-havioral agency theory, this latter research stream hasattempted to bridge agency theory (as formulated by

2 Note that this literaturehas alsooutlined caveats for theefficiency of stock options. Hall and Murphy (2002) sug-gested that options align interests more effectively if theyare valuable (exercise price below market price) whengranted, due to the marginal losses the CEO will incur ifstock prices decline in this situation.


financial economists) with behavioral research exam-iningdecisionmakingunderuncertainty.However, thisbehavioral approach to agency is a nascent field of re-search, with a limited number of theoretical and em-pirical studies explicitly (that is, studies combiningbehavioralandagency theories)usingbehavioral theoryto examine agent risk taking and the agency problem ofrisk sharing. We now review theoretical and empiricalresearch in this field from psychology and behavioralresearch that provided the platform.

Theory. The behavioral approach to agency theoryhasdrawnonawide rangeof research frombehavioralscience, sociology, and psychology examining choicebehavior and individual risk preferences (Camerer,1989; Cohen, Jaffray, & Said, 1987; Kahneman &Tversky, 1979, 1984; Lopes, 1984; Tversky, Sattath, &Slovic, 1988; Weymark, 1981). The idea that an in-dividual’s choices will depart from the rationalpredictions of expected utility theory (Bernoulli,1738/1954; von Neumann & Morgenstern, 1947) hasa long tradition in behavioral research (Simon, 1955).More recent work is collectively known as prospecttheory anddrawsmostly on experimental researchonindividual choice behavior under uncertainty.

Recent advancements in this area have demon-strated that an individual’s risk preferences are con-text dependent and that individual choices underuncertainty systematically depart frompredictions ofrational choicemodels (Kahneman, 2012; Kahneman& Tversky, 1979; Luce & Fishburn, 1991). In particu-lar, behavioral research on choice under uncertaintyindicates that individuals are profoundly affectedin their decision making by cognitive limitationsand often rely on decision heuristics that lead tochoices that appear less rational or less consistentacross choice situations than assumed by econo-mists (Shapira, 1995). This is reflected by prospecttheory’s value function, which implies that in-dividuals value additional wealth more when theyare below their reference point. Further, if an in-dividual’s expected outcome is framed as a loss—when the expected outcome is below the subjectivelydetermined reference point—individuals will takemore risk to avoid this loss, creating the potential foraccentuating the loss. Yet if the expected outcome isframed as a gain (an expected outcome above theirreference point) theywill take less risk to avoid losingwealth (Kahneman & Tversky, 1979).

In addition to the value function, prospect theoryprovides a probability weighting function, demon-strating that objective probabilities and decisionweights will typically not reflect the probability as-sociatedwith the outcome. The probabilityweighting

function provided by Kahneman and Tversky(1979) suggests that individuals will underweightmost probabilities (especially larger ones), but as theprobability approaches zero they will overweight theevent. Changes in decision weights corresponding tochanges in probabilities are larger as probabilitiesapproach certainty (i.e., 0% or 100%) (Allais, 1953)and are smaller in the middle range of probabilities(Fennema & Wakker, 1997). An additional aspect ofthis body of research is the principle of diminishingsensitivity, suggesting that individuals value incre-mental gains less as their subjectively determinedgain increases (Tversky&Kahneman, 1992). Decisionweights and diminishing sensitivity are importantaspects of prospect theory that behavioral agency re-search has been criticized for failing to incorporate(Holmes et al., 2011).

An independent line of theorizing developed byorganizational behaviorists created predictions thatappear to correspond to those of prospect theory.The behavioral theory of the firm (Cyert & March1963/1992) suggests that organizational behavioris performance dependent, such that performancebelow aspirations leads organizations to engage insearch behavior (e.g., Bromiley, 2009; Camerer, 1989;Cohen, Jaffray, & Said, 1987; Weymark, 1981). Thisfailure to achieve desired performance goals encour-ages behavior that some scholars have equated to risktaking (Wiseman & Bromiley, 1996). This theory hasenrichedour understanding of risk taking by assistingus in understanding how the target (or aspiration)is determined and the inherent limitations of anindividual’s ability to process all relevant informa-tion to inform risk choices (bounded rationality). Ithas also informed prospect theory scholars to con-sider aspirations as a referent for determining gainand loss contexts (Lopes & Oden, 1999).

Other research grounded in psychology, socialpsychology, and sociology has also explored in de-tail individual risk behavior, such as how risk tak-ing may differ across financial, social, and ethicaldomains (Weber, Blais, & Betz, 2002); how it maybe affected by social structures (Westphal & Zajac,2013); how it may be influenced by personalitytraits (Herrmann & Nadkarni, 2014; Li & Tang, 2010;Lopes, 1984, 1987; Simon, 1955; Wowak & Hambrick,2010); and how contextual variables such as anxiety(Mannor et al., 2016) and vulnerability (Martinet al., 2013) may affect risk-taking behavior. Ingeneral, however, prospect theory’s framing effectand the concept of loss aversion have assumeda more prominent role in the behavioral agencyliterature.


Infusing traditional agency theory with the afore-mentioned research from psychology and behavioralscience, the behavioral agency model relaxes tradi-tional agency theory’s restrictive assumption of con-sistent preference orderings, fundamentally alteringpredictions of agent risk preferences (Gomez-Mejiaet al., 2000; Wiseman & Gomez-Mejia, 1998). Specifi-cally, this approach relaxes the assumption that in-dividuals consistently prefer less risk to more andmore wealth to less. For instance, agents’ subjectivelydetermined goals (equating goals with prospect theo-ry’s reference levels) for their compensation-relatedwealth will affect how they value the compensationthey receive and thus their risk preferences (Heath,Larrick, & Wu, 1999). This is based on prospect theo-ry’s value function, which suggests that individualswillvalueadditionalwealth that satisfies awealthgoalmore thanwealth that exceeds that goal (Kahneman&Tversky, 1979; Lopes, 1984). It follows that agentswillnot value all forms of pay equally, even when theyappear to have the same expected value. Agents mayendow and therefore assign higher values to someforms of compensation over others simply becausethe agent perceives that they have claimant rights tothat pay (the endowment effect), regardless of whethertheyhold legal rights to it (Franciosi,Kujal,Michelitsch,Smith, & Deng, 1996; Thaler, 1980).

Recognizing differential valuation of pay suggeststhat stock options would indeed have downside riskpotential, which creates risk bearing and weighs onrisk taking (Beatty&Zajac, 1994).However, agents arealso expected to weigh potential for further gainsagainst these potential losses when making deci-sions that influence theirwealth (Martin et al., 2013;Tversky & Kahneman, 1992). Thus stock options,like equity ownership, reflect a mixed gamble inwhich gains as well as losses are possible (Martinet al., 2013).

Empirical findings. Drawing on behavioral de-cision research and the aforementioned theory,various empirical studies in this streamhave soughtto predict individual and agent risk behavior inresponse to equity-based pay (Devers et al., 2008;Larraza-Kintana et al., 2007; Martin et al., 2013;Sanders, 2001; Sanders & Carpenter, 2003; Sanders&Hambrick, 2007). Common themes emerging fromthis empirical literature are the role of loss aversion,the influence of agent risk bearing, and the degreeto which the agent has endowed different typesof compensation-related wealth. Risk bearing is ameasure of wealth at risk of loss and, according tothe concept of loss aversion, should be negativelyrelated to risk taking (Wiseman & Gomez-Mejia,

1998). Providing support for this behavioral ap-proach to the agency problem, risk bearing has beenfound to negatively influence agent risk taking,or decisions that may threaten accumulated equitywealth (Devers et al., 2008; Larraza-Kintana et al.,2007; Lim & McCann, 2013; Martin et al., 2013;Sanders, 2001; Sanders & Carpenter, 2003; Zhang,Bartol, Smith, Pfarrer, & Khanin, 2008). Empiricalsupport has also been provided for the behavioraltheory of the firm’s predictions with regard to as-pirations and performance in the context of CEOrisk behavior (Harris & Bromiley, 2007).

