The economic psychology of incentives new design principles for executive pay

When it comes to agencytheory, however, we need not dwell on the merits or otherwise ofFriedman’s argument because there is a further problem.The origins of agency theory lie in a series

All rights reserved No reproduction, copy or transmission of this

publication may be made without written permission.

No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6–10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages The author has asserted his right to be identified as the author of this work in accordance with the Copyright, Designs and Patents Act 1988 First published 2015 by

PALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC,

175 Fifth Avenue, New York, NY 10010.

Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world.

Palgrave®and Macmillan®are registered trademarks in the United States, the United Kingdom, Europe and other countries.

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin.

A catalogue record for this book is available from the British Library.

A catalog record for this book is available from the Library of Congress.ISBN 978-1-349-68142-6 ISBN 978-1-137-40925-6 (eBook)DOI 10.1007/978-1-137-40925-6

List of Tables and Figures vi

Appendix 3: Mathematical Proof of the Importance of

2.1 Key themes and exemplary quotes from the

3.1 Assumptions about the nature of man under positive

3.2 Overview of positive agency theory vs behavioural

Figures

3.1 An agent’s job performance and work motivation cycle 48

7.1 Summary and formal restatement of behavioural

vi

The directors of such companies, however, being the managers rather of other people’s money than of their own, it cannot well be expected, that they should watch over it with the same anxious vigilance with which the partners in a private copartnery frequently watch over their own Like the steward of a rich man they are apt to consider attention to small matters as not for their master’s honour, and very easily give themselves

a dispensation from having it Negligence and profusion, therefore, must always prevail, more or less, in the management of the affairs of such a company.

Adam Smith (1776) An inquiry into the nature and causes of the wealth of nations London: Strahan and

Cadell Book V: Chapter 1: Part III Article 1 “Of thePublic Works and Institutions Which Are Necessaryfor Facilitating Particular Branches of Commerce”

The pleasure which we are to enjoy in ten years hence, interests us so little in comparison with that which we enjoy today, the passion which the first excites, is naturally so weak in comparison with that violent emotion which the second is apt to give occasion to, that one could never be any balance to the other, unless it was supported by a sense of propriety.

Adam Smith (1759) The theory of moral sentiments London: Miller; Edinburgh: Kincaid and

Bell Part IV: Chapter 2 “Of the Beauty Whichthe Appearance of Utility Bestows Uponthe Characters and Actions of Men”

Adam Smith identified the agency problem in public corporations

as long ago as 1776 in the Wealth of Nations, the first great

spe-cialist work of economics Yet Adam Smith was also a behavioural

scientist, writing in 1759 in The Theory of Moral Sentiments about

vii

motivation, loss aversion, fairness and inter-temporal choice, amongother things.1 That agency problems in public companies still existtoday, and that the kinds of incentive mechanisms advocated byscholars and implemented by practitioners have not achieved theirobjective of solving these problems, is surely not in dispute: evenJohn Roberts, one of the world’s leading organisational economistsand a well-known agency theorist, has commented that agency the-ory did not perform well during the financial crisis of 2008–10.2

Nevertheless, it would be unwise to insist that the principal–agentmodel is now so fundamentally flawed that it should be rejected inits entirety: agency theory comes from a long and highly reputablescholarly tradition which has, over the last 40 years, influencedmanagement scholars and legal theorists, as well as economists.The objective of this book is to propose a new version of agencytheory which is built on more robust assumptions about humanbehaviour It develops a line of thinking whose origins can be tracedback to an article written by Robert Wiseman and Luiz Gomez-Mejia

which was published in the Academy of Management Review in 1998.3

My hope is that, by extending the theory and providing empiricalevidence in its support, I might encourage other scholars to build onthe behavioural agency model and test it with new data Ultimately,

of course, my goal is to have an effect on management practice and tocontribute new ideas about executive reward that business will sup-port, which might also help to win back the confidence of the public

in public corporations and their managers It would indeed be verydisappointing if “negligence and profusion” were always to prevail inthe management of the affairs of public companies

This monograph brings together in one place the research I havecarried out during the past six years with various collaboratorsand co-authors into what has become known as the “behaviouralaspects of senior executive reward systems”, commencing with mydoctoral thesis submitted under that title in part fulfilment of the

1 Ashraf, N., Camerer, C., & Loewenstein, G (2005) Adam Smith, behavioral

economist Journal of Economic Perspectives, 19(3), 131–145.

2Roberts, J (2010) Designing incentives in organizations Journal of tional Economics, 6, 125–132.

Institu-3 Wiseman, R., & Gomez-Mejia, L (1998) A behavioral agency model of

managerial risk taking Academy of Management Review, 23(1), 133–153.

requirements of the degree of Doctor of Business Administration(DBA), awarded by the University of Surrey in 2011 Various parts ofthis book have previously been published in academic journals andare reproduced here with permission of the co-authors and publishers

of the respective journals

Pepper, A., Gore, J., & Crossman, A (2013) Are long-term tive plans an effective and efficient way of motivating senior

incen-executives? Human Resource Management Journal, 23(1), 36–51.

Reproduced with the permission of John Wiley & Sons

Pepper, A., & Gore, J (2014) The economic psychology of

incen-tives – An international study of top managers Journal of World Business, 49(3), 289–464 Reproduced with the permission of

Elsevier

Pepper, A., & Gore, J (2012) Behavioral agency theory: New

foun-dations for theorizing about executive compensation Journal

of Management, dx.doi.org: 10.1177/0149206312461054 Repro- t

duced with the permission of Sage Publications

Pepper, A., Gosling, T., & Gore, J (2015) Fairness, envy, guilt and

greed: Building equity considerations into agency theory Human Relations, dx.doi.org/:10.1177/0018726714554663 Reproduced

with the permission of Sage Publications

Many people deserve recognition for their contribution to thiswork, foremost the co-authors of the journal articles around whichthe book is based, namely Julie Gore and Alf Crossman at Sur-rey Business School, University of Surrey, and Tom Gosling atPricewaterhouseCoopers in London Their contributions to the orig-inal articles are noted at the beginning of the relevant chapters.Without their help and support this work would not have beenpossible