A further theoretical insight from empirical re-search is the value of prospect theory’s framingconcept and the associated use of reference pointswhen predicting executive and firm behavior. Theexecutive (or agent) is likely to have a goal (or refer-ence point, in prospect theory’s language) in mindfor the value of his or her equitywealth (Devers et al.,2008; Harris & Bromiley, 2007; Zhang et al., 2008). Infact, individuals have been demonstrated to havemultiple goals and reference points (Lehner, 2000).These findings provide support for behavioral agen-cy’s prediction that the agent’s goals influence his orher riskpreferences inwaysnotpredictedby expectedutilitymodels of agency (see, for instance,Holmstrom,1979). Said differently, the ability to predict agent risktaking appears to have been enhanced by incorporat-ing problem framing and behavioral theory moregenerally within agency theory’s frameworks.

Empirical behavioral research has also examinedthe contextual situations that influence the effi-ciency of incentive alignment andmonitoring. Thesestudies have revealed that incentive alignment ismore effective than monitoring as a means of con-trolling agency costs, that higher-risk firms are lesslikely to use incentive compensation due to the costsof having the agent bear this risk, and that highermonitoring costs due to more complex operationslead to greater use of incentive alignment (Beatty &Zajac, 1994; Tosi, Katz, &Gomez-Mejia, 1997). Thesefindings provide some empirical validation forpropositions emerging from traditional agency the-ory, as outlined by Eisenhardt (1989). However, thefailure to find a strong relationship between firmperformance and the use of incentive alignment (orboard monitoring) has not been supportive of tradi-tional agency theory (Daily, Dalton, & Cannella,2003; Kolev et al., in press; Tosi, Werner, Katz, &Gomez-Mejia, 2000).

In sum,organizationalandbehavioral scholarshavecombined traditional agency theory with behavioraldecision research, embracing the concepts of agent


loss aversion, endowment, and prospect theory’s val-uation curve to predict agent risk behavior. Empiricalbehavioral research remains nascent, yet findings ap-pear to be supportive of behavioral agency theoreticalframeworks.

Comparing Economic and Behavioral Views ofRisk Sharing

Both financial economists and behavioral scholarsdraw on the aforementioned seminal work of agencytheorists such as Jensen, Meckling, Fama, andMurphy. The inspirational work of these financialeconomists provided the foundations of agencytheory, paving the way for the subsequent large vol-ume of research—in both finance and behavioralstreams—by scholars who have gone in search ofmechanisms for curbing the opportunistic behaviorof managerial agents and resolving the contractingproblems of moral hazard and adverse selectionthrough some formof risk sharing. Thus, both groupsof scholars have used as a starting point the as-sumptions of classical agency theorists, includingan opportunistic agent, bounded rationality and in-formation asymmetry. However, since Eisenhardt’s(1989) lucid explanation of agency theory and itssubsequent kindling of interest for behavioral schol-ars, the two groups—finance and behavioral science—have diverged significantly in terms of their methodsand theory. As we elaborate below, this divergencederives primarily from different conceptions of agentrisk preferences.

Theoretical differences. Finance scholars haveheld steadfastly to the original assumptions of neo-classical economics in the studyof agent riskbehaviorand, inparticular, theassumptionsof agent rationalityand self-interest, as the following quotes indicate(italics added for emphasis):

A considerable body of theory posits that employeestock options (ESOs) offer incentives to risk-aversemanagers to invest in high-risk high-return projectson behalf of risk-neutral shareholders. (Rajgopal &Shevlin, 2002, p. 146)

The sensitivity of stock options’ payoff to return vol-atility, or vega, provides risk-averse CEOs with anincentive to increase their firms’ risk more by in-creasing systematic rather than idiosyncratic risk.(Armstrong & Vashishtha, 2012, p. 70)

The apparent rejection of behavioral theory in thestudy of the agency problem by finance scholars is atodds with other streams of accounting and financeresearch that have adopted a behavioral approach

(e.g., Barberos & Thaler, 2003; Farrell, Goh, &White,2014; Fung, 2015; Thaler, 2005). The field of behav-ioral finance is yet to be applied to the study of agentrisk taking and incentive alignment. Thus, the focusof finance research continues to be on a contractingproblem of risk sharing based on the assumption thatagents are overinvested in their firms, leading them toavoid risk. This has led to a focus on bridging the pur-portedriskdifferentialbetweenagentandprincipalsbyencouraging the agent to adopt a risk posture closer tothat of the firm’s principals through risk sharing. A re-lated concern in this literature has been balancing thebenefits of risk sharingwith the costs, given that agentsare assumed to require higher compensation for accept-ing greater risk (Hoskisson et al., 2009).

The theoretical approaches of behavioral scholarsdiffer on various fronts. First, according to behav-ioral decision research—which has formed the basisof the management theorizing—individual prefer-ence orderings, including those regarding risk, arenot consistent across decision contexts (Kahneman&Tversky, 1979; Tversky & Kahneman, 1992). In par-ticular, an agent’s risk preferences are a function ofwealth relative to wealth goals and prospects forchanges towealth (Gomez-Mejia et al., 2000;Wiseman& Gomez-Mejia, 1998). Both streams continue to ex-amine antecedents to agent risk taking using thesedifferent theoretical platforms.

A second point of contention is the normativeview among finance scholars that higher executiverisk taking is good for shareholders and thereforeshould be incentivized to the point where the an-ticipated benefits of risk sharing no longer exceedthe cost of compensating agents for accepting risk-sharing agreements. Despite the fact that the capitalasset pricing model (CAPM) has proven valuable inpricing debt or equity in the capitalmarkets throughthe prescription that higher risk demands higherreturns, when the relationship between risk andreturn is studied longitudinally using panel data(Andersen, Denrell, & Bettis, 2007; Bromiley, 1991)or in cross-section (Bowman, 1980; Wang, Yan, &Yu, 2012), empirical evidence from managementresearch suggests that risk and return may be nega-tively related for themajority of industries and someinvestors.3 Reinforcing these empirical findings is

3 This empirical finding has been questioned on thebasis of endogeneity and identificationproblemswith risk-return models, yet after correcting for these issues, thenegative relationship has been found to persist (Henkel,2009). For a detailed review of the risk-return literature,refer to Nickel and Rodriguez (2002).


the overwhelming view among regulators and busi-ness leaders that excessive and careless risk takingcontributed to the depth of the Great Recession(Basel Committee, 2009; Bushman&Williams, 2015;Geithner, 2009). These views suggest that agent–principal contracts that encourage greater agent risktaking—assumed to reduce agency costs by financescholars—may trigger unintended negative conse-quences for shareholder wealth as well as society asa whole.

Third, finance research examining optimal agentcontracting has assumed that all forms of wealth arefungible. By contrast, behavioral research has dem-onstrated that individuals (including agents) placedifferent values on different forms of wealth (Devers,Wiseman, & Holmes, 2007). How the agent valueswealth depends on whether the agent’s wealth goalshave been achieved and the agent’s subjective esti-mations of prospective changes to current wealth(Larraza-Kintana et al., 2007). For example, agentsmay shift their risk preferences over time as the valueof their stock option portfolio adjusts due to changesin the firm’s stock price (Martin et al., 2013). Last,a recent stream of research focusing on family prin-cipals and family agents shows that for these parties,the framing of gains and losses may be socioemo-tional rather than financial in nature (Berrone et al.,2010, 2012; Gomez-Mejia et al., 2001, 2003, 2007,2010b, 2011, 2014, in press; Le Breton-Miller &Miller, 2013). These behavioral findings suggest thatagent responses to financial incentives may be morenuanced and complex than presumed by financialmodels of agent compensation.

Empirical differences. The two streams of litera-ture differ markedly in their methods as well. First,the exogenous variables used in finance research,such asdelta, gamma, and vega of stock options usedto predict agent risk taking and model the optimalcontract, have not found their way into the behav-ioral literature. Further, finance scholars suggest thatagents can opportunistically manipulate their con-tractsbasedon the riskorcharacteristicsof theiroptionportfolios to maximize compensation (e.g., Rajgopal &Shevlin, 2002). These features of finance research aresymptomatic of an attention to precise technical de-tails and more extensive use of inductive quantitativeanalysis, which distinguishes this literature from thatof organizational scholars.

The finance approach contrasts with most behav-ioral work that focuses on the value of stock optionawards inpredicting agent risk taking (e.g., Carpenter,2000; Lim&McCann, 2013; Sanders, 2001; Sanders &Hambrick, 2007). However, some behavioral scholars

have challenged this focus on the value of optionawards by finding that agents subjectively value stockoptions differently than the objective values of thoseoptions reported in proxy statements (Devers et al.,2007). Others have found evidence that accumulatedcash value of in-the-money options plays a strongerrole in determining agent risk taking than the value ofoption awards calculated in the year of their award(Devers et al., 2008). More recently, Martin et al.(2013) compared the existing cash value of stock op-tions against prospective futurevalue of those optionsto examine how both values might interact to influ-ence risk behavior. This research comes closer tobridging behavioral and financial research, given thatbehavioral scholars are considering the possibilitythat the agent may take additional risk in response tooption grants, consistent with Jensen and Meckling’s(1976) original assertion.

An area of intersection between behavioral andfinance examinations of agency are the measuresof agent risk taking, which are similar across bothfields. These include R&D, capital expenditure(CAPX), leverage (Devers et al., 2008;Kish-Gephart &Campbell, 2015; Larraza-Kintana et al., 2007; Martinet al., 2013), and adoption of new technology (Li &Tang, 2010). These types of investments (ones thatincrease a firm’s fixed costs) have been referred to asvariance-increasing investments in the finance lit-erature (Rajgopal & Shevlin, 2002). Interestingly,finance research has revealed that stock options’convexity has a negative effect on CAPX but a posi-tive effect on R&D and leverage (Coles et al., 2006).This raises questions over the use in behavioral lit-erature of thesemeasures as proxies forCEOstrategicrisk taking.

A major difference in methods between financeand behavioral examinations of agency rests on theuse of formal modeling. Finance scholars have usedformal modeling in their search for an optimal con-tract that balances the costs and benefits of risksharing arrangements. In contrast, behavioral re-search has made less use of formal modeling and hasoverlooked the idea that compensation for additionalrisk bearing is necessary. For instance, behavioralscholars have examined factors related to an in-dividual’s personality and how those factors maydrive agent behavior (Hayward & Hambrick, 1997;Herrmann & Nadkarni, 2014; Li & Tang, 2010). Fur-ther, empirical behavioral research has demonstratedthe potentially destructive effects of encouraginggreater risk taking and thus the potential perverseincentives created by granting options to executives(Denya et al., 2005;Martin et al., 2015; Sanders, 2001;


Sanders & Hambrick, 2007; Sitkin & Pablo, 1992).While the destructive effect of options has been re-cently addressed by accounting scholars (Flor et al.,2014) and finance scholars have considered factorsthat may result in value destruction (e.g., Harford,Humphery-Jenner, & Powell, 2012), they have yet toconsider the value-destroying effect of options whenmodeling optimal agent contracting.

PROGRESS REPORT

Asdescribedabove,much theorizingandempiricalanalysis has been undertaken across the disciplineswith regard to the contracting problem of risk sharingand understanding antecedents to executive (agent)risk taking. The central concern of finance agencyresearchhas been specifying the optimal contract thatwill induce agent effort through incentive alignmentand reduce agency costs; the focus of behavioral re-search has been explaining how agents respond todifferent risk-sharing arrangements under differentconditions of uncertainty. In this section we review(1) theoretical and empirical progress and (2) how thisprogress has benefited practice.

Theoretical and Empirical Progress

Has the abundance of theorizing and empirical re-search succeeded in enhancing our understanding ofhow we can limit the opportunistic actions of mana-gerial agents by successfully aligning their risk be-havior with that desired by the firm’s shareholders?One can argue that advancements in our under-standing of antecedents to agent risk behavior haveprovided the theoretical basis to limit agency costsassociatedwith risk profile divergence. This has beenachieved by identifying levers available to share-holders to regulate agent risk taking—or strategicchoices that influence the firm’s risk profile. For ex-ample, behavioral scholars have found support fortheory drawing on prospect theory and the concept ofmixed gambles, demonstrating that CEOs’ risk be-havior will depend on their estimates of what they togain relative to what they have to lose (Martin et al.,2013; Tversky &Kahneman, 1992). Similarly, financescholars have developed a strong understanding ofhow a CEO’s option portfolio—and in particular, thedynamic characteristics of options as representedby the “option Greeks”—influence his or her riskbehavior (e.g., Coles et al., 2006; Guay, 1999). Thus,advancements in understanding of antecedents toagent risk taking appear to have been of theoreticaland practical value.

Despite this progress, open questions remain. Forinstance, our understanding of antecedents to risktaking other than compensation design remainslimited. Such alternate antecedents could includethe choice of CEO, executive investments outside ofthe firm, and personality traits. These antecedents toagent risk taking and agency costs are underexploredand have great potential as future avenues of agencyresearch.

Practical Implications

Practical value of research acts as a good test oftheoretical progress, and thus scrutiny of the theory’spractical implications allows us to build on the priorsection (Corley & Gioia, 2011). Questioning whethertheoretical advancements have translated into valuefor practitioners is a particularly salient question inlight of theblatant governance failures that havebeenassociated with the Great Recession. Many prom-inent public figures have unambiguously attributedthe depth of the Great Recession and subsequenteconomic malaise to excessive and careless risktaking in the years leading up to 2007 (Bushman &Williams, 2015). One can point to the huge di-vergence in the financial fates of executives who ledfirms prior to their demise and the firm’s share-holders as further evidence of the failures of corpo-rate governance—or failure to align the fortunes ofagent and principal—in practice. As an extreme ex-ample, Charles Prince left his position as CEO ofCitigroup with stock and options of approximately$160 million while the firm he led incurred manybillions of dollars of losses and required a bailout.This and the various other bonuses granted to exec-utives at the height of the financial crisis suggest thatincentive alignment of the fates of principal andagent has been lacking. This apparent failure ofincentive alignment in practice once more suggestslack of appropriate guidance or failure to influencepractice with existing theoretical guidance.

These examples of governance failures and theassociatedagencycosts raisequestionssuchaswhetheragency research contributed to these failures andwhether such research could have mitigated (or pre-vented) theexcessive risk taking thathasbeenthe focusof public debate. There are obviously no clear answersto these questions; all possibilities are shrouded insubjective opinion and therefore open to criticism. Itappears, however, that the excessive risk taking—particularly within (but certainly not restricted to) fi-nancial services—was not anticipated by the boardsor compensation committees that are charged with


ensuring that executives behave in the interests of thefirm’s principals.

We noted above that the theory allows for pre-dicting CEO risk behavior in response to the stockand stock options thatwere large components of totalpay (e.g., Devers et al., 2008; Lim & McCann, 2013;Sanders & Hambrick, 2007). Yet it appears that thesetheoretical insights were not utilized by boards andcompensation committees. This suggests that thesepractitionerswereeither awareof the theoryyet choseto ignore it due to internal politics or regulatory con-straints, or were ignorant of recent theoretical de-velopments that could have informed their decisionsregarding the rewarding of key agents. With regard tothe first explanation, as noted by Bebchuk and Fried(2004), CEO power may have ensured that the com-pensation contract was designed to achieve the ob-jective of CEO enrichment rather than encouragingsound risk taking in the interests of a wider group offirm stakeholders. Government policies with regardto compensation may also have restricted the abilityor willingness to design compensation contracts thatachieved better shareholder outcomes (such as taxlegislation). If the second explanation (not aware ofacademic research’s insights) is valid, it would reflectpoorly on business academia for not having themechanisms in place to communicate the research orperhaps for not having the credibility to capture theattention of practitioners.