Numerous others have contributed directly or indirectly to thisbook In addition to my co-authors, these include Jon Terry, SeanO’Hare and Gemma Carr at PricewaterhouseCoopers; Wida Amani,Rebecca Campbell and Lori Peterson at the London School of Eco-

nomics (LSE); Tony Dundon of Human Resource Management Journal; Wayne Cascio of Journal of World Business; Patrick Wright of Journal

of Management; and Mathew Sheep of Human Relations, along with

eight anonymous reviewers

I am especially grateful to Julie Gore of the Surrey Business School,who originally became involved in this work as the supervisor of

my DBA thesis, and who subsequently became my researcher laborator and co-author Julie has also been my academic mentor,and her guidance and support as I went through a career changefrom consultant to academic have been of great importance to me

col-I also owe a great deal to the Foundation for Management Education,and especially its director Mike Jones, who supported my move fromPricewaterhouseCoopers to LSE in 2008, and to Saul Estrin at LSE fortaking a risk and taking me on This book is dedicated to them

Introduction

Agency theory has been a major component of the economic theory

of the firm since the 1970s It has also come to dominate academicthinking about executive reward Agency theory asserts that the inter-ests of shareholders (who, for these purposes, are the principals in theprincipal–agent relationship) and top managers (their agents) are dif-ferent, and at times radically diverge In order to align the interests

of principals and their agents, boards of directors, acting on behalf

of shareholders, create incentive contracts which reward executivesfinancially if shareholders’ returns increase, but not otherwise; or sothe theory goes Agency theory also postulates the active monitor-ing of the actions of top managers by shareholders through a processwhich we now know as corporate governance

The main thesis of this book is that standard agency theory isflawed, so that it is neither a good positive theory of senior executivereward (it does not explain what actually happens) nor a good nor-mative theory (it does not provide helpful guidance as to what shouldhappen) The book explores what happens if the behavioural assump-tions on which agency theory is based are modified and broughtmore closely in line with actual behaviour

Standard agency theory assumes that firms are profit seeking, thatagents are both rational and rent seeking and that there is no non-pecuniary agent motivation An agent’s utility is assumed to bepositively contingent on pecuniary incentives and negatively con-tingent on effort, and effort is assumed to increase monotonicallywith additional reward In practice we know that these things are

at best partial truths: not all firms are profit seeking all of the time;

1

some firms are sometimes focused only on survival; some firms have

a social purpose; some firms recognise a wider group of stakeholdersother than shareholders Human agents are, in the words of HerbertSimon, “boundedly rational” rather than fully rational:1we estimate;

we take mental shortcuts in calculations; we make mistakes; we forgetthings; we use language in ways which are far from clear.2Motivation

is also a far more complex psychological phenomenon than standardagency theory admits We are motivated by many things other thanmoney: by doing a good job; the pleasure of giving; friendship; love;duty; religious belief The relationship between intrinsic motivation,doing something for its own sake, and extrinsic motivation, doingsomething for reward, is not straightforward Some experts believethat in certain circumstances increasing financial rewards “crowdsout” intrinsic motivation.3

Economists have long argued that the lack of realism in some oftheir assumptions does not in itself undermine the scientific valid-ity of their theories Milton Friedman famously wrote that the test of

a good economic theory is its ability to predict future outcomes, notany correspondence between its assumptions and mechanisms on theone hand, and reality on the other.4This principle has become insti-tutionalised in standard neoclassical economic thinking, although itincreasingly seems rather odd: it is hard to imagine a natural scientist

1 Simon (1945/1997).

2 There are many different definitions of “bounded rationality” As well as Herbert Simon, I have been heavily influenced by Oliver Williamson, who explains that rationality is subject to neuro-physiological rate and storage lim- its on the powers of agents to receive, store, retrieve and process information without error (Williamson 1975, p.21) Williamson also talks about a further element of bounded rationality, which he calls “language limits”, being the constraints on individuals to communicate comprehensively in such a way that they are fully understood by others, but this element is not really relevant

to the current article Foss (2010) provides an elegant summary description of bounded rationality, which he describes in terms of limitations in the human capacity to process information and attempts to economise on mental effort

by relying on shortcuts or heuristics, and as a consequence of the fact that cognition and judgement are subject to a wide range of biases and errors.

3 The most notable proponent of this view is Bruno Frey, a Swiss economist, formerly Professor of Economics at the University of Zurich, and most recently Professor of Economics at Zeppelin University, Friedrichshafen, Germany.

4 Friedman (1953/2008).

trying to advance a similar argument When it comes to agencytheory, however, we need not dwell on the merits or otherwise ofFriedman’s argument because there is a further problem.

The origins of agency theory lie in a series of theoretical paperspublished in the 1970s.5In 1990, empirical work by Michael Jensenand Kevin Murphy, both agency theorists, failed to establish a conclu-sive link between CEO pay and stock price performance.6This caused

a number of scholars to argue that companies would perform ter if executives were provided with greater financial incentives, yetthe logical flaw and sleight of hand here should be apparent Theprincipal–agent framework, which had begun as a positive theoryabout executive compensation (i.e., seeking to explain what hap-pened in practice) was being turned instead into a normative theory(i.e., recommending what should happen) when the explanatorypowers of the model were found to be wanting Ten years after Jensenand Murphy’s empirical findings were reported, a meta-analysis ofover a hundred empirical studies concluded that incentive alignment

bet-as an explanatory agency construct for CEO pay wbet-as at best weaklysupported by the evidence.7A subsequent literature review suggeststhat agency theory has apparently fared no better in later empiri-cal research.8We are driven to the conclusion that, when it comes

to executive compensation and behaviours, agency theory is not agood predictor of outcomes, as well as being based on unrealisticassumptions

In recent years Milton Friedman’s principle has been challenged

by a number of behavioural economists and economic psychologists.Many trace the origins of this schools of thought to the publica-tion of a seminal article by Daniel Kahneman and Amos Tversky

in 1979 entitled “Prospect theory: an analysis of decisions underrisk”,9although arguably the roots of behavioural economics can betraced further back to Herbert Simon’s work on bounded rationality,Harvey Leibenstein’s work on x-efficiency and even to the American

5 Spence & Zeckhauser (1971), Alchian and Demsetz (1972), Ross (1973) and Jensen and Meckling (1976).

6 Jensen & Murphy (1990).

7Tosi et al (2000).