Finally, perhaps agency scholars can be criticizedfor propagating the view that agent (or CEO) risktaking should be encouraged. This is an implicationof agency theory’s assumption that CEOs eschewoptimal risk taking and thus must be incentivized totake more risk to satisfy the risk-neutral shareholderprincipals. This paradigm continues to guide financescholars. If practitioners are guided by the advice andassumptions of financial economics, it is possible thatthey are (orwereprior to theGreatRecession) anchoredto and guided by this perspective. In the post-recessionera,“risk” is sometimesconsidered tobea “dirtyword”in corporate boardrooms, as prudent risk taking be-comes the dominant theme (Geithner, 2009). This em-phasizes the need for agency scholars, regardless oftheir field, to elucidate the caveat that agents’ attitudestoward risk may not necessarily be misaligned withthose of shareholders and thus should not necessarilybe discouraged, but instead should be understoodwithin the context in which the agent is operating.

This review of the practical implications of agencyresearch provides various ideas for future studies.First, theproblems identifiedwith thepractical utilityof agency research suggest that there may be value in

revisiting assumptions and definitions that havedriven much of the agency research in an attempt toimprove the predictive validity of the theoreticalframeworks used. For instance, the assumption ofa risk differential between agents and principals restspartly on viewing principals as universally risk neu-tral. Given that agents have been shown to exhibitdiverse risk profiles, recognizing that principals mayexhibit similar diverse attitudes toward risk raisesthis question: When is risk preference divergencebetween agent and principal most likely to occur?

Another question emerging from the above reviewof practical implications is this: When will agent–principal risk preference divergence be good or badfor shareholders or other firm stakeholders? Firmswith greater risk profile divergence (based on theassumptions of a risk-averse manager and risk-neutral shareholders) before the recent crisis mayhave been less exposed to the negative performanceimplications of the financial downturn that began in2008, challenging the assumption that risk profiledivergence is bad for shareholders. Along similarlines, we must ask what constitutes “prudent” risktaking in practice. Because of an explicit or impliedrationality assumption and dogmatic adherence toa “high risk/high return” paradigm, financial econ-omists are generally less concerned with a distinc-tion between good and bad risk taking. Behavioralscientists also have much to contribute to a betterunderstanding of the concept of “prudent” agent risktaking and how it may be encouraged (or otherwise)by equity-based pay.

Finally,due toconstraintsonaccess todata, therehasbeen limited research on risk-sharing effects on be-havior, controlling for the agent’s entire wealth portfo-lio. Clearly, agents seek to limit their exposure tofirm-specific risk by, for example, exercising optionsand selling the resulting stock, thus limiting theportionof personal wealth that is exposed to firm risk. Assess-ing risk sharing in the context of total agentwealth is anarea ripe for further examination (Zajac, 2006).

A RESEARCH AGENDA BASED ONCROSS-FERTILIZATION

What is evident from the difference in approachesof finance and behavioral scholars is that the twostreams have operated mostly in separate silos withlimited acknowledgment of each other’s contribu-tions. One could argue that this approach has im-peded our combined progress in advancing agencytheory and our understanding of antecedents toagent risk taking and how to optimize the use of


equity-based pay. Certainly, practitioners have em-braced cross-functional cooperation, given the needto ensure goal alignment and todealwith the obviousinterreliance across functions—including finance,operations, marketing, in-house strategic planning,and treasury—in pursuing the firm’s overarchingstrategic goals. Yet this cross-functional approach toproblem solving appears to be seldom applied inbusiness scholarship.

The failure to more explicitly acknowledge andintegrate the different streams of research is drivenpartly by fundamental differences in philosophiesaboutwhat constitutes a “contribution.”But this doesnot negate the possibility of cross-fertilization of ideasacross these streams. Below we outline how bothstreams could benefit and research opportunities thatcould evolve from combining the findings and ad-vancements of both streams.

First, behavioral research could benefit fromutilizing the impressively rigorous findings andmethods that have emerged from the finance stream.For instance, the finding that executive stock optionsinfluence capital expenditures differently fromotherdecisions (such as R&D spending and leverage poli-cies) could help behavioral scholars to refine theiroperationalizations of risk taking (see Coles et al.,2006). That insight came from focusing on variousoption dynamics, such as the exponential growth ofoption returns (cf. Coles et al., 2006). The robustfindings that the characteristics of options repre-sented by “the Greeks” (i.e., delta, vega, and gamma)influence agent risk behavior have been largely ig-nored by behavioral scholars, who have focusedinstead on the value of annual option pay or theaccumulated value of options.

A reason for not examining these option character-istics can be defended by behavioral researchers on thegrounds that these are not metrics that the agent canreadily observe and that therefore, due to bounded ra-tionality, the agent is less likely to be cognizant of suchmetrics when making decisions under uncertainty.Behavioral researchers are concerned with heuristicsthat directly influence the agent’s calculus of wealth atrisk of loss versus prospective wealth to be gained(i.e., amixedgamble;Martinetal., 2013).Thechallengefor behavioral researchers lies in building a bridgebetween their heuristics and the exogenous variablesused by finance scholars. The robust findings emerg-ing from finance research suggest that agents may bemore sophisticated in understanding the metrics as-sociatedwith their optionportfolios thanmanagementscholars have given them credit for. For instance,agents (especially those with financial advisers, such

as CEOs) may have at least a basic understanding ofhow the value of their options changes in response tostock price changes (gamma, delta) or changes involatility (vega). Thus, it may be worth revisiting theheuristics associatedwith executive stockoptions thatsubsequently influence agent behavior.

As an example of the possibilities for utilizing thedynamic qualities of stock options in behavioral re-search, behavioral scholars could use the concept ofconvexity, known as gamma, to understand whyagents may take more risk in the presence of highlevels of current equity wealth. Convexity capturesthe degree of the exponential increase in the agent’soption wealth as stock price increases (e.g., Chava &Purnanandam, 2010; Cohen et al., 2000; Guay, 1999;Rajgopal & Shevlin, 2002). The convex nature ofstock option payoffs could be used to delve furtherinto how agents may respond to stock options overthe life cycle of those options. The sensitivity of thevalue of options to volatility (vega) could also be ofinterest to behavioral scholars given the conflict ofinterest this creates between agent and principal;that is, higher volatility of a firm’s stock options willenhance the wealth of an agent holding them. Yetthis volatility also creates costs for the shareholderprincipal due to higher funding costs and often neg-ativeperformanceconsequencesof earningsvolatility(Andersen et al., 2007; Henkel, 2009). Thus, the pos-itive relationship between volatility and agent optionwealth highlights potentially perverse incentives thatare underexplored by behavioral research.

Behavioral research could also benefit from em-bracing the concept of agent hedging (managing therisk of adverse price movements), which has been ex-plored by finance research (e.g., Gao, 2010). Althoughsteps in this directionhavebeenmade (cf.Martin et al.,2013), the vast majority of behavioral research con-tinues toview theCEOagent as apassive actor. It couldbe argued that golden parachutes (large executiveseverance payments) act as a hedge against failed risktaking, but despite being of interest to managementscholars in other research contexts (Wade, O’Reilly, &Chandratat, 1990) they are yet to be integrated withagency research examining agent risk behavior.

The use of behavioral theory by finance scholars hasbeen limited. In particular, there is often a failure toembrace the robust findings from behavioral decisionresearchdemonstrating that individualsare lossaverseand their riskpreferences are contextdependent.4This

4 As noted below, an exception is the work by RichardThaler and colleagues (see Barberos & Thaler, 2003, fora review).


is reflected in the theorizing and models based on as-sumptions of a risk differential between agents andprincipals and rigorous definitions for risk aversionand risk neutrality (while behavioral scholars tend touse the terms somewhat more loosely). Recognitionthat agent risk preferences are reference dependentand thatmanagerial agents are prescientwith regard toboth the upside and downside consequences of risktaking could substantially alter theorizing with regardto agent risk behavior and the optimal principal–agentcontract. For example, viewing stock options as amixed gamble (Bromiley, 2009) recognizes that indi-vidualsmay framehowtheyvalue thecurrent intrinsicvalue of their options relative to a projection of whatthey may be worth in the future (Martin et al., 2013).Incorporating prospect theory–driven modificationssuch as this into formal modeling of the principal–agent contract (commonly used by finance scholars)could greatly assist in furthering our understandingof the optimal use of stock and options in executivecompensationcontracts.This integrationofbehavioraldecision theory with the econometric rigor previouslyapplied by financial economists to the study of agentrisk taking has the potential to significantly advanceknowledge regarding principal–agent risk profile di-vergence and associated agency costs.

A further theoretical development that may benefitfinance scholarship is the notion from behavioral re-search that it is also necessary to explicitly acknowl-edge and examine the possibility that opportunisticbehavior is not the sole domain of managerial agents.As explicated by Werder (2011), it is possible thatvarious stakeholders may behave opportunistically.This necessitates the study of how risk sharing occursamong actors other than shareholders and managers,and the implications of this risk sharing on contribu-tions to firm value. It also suggests that the influenceof shareholders should be a further subject of study,given their scope for opportunistic behavior at theexpense of other firm stakeholders.

An example of both behavioral and financialeconomics research benefiting from integration andcross-fertilization of empirical and theoretical ad-vances from each field is the study of intertemporalchoice and frequency of asset performance evalua-tion. This research stream was inspired by thequestion of why investors hold bonds when stocksprovide returns that are substantially higher (theequity premium puzzle; Benartzi & Thaler, 1995).Integrating the behavioral concepts of loss aversionand framing, this research found evidence that de-cision makers (investors) accept higher risks whenaggregation leads to appraisal of investments less

frequently and over longer time frames (Benartzi &Thaler, 1995; Loewenstein & Thaler, 1989; Thaleret al., 1997). The idea that investors (and individ-uals) are less sensitive to losses incurred over thelonger term than more immediate losses—myopicloss aversion—has been applied by behavioralscholars to the study of family firm decision mak-ing (Chrisman & Patel, 2012) and the CEO’s re-sponse to equity-based pay (Martin, Wiseman, &Gomez-Mejia, in press).

Last, both fields would benefit from further ex-ploring the role of nonfinancial utilities of decisionmakers when predicting risk behaviors of execu-tives. As noted earlier, most of this work has focusedon family owners and family agents, and argues that“socioemotional wealth” (SEW) represents an im-portant stock that these individuals want to protecteven if this involves the trade-off of accepting lowereconomic returns (Gomez-Mejia et al., 2014, in press;Martin & Gomez-Mejia, in press). That is, these in-dividuals may be loss averse to both SEW and eco-nomic wealth, yet in their frame of reference SEWtends to have a higher priority except when firmsurvival is in question (Gomez-Mejia et al., 2007,2011). Some of this theoretical development mayalso be pertinent to realms beyond the family contextand opens the door to much-needed research involv-ing “mixed gambles” between financial and non-financial utilities. For instance, Zona et al. (in press)argued that executives often engage in interlocks togain prestige and influence in the industry ratherthan for financial reasons. Likewise, Chatterjee andHambrick (2007) argued that top executives of largecorporations are often driven by nonfinancial goals(such as empire building, narcissism, and power)and that this helps definewhat theymay consider asa gain or a loss.

CONCLUSION

It is evident that both the finance and behavioralfields could gain from paying greater heed to thefindings and methods of the other field. We suggestthat the failure to do so has impeded the progressof both groups of scholars. Despite these impedi-ments, much progress has been made in predictingagent risk taking, understanding the behavioraleffects of equity-based pay, managing agent riskincentives, and in theory understanding how tomitigate agency problems. Yet these advancementshave failed to limit significant governance failuresand related agency costs in practice. We have out-lined an agenda for greater cross-fertilization of the


two fields and trust that this will expedite furtheracademic advancement.

REFERENCES

Aggarwal, R. K., & Samwick, A. A. (2002). Executivecompensation, competition, and relative performanceevaluation: Theory and evidence. Journal of Finance,54, 1999–2043.

Ai, H., & Li, R. (2015). Investment and CEO compensationunder limited commitment. Journal of FinancialEconomics, 116(3), 452–472.

Allais, M. (1953). Le comportement de l’homme rationneldevant le risque: Critique des postulats et axiomes del’Ecole Americaine. Econometrica, 21(4), 503–546.

Amihud, Y., & Lev, B. (1981). Risk reduction as a manage-rial motive for conglomerate mergers. Bell Journal ofEconomics, 12, 605–617.

Anantharaman, D., & Lee, Y. G. (2014). Managerial risktaking incentives andcorporatepensionpolicy. Journalof Financial Economics, 111(2), 328–351.

Andersen, T. J., Denrell, J., & Bettis, R. A. (2007). Strategicresponsiveness and Bowman’s risk-return paradox.Strategic Management Journal, 28, 407–429.

Armstrong, C. S., Larcker, D. F., Ormazabal, G., & Taylor,D. J. (2013). The relation between equity incentivesand misreporting: The role of risk-taking incentives.Journal of Financial Economics, 109(2), 327–350.

Armstrong, C. S., & Vashishtha, R. (2012). Executive stockoptions, differential risk-taking incentives, and firmvalue. Journal of Financial Economics, 104(1), 70–88.

Barberos, N., & Thaler, R. (2003). A survey of behavioralfinance. InG.M.Constantinides (Ed.),Handbook of theeconomics of finance (pp. 1053–1128). Amsterdam:Elsevier.

Basel Committee on Banking Supervision. (2009, July).Enhancements to the Basel II framework. Basel,Switzerland: Bank for International Settlements.

Beatty, R. P., & Zajac, E. J. (1994). Managerial incentives,monitoring, and risk bearing: A study of executivecompensation, ownership, and board structure in ini-tial public offerings.Administrative ScienceQuarterly,39, 313–335.

Bebchuk, L., & Fried, J. (2004). Pay without performance:The unfulfilled promise of executive compensation.Cambridge, MA: Harvard University Press.

Benartzi, S., & Thaler, R. H. (1995). Myopic loss aversionand the equity premium puzzle. Quarterly Journal ofEconomics, 110(1), 73–92.

Bergstresser, D., & Philippon, T. (2006). CEO incentives andearningsmanagement. JournalofFinancialEconomics,80(3), 511–529.

Bernoulli, D. (1954). Exposition of a new theory on themeasurement of risk (L. Sommer, Trans.). Econo-metrica, 22, 23-36. (Original work published 1738.)

Berrone, P., Cruz, C., & Gomez-Mejia, L. R. (2012). Socio-emotional wealth in family firms: Theoretical dimen-sions, assessment approaches, and agenda for futureresearch. Family Business Review, 25(3), 258–279.

Berrone, P., Cruz,C.,Gomez-Mejia, L.R., &Larraza-Kintana,M. (2010). Socioemotional wealth and corporate re-sponses to institutionalpressures:Do family-controlledfirms pollute less? Administrative Science Quarterly,55(1), 82–113.

Bertrand,M., & Schoar, A. (2003).Managingwith style: Theeffect ofmanagers on firmpolicies.Quarterly Journal ofEconomics, 118, 1169–1208.

Bettis, J. C., Bizjak, J. M., & Lemmon, M. L. (2001). Mana-gerial ownership, incentive contracting, and the use ofzero-cost collars and equity swaps by corporate in-siders. Journal of Financial and Quantitative Analy-sis, 36(3), 345–370.

Bowman, E. H. (1980). A risk–return paradox for stra-tegic management. Sloan Management Review, 21,17–31.