8 See Frydman & Jenter (2010).

9 Kahneman & Tversky (1979).

institutional economists writing at the end of the 19th and beginning

of the 20th centuries.10Intriguingly both Kahneman and Simon wonthe Nobel Prize in Economic Sciences for their work on the founda-tions of behavioural economics, even though neither of them wouldhave described themselves as “economists”.11

Behavioural economics is seen by some as a heterodox, evenheretical, activity, which is not worthy of association with standardeconomic science, but this is increasingly an extreme view Manybehavioural economists are trying to integrate their work with themainstream neoclassical tradition, in order to strengthen and enrich

it Matthew Rabin argues that behavioural economics is increasinglybecoming “normal science”, in the sense in which that phrase is used

by Thomas Kuhn in The structure of scientific revolutions.12 That tion is entirely consistent with the thesis of this book, that if thebehavioural assumptions of agency theory are modified and mademore consistent with actual observed behaviour, then a new set ofpredictions can be inferred by deduction These can be tested empir-ically with, it is further believed, a better chance of being shown to

posi-be in accordance with the observable facts The result is, arguably, abetter (behavioural) agency theory, with greater explanatory powerand more valuable applications for management practice

11 Simon won the Nobel Prize in Economic Sciences in 1978 for “pioneering research into the decision-making process within economic organizations” Kahneman’s Nobel Prize was awarded in 2002 “for having integrated insights from psychological research into economic science, especially concerning human judgment and decision-making under uncertainty” Tversky was not eligible for the prize because of his premature death in 1996.

12 Kuhn (1962/1996) Matthew Rabin made the remarks about normal ence at a seminar delivered at the London School of Economics and Political Science in February 2011.

sci-successful long-term incentive plans (LTIPs) were proving to be inmeeting their two primary objectives of aligning the interests ofshareholders and senior executives, and of motivating executives.Criticisms included the assertion that complex designs made LTIPsvery hard to understand (objections by executives and investors),performance targets were perceived to be undemanding (objec-tions by investors) or too demanding (objections by executives),the performance of comparator companies had an undue impact

on performance targets (executives) and the total amounts mately paid out were often perceived to be too high (some investorsand the public generally) One of the paradoxes about LTIPs isthat, self-evidently, all these points of view could not be easilyreconciled

ulti-The research began with an exploratory study, which was described

in an article published in the journal Human Resource Management in

January 2013, and which forms the basis of Chapter 2 This first cle was followed by a more rigorous working-out of the theoreticalimplications of the revised set of assumptions about agent behaviour

arti-in a theoretical paper, which was published arti-in the Journal of agement in 2012, and which forms the basis of Chapter 3 The ideas

Man-in Chapters 2 and 3 also owe much to the thesis which I ted in part fulfilment of the requirements of the degree of Doctor ofBusiness Administration at the University of Surrey in 2010

submit-An aside here, which I hope will explain the motivation behindthis book and the way it has been framed I am a late entrant toacademia Between 1981, when I left university for the first time, and

2008 I worked for PricewaterhouseCoopers (PwC; formerly Coopers &Lybrand) as an accountant, tax adviser and latterly as an HR consul-tant Much of my time was spent advising companies on executivereward strategies and LTIPs Accounting and tax law are essentiallyclosed systems – decisions, and hence advice about decisions, arelargely based on the interpretation of written codes Reward strate-gies, on the other hand, are open systems, and best practices in paydesign are informed by theories drawn from a wide range of academicdisciplines, by empirical work, by prior experience and sometimes byoriginal thinking Given the open-ended nature of the issues I foundmyself facing as an adviser to companies on reward, I turned tothe academic literature for ideas The literature on executive reward

is extensive, and written in many traditions, including economics,social psychology and management I became increasingly fascinated

by this literature, prompting me to go back to university to undertake

a part-time doctorate at Surrey, and eventually, when the opportunityunexpectedly arose, to leave PwC altogether to take up an academicposition at the London School of Economics At the same time as

I made this move, I was also becoming aware that the academic study

of management, and management practice, often proceed in paralleluniverses Unlike medicine and law, management practitioners gen-erally carry out their activities without any recourse to the academicthinking which may be relevant to the decisions they are making.Equally, management academics spend far too much time studyingphenomena that are not central to the day-to-day problems faced bypractising managers The result is a missed set of opportunities onboth sides Hence I decided to make building bridges between thestudy of management in academia and the practice of managementone of the objectives of the latter part of my career

That objective informs this book in a number of ways I believethat more management research should be made relevant and acces-sible to practitioners: too often good ideas are buried in technicalpapers published in journals read only by other academics At thesame time, I have developed a dislike of the type of intellectuallypre-digested practitioner material written by some academics, to befound particularly in bookshops in airport terminals There are lots

of highly intelligent business people, indeed many who might havebeen academics themselves had they chosen a different path earlier

in their careers, who are perfectly capable of understanding complexsocial scientific ideas This book is intended for them, as well as foracademics and management students While I have endeavoured toavoid unnecessary academic language and have included some ofthe more esoteric material in footnotes, equally I have not tried tosimplify the text unduly

To return to the agenda of the book – after completing the workthat underpins Chapters 2 and 3, I was fortunate enough to attractthe interest of some of my former colleagues in the Reward group

at PwC They agreed to support my research programme by ing funding for a research agency, Forbes Intelligence, to issue anelectronic version of my survey instrument, which had been devel-oped for the earlier study, to a panel of international executives,

provid-thus helping to construct a unique and extremely rich dataset Theempirical findings of this larger study were published in 2012 in aPwC report entitled “Making executive pay work – the psychology

of incentives”,13 and subsequently in two other academic journalarticles.14These two articles form the basis of Chapters 4 and 6.Chapter 5, which examines intrinsic and extrinsic motivation, andthe “crowding-out” conjecture, is also based on the research carriedout in conjunction with PwC and Forbes Intelligence, but has notpreviously been published Empirical evidence for the crowding-outconjecture in the context of executive pay falls short of an ade-quate scientific proof, even though the conjecture is intellectuallypersuasive and consistent with other empirical research Chapter 5,therefore, summarises the relevant literature, provides some relevantdata from the study and makes a partial case in support of the con-jecture, but for the time being it must be left to the reader to decide

on the merits of the argument Chapter 6 examines the importance

of social comparisons in the determination of top pay, arguing thatsenior executives have similar concerns about fairness as other work-ers As a result, equity within teams becomes an important factor

in the design of top-management pay systems in contradiction, forexample, to one of the postulates of tournament theory I argue, withthe support of my co-authors, that low-powered team-based incen-tives may actually be more effective than high-powered individuallybased incentives Chapter 7 draws the various threads together andsets out the strengths of the various studies, as well as the overall lim-itations and areas for further investigation Importantly, this chapteralso sums up the practical significance of the research, in particularthe implications for the future design of management incentives

How to read this book

Readers will notice that this book describes what in one sense appears

to be a paradigmatic social science research programme, beginningwith an exploratory study, using this for theory construction, before

13 PwC (2012) “Making executive pay work – the psychology of tives” http://www.pwc.co.uk/human-resource-services/publications/making- executive-pay-work.jhtml

incen-14Pepper & Gore (2014), Pepper et al (2015).

testing the theory empirically in a large-scale quantitative study Thisdescription disguises the fact that the research was not necessarilyplanned as carefully in advance as this description might suggest.

It progressed by evolution as much as by design Such is the nature

of much social scientific research

The book can therefore be read as the story of a social scienceresearch project, starting with a preliminary study, followed by thedevelopment of a full theoretical framework, then a more exten-sive empirical study and so on Alternatively, because a number ofthe chapters are based on articles published previously in academicjournals, most chapters are largely self-standing Thus, it is perfectlypossible for the reader to dip in and out of the book, or to read thechapters in non-sequential order

Key definitions

The argument in the book is built upon a number of important tions, including terms such as “efficiency”, “effectiveness”, “boundedrationality”, “senior executive” and “top-management team” Theseterms are defined at the most appropriate point in the develop-ment of the argument; however, for the reader’s convenience keydefinitions are summarised here in Table 1.1

to economise on mental effort by relying on shortcuts or heuristics; and that cognition and judgement are subject

to a wide range of biases and errors.

Efficiency and

effectiveness

An action, event, plan, policy or programme is

considered to be efficient if it causes inputs to be

minimised for a given set of outputs or outputs maximised for a given set of inputs; on the other hand,

an action, event, plan, policy or programme is

considered to be effective if it is capable of achieving its

intended objectives, what the objectives are being exogenous to the theory.

incentives

Long-term incentives include share-based incentives such as stock options, restricted stock and performance shares, as well as equity-linked cash-based incentives such as phantom options and stock appreciation rights.

In the UK LTIPs typically take the form of an award of deferred shares which vest over a three-year period conditional upon the satisfactory achievement of a number of financial performance targets.

Senior executive The top managers of a firm who through their actions

are capable of affecting the company’s profits, share price, reputation and market positioning.

Top-management

team

The group of very senior executives who are responsible for defining and executing a firm’s strategy This group, which includes the chief executive officer (CEO), the chief operating officer (COO), the chief financial officer (CFO), divisional heads and other heads of function, is sometimes referred to as the “management board”,

“operating board”, “executive committee” or “general management committee”.

A note on referencing

Social scientists typically include extensive references to prior demic literature in the main text of their work as a way of buildingtheir intellectual case While this is an important part of validatingtheoretical claims, it does not always make easy reading In this book,therefore, I have as far as possible consigned references to footnotes

aca-I prefer these to endnotes for ease of accessibility, to avoid the readerhaving to scrabble back and forth between different pages, which washard enough to do even before the advent of e-books

Long-Term Incentive Plans

In 1995 the Greenbury report1 recommended that UK companiesshould adopt performance-related long-term incentive plans (oftenknown simply as “LTIPs”) for senior executives, preferring them totraditional share options The report pointed out that stock optionshad a number of shortcomings: they sometimes led to windfall gainssimply as a result of general movements in share prices and did notencourage directors to build up significant shareholdings in theiremploying companies Reuters Group plc was the first UK listedcompany to adopt the new style of LTIP in 1993 Many other UK com-panies followed suit after 1995, influenced by the Greenbury report

as well as the withdrawal of tax relief for share options granted overshares with a market value in excess of £20,000 in the 1995 budget.Since that time, LTIPs have become a major component of seniorexecutive reward systems in UK listed companies By 2012 long-termincentives comprised nearly 50% of the total earnings of executives

in the FTSE 350, up from just under 40% in 2006.2

While designs vary, in the UK long-term incentives typically takethe form of an award of deferred shares which vest over a three-year

Parts of this chapter were previously published as Pepper, A., Gore, J., & Crossman, A (2013) Are long-term incentive plans an effective and efficient

way of motivating senior executives? Human Resource Management Journal, 23(1), 36–51, and are reproduced here with the permission of John Wiley and

Sons, and of my co-authors.

1Greenbury, R (1995) Director’s remuneration: Report of a study group chaired by Sir Richard Greenbury London: Confederation of British Industry.

2 Based on data obtained from Income Data Services (“The director’s pay report” 2006 and 2013) London: Thomson Reuters.

10

period conditional upon the satisfactory achievement of a number

of financial performance targets These are often relative measures,benchmarked against either an index or the financial performance

of a number of comparator companies, so that the extent to whichawards vest is dependent upon a company’s financial performancerelative to the market

Long-term incentives have two primary objectives: first, to alignthe interests of executives and shareholders in order to minimiseboth agency risk and the associated agency costs (the alignmentobjective); secondly, to recruit, retain and motivate senior exec-utives to maximise their effort and give high performance (themotivation objective) For some years there has been disquiet abouthow successful LTIPs are in meeting these two objectives Criti-cisms by executives, investors or the public generally include: theassertion that complex designs make LTIPs very hard to under-stand; performance targets are perceived to be undemanding, or toodemanding; the performance of comparator companies has an undueimpact on performance targets; and the total amounts ultimatelypaid out are perceived to be too high One of the paradoxes aboutLTIPs is that self-evidently these points of view cannot be easilyreconciled

This chapter examines whether LTIPs are an effective and efficient3

way of motivating senior executives, while at the same time ing other behavioural aspects of senior executive reward systems

explor-It argues that it is short-sighted to focus on the alignment objectivewithout also considering the motivation objective, on the basis thatthe interests of shareholders and executives cannot be aligned if exec-utives are not properly motivated to maximise their effort and givehigh performance It proposes that more attention should be paid tothe motivation objective and inequity aversion by economists andother management theorists

The rest of the chapter is organised as follows It describes a tative research study which was designed, on an essentially inductive

quali-3 For the purposes of this chapter, something is considered to be “efficient”

if it causes inputs to be minimised for a given level of output and “effective”

if it is capable of achieving its intended objectives There is a strong logical connection between the two terms A more detailed explanation is provided

at the start of Chapter 3.