Breton-Miller, L., & Miller, D. (2013). Socioemotionalwealth across the family firm life cycle: A com-mentary on “Family Business Survival and the Roleof Boards.” Entrepreneurship Theory and Practice,37(6), 1391–1397.

Bromiley, P. (1991). Testing a causal model of corporaterisk taking and performance. Academy of Manage-ment Journal, 34(1), 37–59.

Bromiley, P. (2009). A prospect theory model of resourceallocation. Decision Analysis, 6(3), 124–138.

Burns, N., & Kedia, S. (2006). The impact of performance-based compensation on misreporting. Journal ofFinancial Economics, 79(1), 35–67.

Bushman, R. M., & Williams, C. D. (2015). Delayed ex-pected loss recognition and the risk profile of banks.Journal of Accounting Research, 53(3), 511–553.

Camerer, C. F. (1989). An experimental test of severalgeneralized utility theories. Journal of Risk and Un-certainty, 2, 61–104.

Carpenter, J. N. (2000). Does option compensation increasemanagerial risk appetite? Journal of Finance, 55(5),2311–2331.

Chatterjee, A., & Hambrick, D. (2007). It’s all about me:Narcissistic CEOs and their effects on company strat-egy and performance. Administrative Science Quar-terly, 52, 351–386.

Chava, S., & Purnanandam, A. (2010). CEOs versus CFOs:Incentives and corporatepolicies. Journal of FinancialEconomics, 97(2), 263–278.


Chrisman, J. J., & Patel, P. C. (2012). Variations in R&Dinvestments of family and nonfamily firms: Be-havioral agency and myopic loss aversion per-spectives. Academy of Management Journal, 55(4),976–997.

Coffee, J. C., Jr. (1988). Shareholders versus managers:The strain in the corporate web. In J. C. Coffee Jr.,L. Lowenstein, & S. Rose-Ackerman (Eds.), Knights,raiders, and targets: The impact of the hostile takeover(pp. 77–134). New York: Oxford University Press.

Cohen, D. A., Dey, A., & Lys, T. Z. (2013). Corporate gov-ernance reform and executive incentives: Implica-tions for investments and risk taking. ContemporaryAccounting Research, 30(4), 1296–1332.

Cohen, M., Jaffray, J., & Said, T. (1987). Experimentalcomparisons of individual behavior under risk andunder uncertainty for gains and for losses. Organi-zational Behavior and Human Decision Processes,39(1), 1–22.

Cohen, R. B., Hall, B. J., & Viceira, L. M. (2000). Do execu-tive stock options encourage risk taking? (WorkingPaper). Cambridge, MA: Harvard Business School.

Coles, J. L., Daniel, N. D., & Naveen, L. (2006). Managerialincentives and risk-taking. Journal of Financial Eco-nomics, 79(2), 431–468.

Core, J., & Guay, W. (1999). The use of equity grants tomanage optimal equity incentive levels. Journal ofAccounting and Economics, 28(2), 151–184.

Corley, K. G., & Gioia, D. A. (2011). Building theory abouttheory building: What constitutes a theoretical con-tribution? Academy of Management Review, 36(1),12–32.

Cyert, R. M., & March, J. G. (1963/1992). A behavioraltheory of the firm (1st/2nd ed.). Cambridge, MA:Blackwell Publishing.

Dai, Z., Jin, L., & Zhang, W. (2014). Executive pay–performance sensitivity and litigation. ContemporaryAccounting Research, 31(1), 152–177.

Daily, C. M., Dalton, D. R., & Cannella, A. A. (2003).Corporate governance: Decades of dialogue anddata. Academy of Management Review, 28(3),371–382.

Davis, J. H., Schoorman, F. D., & Donaldson, L. (1997). To-ward a stewardship theory of management. Academyof Management Review, 22(1), 20–47.

Demsetz, H., & Lehn, K. (1985). The structure of corporateownership: Causes and consequences. Journal ofPolitical Economy, 93(6), 1155–1177.

Denis, D. J., Denis, D. K., & Sarin, A. (1997). Ownershipstructure and top executive turnover. Journal ofFinancial Economics, 45(2), 193–221.

Denya, B., Gomez-Mejia, L. R., De Castro, J., &Wiseman, R.(2005). Incentive alignment or perverse incentives? Abehavioral view of stock options. Management Re-search, 3(2), 1–17.

Devers, C. E., McNamara, G.,Wiseman, R.M., & Arrfelt, M.(2008). Moving closer to the action: Examining com-pensation design effects on firm risk. OrganizationScience, 19(4), 548–566.

Devers, C. E.,Wiseman, R. M., & Holmes, R.M. (2007). Theeffects of endowment and loss aversion in managerialstock option valuation. Academy of ManagementJournal, 50(1), 191–208.

Eisenhardt, K. M. (1989). Agency theory: An assessmentand review.Academy ofManagement Journal, 14(1),57–74.

Fama, E. F. (1980). Agency problems and the theory of thefirm. Journal of Political Economy, 88(2), 288–307.

Farrell, A. M., Goh, J. O., &White, B. J. (2014). The effect ofperformance-based incentive contracts on system 1and system2processing in affectivedecision contexts:fMRI and behavioral evidence. Accounting Review,89(6), 1979–2010.

Fennema,H.,&Wakker,P.P. (1997).Originalandcumulativeprospect theory: A discussion of empirical differences.Journal of Behavioral Decision Making, 10, 53–64.

Flor, C. R., Frimor, H., & Munk, C. (2014). Options in com-pensation: Promises andpitfalls. Journal ofAccountingResearch, 52(3), 703–732.

Franciosi, R., Kujal, P.,Michelitsch, R., Smith, V., & Deng,G. (1996). Experimental tests of the endowment ef-fect. Journal of Economic Behavior & Organization,30(2), 213–226.

Fung, M. K. (2015). Cumulative prospect theory and man-agerial incentives for fraudulent financial reporting.Contemporary Accounting Research, 32(1), 55–75.

Gao, H. (2010). Optimal compensation contracts whenmanagers can hedge. Journal of Financial Economics,97, 218–238.

Geithner, T. (2009, March 26). Written testimony, HouseFinancial ServicesCommitteehearing.U.S.Departmentof Treasury. Retrieved from https://www.treasury.gov/press-center/press-releases/Pages/tg71.aspx

Gomez-Mejia, L.R., &Balkin,D.B. (1992).Thedeterminantsof faculty pay: Anagency theory perspective.Academyof Management Journal, 35(5), 1–36.

Gomez-Mejia, L. R., Berrone, P., & Franco-Santos,M. (2010a).Compensationandorganizationalperformance:Theory,research, practice. Armonk, NY: M. E. Sharpe.

Gomez-Mejia, L. R., Campbell, J., Martin, G., Hoskisson,R.E.,Makri,M.,&Sirmon,D.G. (2014). Socioemotionalwealth as a mixed gamble: Revisiting family R&D


https://www.treasury.gov/press-center/press-releases/Pages/tg71.aspx

https://www.treasury.gov/press-center/press-releases/Pages/tg71.aspx

investments with the behavioral agency model. Entre-preneurship Theory and Practice, 38, 1351–1374.

Gomez-Mejia, L. R., Cruz, C., Berrone, P., & De Castro, J.(2011). The bind that ties: Socioemotional wealthpreservation in family firms. Academy of Manage-ment Annals, 5, 653–707.

Gomez-Mejıa,L.R.,Haynes,K.T.,Nuñez-Nickel,M., Jacobson,K. J., &Moyano-Fuentes, J. (2007). Socioemotionalwealthand business risks in family-controlled firms: Evidencefrom Spanish olive oil mills. Administrative ScienceQuarterly, 52(1), 106–137.

Gomez-Mejia, L. R., Larraza-Kintana, M., & Makri, M.(2003). The determinants of executive compensationin family controlled companies. Academy of Man-agement Journal, 46(2), 226–237.