basis, to identify major research themes It introduces the theoreticalframework, focusing on agency theory and work motivation, and setsout three hypotheses The methodology and findings of the study aredescribed and discussed The chapter concludes by considering theimplications of the findings for the development of agency theory as

it applies to senior executive reward

Agency theory

The investigations described in this chapter take agency theory astheir starting point on the basis that, as indicated in Chapter 1, it isthe dominant framework for examining senior executive reward,4but

we examine a number of the underlying behavioural assumptions,specifically those relating to motivation and fairness Agency theoryfocuses on the separation of ownership and control and hence on theimportance of incentive contracts to align the interests of sharehold-ers and managers The underlying assumptions are that organisationsare profit seeking, that agents are both rational and rent seeking, andthat there is no non-pecuniary agent motivation It is assumed that

an agent’s utility is positively contingent on pecuniary incentives andnegatively contingent on effort It is postulated that effort and moti-vation increase monotonically with additional reward The pay–effortfunction is therefore presumed to be a straight line with a positivegradient proceeding from bottom left to top right

The principal–agent model places less emphasis on the objective

of motivating agents (whether extrinsically or intrinsically) than

it does on alignment David Kreps contends that it is not sary to postulate the concept of intrinsic motivation on the basisthat what is called intrinsic motivation may in fact be no morethan a series of vaguely defined extrinsic motivators.5 Besley andGhatak argue that there is such a thing as a “motivated agent” whoseeconomic behaviour is affected by intrinsic motivation, but theirargument is restricted to employees of public sector or non-profit

neces-4 See William Bratton’s article “The academic tournament over executive pensation” (2005) for an analysis of the comparative merits of agency theory and the managerial power hypothesis advanced by Bebchuk, Fried & Walker (2002).

com-5 Kreps (1997).

organisations whose activities coalesce around a “mission”.6Deci andRyan argue that the importance of intrinsic motivation should not beunderestimated.7 They challenge the idea that intrinsic and extrin-sic motivation are either independent or additive, arguing insteadthat contingent monetary rewards might actually cause a reduction

in intrinsic motivation In a similar way, Frey and Jegen postulatethat in some cases extrinsic motivation can “crowd-out” intrinsicmotivation, particularly if monetary incentives are badly designed.8

They argue for a strong form of crowding-out, whereby an increase

in extrinsic reward leads to an overall reduction in total tion A weaker form of crowding-out can alternatively be postulated,whereby the level of total motivation is maintained only if theincrease in extrinsic reward more than compensates for the reduction

motiva-in motiva-intrmotiva-insic motivation Weak crowdmotiva-ing-out is consistent with theeconomic concept of the diminishing marginal utility of increasingwealth

The theory of work motivation most commonly used in tions into the motivational impact of pay and monetary incentives

investiga-is expectancy theory In the 1960s, the American psychologinvestiga-ist VictorVroom turned an economic theory of rational choice (expected utilitytheory) into a psychological theory of motivation (expectancy the-ory).9Steel and König propose a version of expectancy theory called

“temporal motivation theory”, which combines expectancy theorywith hyperbolic discounting and prospect theory.10Temporal motiva-tion theory postulates that the motivation of a person to carry out act

i is the product of his expectancy that act i will lead to outcome k, via

j, and the value which he attaches to k, discounted for any time delaybetween the occurrence of act i and outcome k Therefore, motivation

6 Besley & Ghatak (2005).

7 Deci & Ryan (1985).

8 Frey & Jegen (2001).

9 Vroom proposes that motivation is a function of the subjectively perceived probability that an agent’s actions will lead to a particular outcome, and the value which the agent places on that outcome (Vroom 1964).

10 Temporal motivation theory is put forward by Steel & König (2006) bolic discounting was first proposed by Ainslie (1991) and Ainslie & Haslam (1992) Kahneman & Tversky’s article: “Prospect theory – an analysis of decision under risk” was published in 1979.

Hyper-can be understood in terms of expectancy and value, weakened bydelay, with differences for gains and perceived losses.

Based on this theoretical analysis two hypotheses are advanced:Hypothesis 1: Long-term incentives are systematically under-valued by senior executives because of the way that risk, valueand probability are subjectively assessed, the way that the value offuture reward is discounted, and as a result of cognitive responses

by comparing their own situations with other referents Referentsmay be internal (peers, immediate subordinates, immediate superi-ors) or external (people doing equivalent jobs in other organisations)

If people feel that their inputs are fairly and adequately rewarded

by outputs, the equity benchmark being subjectively perceived frommarket norms and other reference points, then they will be happy

in their work and motivated to keep contributing at the same (or

a higher) level However, if the relationship between inputs and puts is not proportionate, then the individual will become dissatisfied

out-11 Shafir, Diamond & Tversky (1997).

12 Akerlof (1982) and Akerlof & Yellen (1990).

13 Adams (1965).

and hence demotivated Frank Michelman translates these ena into economic terms, calling them “demoralisation costs”.14Fehrand Schmidt call fairness “inequity aversion”.15 Gomez-Mejia andWiseman argue that inequity aversion applies equally to senior exec-utives as to other workers.16 Drawing on these theories, the thirdhypothesis is therefore advanced:

phenom-Hypothesis 3: Below a lower threshold level of earnings, inequityaversion resulting from social comparisons of total rewards rel-ative to peers negatively impacts on motivation and leads todemoralisation costs

Empirical research

Paraphrasing Truman Bewley, this inquiry was intended to beexploratory, touching on many issues in order to test existing the-ories, to seek new hypotheses and to see the overall shape of thephenomena.17

The preliminary study comprised a qualitative study of 15 seniorexecutives from companies in the FTSE 350 using semi-structuredinterviews Participants in the study included four CEOs, three exec-utive directors, one other senior executive and seven non-executivedirectors, representing 14 different companies drawn from sevenmajor industry sectors Ages ranged from 40 to 69 with a median age

of 53 Thirteen of the participants were male and two were female.18

The participants were identified through the researcher’s professionalnetwork, a form of convenience sampling Data saturation, the point

at which nothing essentially new is learnt by conducting furtherinterviews, was largely achieved by the end of the 12th interview.19

14 Michelman (1967), p.1214.

15 Fehr & Schmidt (1999), p.819.

16 Gomez-Mejia & Wiseman (1997).

17 Bewley (1999), p.16 Truman Bewley is a neoclassical economist who was brave enough to use a qualitative, inductive, research methodology to inves- tigate a subject, wage rigidity, which had proved to be hard to explain using the standard, largely deductive, method of enquiry preferred by economists.