Gomez-Mejia, L. R., Makri, M., & Larraza-Kintana, M.(2010b). Diversification decisions in family-controlledfirms. Journal of Management Studies, 47, 223–252.

Gomez-Mejia, L. R., Nunez-Nickel, M., & Gutierrez, I.(2001). The role of family ties in agency contracts.Academy of Management Journal, 44, 81–95.

Gomez-Mejia, L. R., Patel, P. C., & Zellweger, T. M. (inpress). In the horns of the dilemma: Socioemotionalwealth, financial wealth, and acquisitions in familyfirms. Journal of Management.

Gomez-Mejia, L. R., Welbourne, T., & Wiseman, R. (2000).The role of risk sharing and risk taking under gainshar-ing. Academy of Management Review, 25(3), 492–512.

Grant, J., Markarian, G., & Parbonetti, A. (2009). CEO risk-related incentives and income smoothing. Contem-porary Accounting Research, 26(4), 1029–1065.

Guay,W. R. (1999). The sensitivity of CEOwealth to equityrisk: An analysis of the magnitude and determinants.Journal of Financial Economics, 53, 47–71.

Hall, B., & Murphy, K. (2002). Stock options of undiversifiedexecutives. Journal ofAccounting andEconomics,33(1),3–42.

Harford, J., Humphery-Jenner, M., & Powell, R. (2012).The sources of value destruction in acquisitions byentrenched managers. Journal of Financial Econom-ics, 106(2), 247–261.

Harris, J., & Bromiley, P. (2007). Incentives to cheat: Theinfluence of executive compensation and perfor-mance on financial misrepresentation. OrganizationScience, 18(3), 350–367.

Hayes, R.M., Lemmon,M., &Qiu,M. (2012). Stock optionsand managerial incentives for risk taking: Evidencefrom FAS 123R. Journal of Financial Economics,105(1), 174–190.

Hayward, M. L., & Hambrick, D. C. (1997). Explaining thepremiums paid for large acquisitions: Evidence of

CEO hubris. Administrative Science Quarterly, 42(1),103–127.

Heath, C., Larrick, R. P., &Wu,G. (1999). Goals as referencepoints. Cognitive Psychology, 38(1), 79–109.

Henkel, J. (2009). The risk-return paradox for strategicmanagement: Disentangling true and spurious effects.Strategic Management Journal, 30(3), 287–303.

Herrmann, P., & Nadkarni, S. (2014). Managing strategicchange: The duality of CEO personality. StrategicManagement Journal, 35(9), 1318–1342.

Holmes, R. M., Bromiley, P., Devers, C. E., Holcomb,T. R., & McGuire, J. B. (2011). Management theoryapplications of prospect theory: Accomplishments,challenges, and opportunities. Journal of Manage-ment, 37(4), 1069–1107.

Holmstrom,B. (1979).Moral hazard andobservability.BellJournal of Economics, 10, 74–91.

Hoskisson, R. E., Castleton,M.W., &Withers, M. C. (2009).Complementarity in monitoring and bonding: Moreintense monitoring leads to higher executive com-pensation. Academy of Management Perspectives,23(2), 57–74.

Jensen, M. C. (1994). Self-interest, altruism, incentives,and agency. Journal of Applied Corporate Finance,7(2), 40–45.

Jensen, M. C., & Meckling, W. H. (1976). A theory of thefirm: Governance, residual claims, and organizationalforms. Journal of Financial Economics, 3(4), 305–360.

Jensen, M. C., & Murphy, K. J. (1990). Performance pay andtop management incentives. Journal of Political Econ-omy, 98(2), 225–264.

John, T. A., & John, K. (1993). Top-management compen-sation and capital structure. Journal of Finance, 48(3),949–974.

Kahneman, D. (2012). The human side of decisionmaking:Thinking things through with Daniel Kahneman.Journal of Investment Consulting, 13(1), 5–14.

Kahneman, D., & Tversky, A. (1979). Prospect theory: Ananalysis of decisions and risk. Econometrica, 47,262–291.

Kahneman, D., & Tversky, A. (1984). Choices, values, andframes. American Psychologist, 39, 341–350.

Kim, Y., Li, H., & Li, S. (2015). CEO equity incentives andaudit fees.ContemporaryAccounting Research, 32(2),608–638.

Kish-Gephart, J. J., & Campbell, J. T. (2015). You don’tforget your roots: The influence of CEO social classbackground on strategic risk taking. Academy ofManagement Journal, 58(6), 1614–1636.

Kolev, K.,Wiseman, R.M., & Gomez-Mejia, L. R. (in press).Do CEOs ever lose? Fairness perspective on the


allocation of residuals between CEOs and share-holders. Journal of Management.

Larraza-Kintana, M., Wiseman, R. M., Gomez-Mejia, L. R.,& Welbourne, T. M. (2007). Disentangling compensa-tion and employment risks using the behavioralagency model. Strategic Management Journal, 28(10),1001–1019.

Lehner, J. M. (2000). Shifts of reference points for framingof strategic decisions and changing risk-return asso-ciations.Management Science, 46(1), 63–76.

Lewellen, K. (2006). Financing decisions when managersare risk averse. Journal of Financial Economics, 82(3),551–589.

Li, J., & Tang, Y. (2010). CEO hubris and firm risk takingin China: The moderating role of managerial dis-cretion. Academy of Management Journal, 53(1),45–68.

Lim, E. N., & McCann, B. T. (2013). Performance feedbackand firm risk taking: The moderating effects of CEOand outside director stock options. Organization Sci-ence, 25(1), 262–282.

Loewenstein, G., & Thaler, R. H. (1989). Anomalies: Inter-temporal choice. Journal of Economic Perspectives,3(4), 181–193.

Lopes, L. L. (1984). Risk and distributional inequality.Journal of Experimental Psychology: Human Percep-tion and Performance, 10, 465–485.

Lopes, L. L. (1987). Between hope and fear: The psychol-ogy of risk. Advances in Experimental Social Psy-chology, 20(3), 255–295.

Lopes,L.L.,&Oden,G.C. (1999).The role of aspiration levelin risky choice: A comparison of cumulative prospecttheory and the SP/A theory. Journal of MathematicalPsychology, 43(2), 286–313.

Luce, R. D., & Fishburn, P. C. (1991). Rank- and sign-dependent linear utility models for finite first-ordergambles. Journal of Risk and Uncertainty, 4, 29–59.

Mannor, M. J., Wowak, A. J., Bartkus, V. O., & Gomez-Mejia, L. R. (2016). Heavy lies the crown? How jobanxiety affects top executive decision making in gainand loss contexts. Strategic Management Journal,37(9), 1968–1989.

Martin,G.P.,&Gomez-Mejia,L.R. (inpress).Therelationshipbetweensocioemotional and financialwealth:Revisitingfamily firm decision making.Management Research.

Martin, G. P., Gomez-Mejia, L. R., &Wiseman, R.M. (2013).Executive stock options as mixed gambles: Revisitingthe behavioral agency model. Academy of Manage-ment Journal, 55(8), 451–472.

Martin, G., Washburn, N., Makri, M., & Gomez-Mejia,L. R. (2015). Not all risk taking is born equal: Thebehavioral agency model and CEO’s perception of

firm efficacy.Human Resource Management, 54(3),483–498.

Martin, G. P., Wiseman, R. M., & Gomez‐Mejia, L. R. (inpress). Going short‐term or long‐term? CEO stockoptions and temporal orientation in the presence ofslack. Strategic Management Journal.

May, D. O. (1995). Do managerial motives influence firmrisk-reduction strategies? Journal of Finance, 50,1291–1308.

Nickel, M. N., & Rodriguez, M. C. (2002). A review of re-search on the negative accounting relationship be-tween risk and return: Bowman paradox. Omega,30(1), 1–18.

Nyberg, A. J., Fulmer, I. S., Gerhart, B., & Carpenter, M. A.(2010). Agency theory revisited: CEO return andshareholder interest alignment. Academy of Manage-ment Journal, 53(5), 1029–1049.