18 The fact that the majority of the participants in the two studies were male reflects the lack of gender diversity in the population of company directors

generally; see Sealy et al (2009).

19This is consistent with the findings of Guest et al (2006).

Data was gathered in a series of semi-structured interviews using

a proforma interview guide A thematic grid was used to develop alist of interview topics based on early work on the literature review

A semi-structured interview approach was preferred to a structuredquestionnaire in order to ensure an appropriate degree of consis-tency, while at the same time retaining enough flexibility to allowparticipants to express their views in full The data were collectedduring in-depth discussions of around one hour in length All inter-views were recorded and full transcripts were prepared using an exter-nal transcription agency In each case confidentiality was assured

In total the transcripts ran to approximately 100,000 words, resenting nearly 17 hours of interview time The transcripts wereanalysed in depth using a form of textual analysis known as tem-plate analysis.20 The interview transcripts were read in detail andall apparently significant phrases highlighted and numbered A tem-plate was then developed, based on the thematic grid and interviewguide, combined with an initial impression of issues arising out ofthe transcripts All significant phrases were coded against the head-ings appearing on the template To some extent this was an iterativeprocess: the template was amended a number of times as new issuesemerged from a deeper reading of the transcripts The templaterequired responses to be categorised and ranked The results (tem-plate headings, answer categories, individual transcript codes andexemplary quotes) were collected in a spreadsheet and summarised

rep-in a table

Financial incentives

The majority of participants regarded financial incentives as tant, but not necessarily very important, to business success Of thetwo participants in the study who rated financial incentives as veryimportant, one, an executive director and evidently by inclination

impor-an entrepreneur, had joined his compimpor-any during its start-up phaseand had helped to grow the business up to and beyond the point

of flotation on the London Stock Exchange The other, a executive director, was on the board of a company which had beenthrough a major turn-around, during which time executives had been

non-20 King (2004).

incentivised with a high-profile private-equity style incentive plan.

In other cases the prevailing view was that most executives are driven

by a sense of achievement, of being part of a successful ment team, of working in a place where they are in tune with theorganisation’s values and objectives, and of building a great company,summarised in the words of one participant as “winning” According

manage-to this majority view, only a small number of executives are primarilymotivated by potential monetary gain, perhaps no more than 10% or20% according to one HR director

Nevertheless, financial incentives clearly do matter Executiveswanted to be valued, to be treated equitably or, as a number ofthem put it, “fairly” Financial incentives are, according to one non-executive, “a necessary but not sufficient condition for motivatingexecutives” As an HR director explained: “the behaviour of the vastmajority of people – including senior executives – can be influenced

by financial incentives” Another CEO said that intrinsic factors, likeachievement, teamwork, status and power, are fundamentally impor-tant but only come into play once you are at or above a minimumthreshold for financial reward

Financial incentives serve a number of purposes: in particular, toprovide opportunities for creating wealth, as a retention mechanism

to discourage executives from looking for employment elsewhere (or

at least to increase their transfer price and thus to deter other nies from targeting them), to strengthen engagement and encouragesustained performance, and as a means of “keeping score” The last ofthese appeared to be especially important in the case of CEOs Chiefexecutives, competitive by nature, want to know how they are doingrelative to their peers Remuneration is an obvious way of measuringthis, as a proxy for wider measures of success Only two intervieweesmentioned the importance of aligning the interests of shareholdersand executives, even though this is the primary reason for long-termincentives according to principal–agent theory In contrast, the use

compa-of LTIPs as a retention mechanism was mentioned most frequently.Short-term incentives (annual bonuses) were generally regarded asvery effective by executives and non-executives alike Participantsdescribed them, in comparison with long-term incentives, as havingmuch better “line of sight”, meaning that the connection betweensuccessful actions and reward is more obvious LTIPs, on the otherhand, were generally seen as at best only partially effective: indeed,

many of the executives in our study felt that LTIPs failed to meet theirmain objectives Various reasons were given for this Commonly citedwas the complexity of most LTIPs One CEO put it rather elegantly:Deferred share schemes are basically somewhat poorly under-stood, and pretty arbitrary In the old days share options wereeasily understood, but pretty arbitrary These new schemes areextraordinarily complex and still pretty arbitrary That’s the

issue

The same CEO described how a divisional finance director had optednot to join an LTIP because he had miscalculated the possible bene-fits, yet had still managed to influence another executive in his deci-sion to sign up to the plan, because his colleague misunderstood theadvice the finance director was giving him A non-executive placedthe onus on boards of directors and HR departments to communicatethe value of LTIPs in terms that executives can understand

A specific problem which participants identified with LTIPs is theuse of comparative performance measures, such as relative totalshareholder return (TSR) As one CEO said: “I don’t know how tomanage relative TSR you don’t wake up in the morning trying to

manage something relative” With comparative performance targetsthe choice of benchmark companies becomes critical An unusuallygood or bad profit or share price performance by another companycan have a disproportionate effect on the basket of comparator com-panies, especially when no payments are made for below medianperformance Takeovers of companies in the comparator group can

be particularly distorting This is the precise opposite of the “line ofsight” argument for short-term incentives: in the case of LTIPs, execu-tives frequently cannot see any causal link between their actions andreward outcomes

The challenge is that investors are driven by relative measures.They are selecting stocks based on relative performance by categoryand are worried about beating the average in the shape of an index.However, an HR director pointed out that the starting positions ofmanagers and investors are not the same: “Most shareholders hold aportfolio and are therefore insulated against the capricious nature ofshareholder returns We as executives are not” Another participant

in the study said: “Investors shouldn’t inflict relative performance

conditions on companies They should say, ‘well that’s our challenge

to manage’ ”

The strong consensus among the executives who were interviewedwas that using absolute performance conditions, designed carefullyand linked to each company’s particular strategic objectives, couldsignificantly enhance the motivational effect of LTIPs The mostappropriate financial metric to use, such as TSR, earnings per share

or earnings before interest and tax, would vary from company tocompany, but in every case the merit of having an absolute measuretrumps relative metrics

Participants in the study cited a number of other problems withLTIPs In particular one participant talked about the insistence ofthe Association of British Insurers, a trade association representinglarge institutional shareholders, that no LTIP payment should bemade unless performance was at or above the median level, which

he referred to as “the tyranny of the median” For reasonably soliddefence stocks which are, as another executive put it, “incremen-tally creating value through incremental good decision-making overtime”, this may result in no LTIP payments The way LTIPs are oftenconfigured appears to favour volatile stocks, where a large amount ofvalue is created in one performance period even if it is lost again inthe next period

The effect of non-paying LTIPs is not merely neutral – it can bepositively demotivating to hold an incentive instrument which youbelieve will never pay out An HR director with particular experience

of this problem described it in the following way: “If you get rewardwrong it is a much bigger de-motivator than it can ever be a moti-vator It’s like walking around a china shop with a sledgehammer inyour hands”

Motivation

The relationship between intrinsic and extrinsic motivation voked some discussion The prevailing view among participants inthe study was that, for senior executives, certain intrinsic factors,especially an orientation towards achievement, are important pri-mary sources of behaviour Power-status and intimacy-teamworkwere also mentioned as significant factors affecting the way peoplebehave In general, however, intrinsic needs or drives were not seen

pro-as substitutes for extrinsic rewards: a substantial minimum level ofremuneration must be provided One CEO put it like this:

Once you are at a threshold level on the financial structures, alevel which is felt to be fair and appropriate to the market, then[intrinsic factors] become really important but if you are at a

significant discount on the monetary part then the other thingswill not make up for it

A number of non-executives commented that very large awardsshould not be necessary to engage and motivate executives Onecompany chairman, commenting specifically on the US market, said:

“I do not believe, nor have I ever observed, that $100 million vates people more than $10 million, indeed more than $1 million”

moti-In practice, intrinsic and extrinsic rewards are evidently closely twined The relationship between the two is complex and hard tounravel As well as providing material benefits, extrinsic rewards arealso important sources of information for executives, signals whichexecutives can use to measure their value relative to their peers, howhighly they are valued by their company boards, and even in somecases their self-worth

inter-Fairness

A significant number of interviewees talked, on an unpromptedbasis, about “fairness” For most of the participants in the studyfairness was primarily a relative concept: as equity theory predicts,one way in which rewards are evaluated is by drawing comparisonswith other people Who these referent persons were was not alwaysclear Executives talked generally about “peers” One CEO referred tosecond-best options: “fairness is relative to other things I might do asopposed to other organisations” Only one participant, also a CEO,thought fairness was a wholly irrelevant concept in the context ofexecutive pay

Summary of the results of the study

Evidence from the study supports the proposition that senior utives systematically undervalue long-term incentives The principalshortcomings of LTIPs which were identified by participants were as

exec-follows First, complexity – you cannot be effectively motivated bysomething which is too complicated to understand; in particular, inthe specific case of relative performance metrics, too much is outsidethe control of executives and for many companies it is difficult topick a fully appropriate group of comparator companies anyway Sec-ond, the tyranny of the median – the fact that there is typically nopayout at all for average performance creates the risk of a “feast orfamine” incentive, executives employed by companies with volatileearnings and share prices sometimes fare better than executives work-ing for companies which are steady performers Third, participantsrecognised the significance of subjective valuation issues, includingtemporal discounting.

One of the ways in which financial incentives are important is thatthey provide a mechanism for “keeping score”, allowing a seniorexecutive to assess how he or she is doing relative to their peersand signalling how they are regarded by their principals The direct-ness of the link between effort, performance and reward was alsoremarked upon, encapsulated in the phrase “line of sight” This cor-roborates the significance of instrumentality, whether an individualcan see a link between effort and performance, one of the principles

of expectancy theory A critical issue here was relative performanceconditions, where the vesting of awards depended not only on thefinancial performance of the executive’s own company (within theexecutive’s line of sight), but also on the relative performance ofcomparator companies (outside the executive’s line of sight).The executives also recognised the existence of a trade-off betweenintrinsic and extrinsic motivational factors This was captured inthe statement made by one of the participants in the study that

a financial incentive is: “a necessary but not sufficient conditionfor motivating a senior executive” Once above a threshold level

of earnings other factors, including status, power and the need forachievement, assume greater importance

The final issue related to social comparisons A notable feature

of the study was the number of executives who talked about theimportance of “fairness” Social comparison is evidently an impor-tant driver of human behaviour across the whole spectrum of society,regardless of income or wealth.21

21 See Tyson & Bournois (2005).

The results of the study are summarised in Table 2.1 Four majorthemes are identified First, the financial cost of an LTIP may begreater than the value perceived by executives because of the waypeople subjectively assess risk, discount future events and estimatevalue Second, the complexity of many LTIPs means that they areoften poorly understood by executives, which impacts upon the per-ception of their value: a person cannot be effectively motivated bysomething which is too complicated to be readily understood Third,the relationship between intrinsic and extrinsic motivation is neitherlinear nor orthogonal: while financial incentives are necessary theyare not sufficient for motivating senior executives; above an upperthreshold level of earnings extrinsic rewards may crowd out intrin-sic motivation; below a lower threshold intrinsic motivation may

Themes Definition Exemplary quotes

“LTIPS are an amount of money with a very high discount attached to it.”

“I think it is inevitable that people attach a lower discount

“The complexity of most deferred share schemes means that they are basically somewhat poorly understood.”

“The direct motivation is not there on a day-to-day basis because of complexity.”

“Relative TSR is meaningless because there is

no line of sight.”

of earnings extrinsic rewards may “crowd out” intrinsic motivation Below a lower threshold intrinsic motivation may be affected by

“demoralisation costs”.

“There are a small number of people who are only motivated

by the monetary gain, maybe 20%.”

“Once you’re above a threshold level on the financial

structures then other stuff

[becomes] really important.”

“The role of money is as a way

of keeping the score.”

“If the amounts are large enough they can make one lose sight of the intrinsic.”

“It seems as if there is a law of diminishing returns.”

“Internal relativity [is] a big issue.”

“The only way I really think about compensation is ‘do I feel fairly compensated relative to

my peers?’ ”

“I believe this is true especially amongst corporate executives who appear to be very sensitive

to differentials with perceived peers.”