Perrow, C. (1986). Economic theories of organization.Theory and Society, 15(1), 11–45.

Rajgopal, S., & Shevlin, T. (2002). Empirical evidence onthe relation between stock option compensation andrisk taking. Journal of Accounting and Economics, 33,145–171.

Ross, S. A. (2004). Compensation, incentives, and the du-ality of risk aversion and riskiness. Journal of Finance,59(1), 207–225.

Sanders, W. G. (2001). Behavioral responses of CEOs tostock ownership and stock option pay. Academy ofManagement Journal, 44(3), 477–492.

Sanders, W. G., & Carpenter, M. A. (2003). Strategic sat-isficing? A behavioral-agency theory perspective onstock repurchase program announcements. Academyof Management Journal, 46(2), 160–178.

Sanders, W. G., & Hambrick, D. C. (2007). Swinging for thefences: The effects of CEO stock options on companyrisk-taking and performance. Academy of Manage-ment Journal, 50, 1055–1078.

Shapira, Z. (1995). Risk taking: A managerial perspective.New York: Russell Sage Foundation Publications.

Shavell, S. (1979). On moral hazard and insurance. Quar-terly Journal of Economics, 93(4), 541–562.

Simon,H.A. (1955). Abehavioralmodel of rational choice.Quarterly Journal of Economics, 69, 99–118.

Sitkin, S. B., & Pablo, A. L. (1992). Reconceptualizing thedeterminants of risk behavior. Academy of Manage-ment Review, 17(1), 9–38.

Smith, C. W., & Stulz, R. M. (1985). The determinants offirms’ hedging policies. Journal of Financial andQuantitative Analysis, 20(4), 391–405.

Smith, C. W., & Watts, R. L. (1992). The investment op-portunity set and corporate financing, dividend, and


compensation policies. Journal of Financial Eco-nomics, 32(3), 263–292.

Thaler, R. H. (1980). Toward a positive theory of con-sumer choice. Journal of Economic Behavior andOrganization, 1, 39–60.

Thaler, R. H. (Ed.). (2005).Advances in behavioral finance(Vol. 2). Princeton, NJ: Princeton University Press.

Thaler, R. H., Tversky, A., Kahneman, D., & Schwartz, A.(1997). The effect of myopia and loss aversion on risktaking: An experimental test. Quarterly Journal ofEconomics, 112(2), 647–661.

Tosi, H. L., Katz, J. P., & Gomez-Mejia, L. R. (1997). Dis-aggregating the agency contract: The effects of moni-toring, incentive alignment, and term in office onagent decision making. Academy of ManagementJournal, 40(3), 584–602.

Tosi, H. L., Werner, S., Katz, J. P., & Gomez-Mejia, L. R.(2000). How much does performance matter? A meta-analysis of CEO pay studies. Journal of Management,26(2), 301–339.

Tversky, A., & Kahneman, D. (1992). Advances in prospecttheory: Cumulative representation of uncertainty.Journal of Risk and Uncertainty, 5(4), 297–323.

Tversky, A., Sattath, S., & Slovic, P. (1988). Contingentweighting in judgment and choice. Psychological Re-view, 95, 371–384.

von Neumann, J. H., & Morgenstern, O. (1947). Theory ofgames and economic behavior (2nd ed.). Princeton,NJ: Princeton University Press.

Wade, J., O’Reilly, C. A., III, & Chandratat, I. (1990). Goldenparachutes: CEOs and the exercise of social influence.Administrative Science Quarterly, 35(4), 587–603.

Wang, H., Yan, J., & Yu, J. (2012). Prospect theory and therisk-return tradeoff. Retrieved from http://papers.ssrn.com/sol3/papers.cfm?abstract_id52070910

Weber, E. U., Blais, A. R., & Betz, N. E. (2002). A domain-specific risk-attitude scale: Measuring risk percep-tions and risk behaviors. Journal of BehavioralDecision Making, 15, 263–290.

Werder, A. V. (2011). Corporate governance and stake-holder opportunism. Organization Science, 22(5),1345–1358.

Westphal, J. D., & Zajac, E. J. (2013). A behavioral theory ofcorporate governance: Explicating the mechanisms ofsocially situated and socially constituted agency.Academy of Management Annals, 7(1), 607–661.

Weymark, J. (1981). Undominated directions of tax reform.Journal of Public Economics, 16, 343–369.

Wiseman, R. M., & Bromiley, P. (1996). Toward a model ofrisk in declining organizations: An empirical exami-nation of risk, performance and decline.OrganizationScience, 7(5), 524–543.

Wiseman, R.M., &Gomez-Mejia, L. R. (1998). A behavioralagency model of managerial risk-taking. Academy ofManagement Review, 23, 133–153.

Wowak, A. J., & Hambrick, D. C. (2010). Amodel of person‐pay interaction:Howexecutives vary in their responsesto compensation arrangements. Strategic ManagementJournal, 31(8), 803–821.

Zajac, E. J. (2006). CEO selection, succession, compensa-tion and firm performance: A theoretical integrationand empirical analysis. Strategic Management Jour-nal, 11(3), 217–230.

Zhang, X., Bartol, K. M., Smith, K. G., Pfarrer, M. D., &Khanin, D. M. (2008). CEOs on the edge: Earningsmanipulation and stock-based incentive mis-alignment. Academy of Management Journal,51(2), 241–258.

Zona, F., Gomez-Mejia, L. R., & Withers, M. C. (in press).Board interlocks and firm performance: Towarda combinedagency-resourcedependenceperspective.Journal of Management.

Geoff Martin ([email protected]) is an associate professorof strategy at Melbourne Business School. He has workedas a chartered accountant, in finance roleswith investmentbanks, and as a strategy consultant.He completedhis Ph.D.in management at IE Business School in Madrid, Spain,before joining Melbourne Business School in 2012. Hisresearch interests cover strategic decision making, execu-tive compensation, risk taking, and corporate governance.He has published in the leading practitioner and academicjournals, including Harvard Business Review, Academyof Management Journal, Strategic Management Journal,Entrepreneurship Theory & Practice, Journal of BusinessEthics, and Human Resource Management.

Robert M. Wiseman ([email protected]) currentlyserves as the senior associate dean and is the Eli BroadLegacy Fellow of Management in the Broad College ofBusiness at Michigan State University. Robert earned hisdoctorate degree in strategic management at the CarlsonSchool of Management at the University of Minnesota. Hisresearch focuses on strategic riskmanagement and the roleof risk in internal corporate governance, with particularemphasis on executive compensation. In particular, hisresearch integrates behavioral research on decision mak-ing and risk perception into principal–agent models ofexecutive compensation. His research has been publishedin a variety of scholarly journals, including StrategicManagement Journal, Academy of Management Journal,Academy of Management Review, Organization Science,Journal ofManagement, and Journal of EconomicBehaviorand Organizations.

Luis R. Gomez-Mejia ([email protected]) is aprofessor of management at the W. P. Carey School of


http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2070910



mailto:[email protected]



Business at Arizona State University. He was the Ray andMilan Siegfried Professor of Management at the MendozaCollege of Business, the Benton Cocanougher Chair inBusiness at Texas A&M University, and a faculty mem-ber at Arizona State University, where he was a Councilof 100 Distinguished Scholar, a Regents Professor, andan Arizona Heritage Chair holder. His research interestscover international management, family business, stra-tegic management, and executive compensation, andhis work has appeared in numerous top journals, in-cluding Academy of Management Journal, Academy ofManagement Review, Strategic Management Journal,

andAdministrative Science Quarterly. He was inductedinto the Hall of Fame of the Academy of Management(given to 33 Academy of Management members out ofapproximately 20,000 members) and has received nu-merous awards for his research, including best paper inAcademy of Management Journal, best paper in Ad-ministrative Science Quarterly, and highest impact pa-per from the entrepreneurship division of the Academyof Management.


Documents

Martin et al AMP 2016