“This is definitely true in my experience as an HR director.”

be affected by demoralisation costs Fourth, social comparisons arecritically important: one way in which rewards are evaluated by indi-viduals is by drawing comparisons with the rewards of other people

Conclusions

The preliminary exploratory research described in this chapter gests that the way senior executives assess probabilities and value issignificantly affected by behavioural factors It challenges the ratio-nal agent assumption which lies at the heart of the principal–agent

sug-model The results are consistent with the findings of an earlier studywhich called into doubt the effectiveness of LTIPs and the agencymodel even though the previous research was conducted largelywithin a conventional microeconomic framework.22 It raises impor-tant questions about how effective, or at least how efficient, LTIPs are

as a way of motivating senior executives

More generally, the study found evidence that as extrinsic rewardsincrease over and above an upper threshold level there is a negativeimpact on intrinsic motivation Conversely, below a lower thresholdlevel, dissatisfaction with extrinsic rewards caused by unfavourablepeer comparisons can negatively impact on intrinsic motivation.These results challenge a second assumption of agency theory, thatthere is no non-pecuniary agent motivation This is consistent withthe positions taken by institutional and behavioural economists such

as Herbert Simon, Harvey Leibenstein, Oliver Williamson, and morerecently Richard Thaler and Dan Ariely, who argue that the set ofmodel triggers for economic action should be extended to includemotivations other than rent-seeking.23

The results of the study appear to be consistent with the threehypotheses Hypothesis 1 is supported by the responses to the ques-tions regarding risk, time and uncertainty; Hypothesis 2 is consistentwith the answers to the questions about intrinsic motivation; andHypothesis 3 is supported by the responses to the questions regardingfairness

A significant theoretical conclusion which can be drawn is thatagency theory, assuming as it does rational, rent-seeking executivesand no-non pecuniary agent motivation, does not, prima facie, inits current form provide a sound basis for modelling senior execu-tive reward A re-theorising of agency theory is therefore proposed,building on the behavioural agency model.24This must:

• avoid the assumption of no non-pecuniary agent motivation andrecognise instead the role of intrinsic motivation;

22 See Buck, Bruce, Main & Udueni (2003).

23 See Simon (1945/1997), Leibenstein (1966), Williamson (1975), Thaler (1991) and Ariely (2008).

24 This was first proposed by Wiseman & Gomez-Mejia (1998).

• take into account the importance of both the motivation andalignment objectives and the interrelationship between them;

• postulate a non-linear pay–effort function which tails off above anupper earnings threshold (because of crowding-out) and below alower earnings threshold (because of demoralisation costs);

• model more realistically the way that agents evaluate non-cashincentives, especially where payment is deferred for a number ofyears;

• recognise the significant role which inequity aversion plays indetermining the motivational impact of earnings

These ideas are developed in Chapter 3

Behavioural Agency Theory

In Chapter 1 it was explained that agency theory has been thedominant theoretical framework for academic research on execu-tive compensation since the mid-1970s Agency theory is one of

a number of theoretical approaches that have been taken by demics in trying to explain executive pay The literature on seniorexecutive reward is now very extensive, drawing on a variety ofscholarly traditions, including economics, law, organisation stud-ies, accounting and finance In addition to the agency approach,theoretical frameworks include tournament theory, human capitaltheory, the managerial power hypothesis, institutional theory, polit-ical theories and theories about fairness.1 There have been a num-ber of extended literature reviews and comprehensive summaries.2

aca-Behavioural research is a relative new feature of this literature.That agency theory has shortcomings has been apparent for sometime Most notably, given Michael Jensen’s role as a leading agencytheorist, empirical work carried out by Jensen and Murphy in 1990failed to establish a conclusive link between CEO pay and stock price

Parts of this chapter were previously published as Pepper, A., & Gore, J (2012) Behavioral agency theory: New foundations for theo-

rising about executive compensation Journal of Management, dx.doi.org: t

10.1177/0149206312461054, and are reproduced here with the permission of Sage Publications, and of my co-authors.

1See Lazear & Rosen (1981), Combs & Skills (2003), Bebchuk et al (2002), Balkin (2008), Ungson & Steers (1984) and Wade et al (2006).

2See Gomez, Mejia & Wiseman (1997), Devers et al (2007) and Gomez-Mejia

et al (2010): 117–140.

26

performance.3 Ten years later, in a meta-analysis of 137 empiricalstudies, another group of researchers similarly found that incentivealignment as an explanatory agency construct for CEO pay was atbest weakly supported by the evidence.4In 2010, Frydman and Saksargued, based on a review of US executive compensation data cover-ing the period 1936–2005, that neither optimal contracting (agencytheory) nor the managerial power hypothesis is fully consistentwith the available evidence John Roberts, another agency theorist,has commented that agency theory performed poorly during thefinancial crisis and has reported various situations where strongincentives are evidently not optimal, as agency theory implies.5Theseinclude when good measures of an agent’s effort or performance arenot available, when multi-tasking is required and when cooperationbetween different agents is necessary, all common situations wheretop management teams are concerned Roberts puts forward argu-

ments in favour of implementing weak rather than strong incentives

in such circumstances

This chapter proposes a new version of agency theory which vides a better explanation of the connection between executive com-pensation, agent performance, firm performance and the interests ofshareholders I call this “behavioural agency theory”, developing aline of argument first advanced by Wiseman and Gomez-Mejia in

pro-1998.6 They proposed that the normal risk assumptions of agencytheory should be varied to incorporate ideas from prospect theory.7

3 Some commentators (e.g., Finkelstein et al 2009) imply that Jensen and Murphy’s empirical evidence is not contrary to agency theory, but suggests instead that it means the (normative) recommendations of agency theory have not been followed in practice An argument in this form, implying that the absence of two factors (incentive pay and high performance) can be interpreted as evidence of a causal connection between the two phenomena (so that more of the first factor will necessarily lead to more of the second)

is hardly justified It also appears to confuse the positive theory of agency (which should be capable of explaining the world as it is) with normative theory In practice, as argued long ago by Herbert Simon in 1957 and demon- strated empirically by Gabaix & Landier in 2008, CEO pay is much more closely correlated with company size than company performance.

4Tosi et al (2000).

5 Roberts (2010).

6 Wiseman & Gomez-Mejia (1998).

7 Kahneman & Tversky (1979) and Tversky & Kahneman (1992).

Sanders and Carpenter8subsequently adopted a behavioural agencyperspective in their examination of stock repurchase programmesand a summary of the literature using the behavioural agencyframework is provided by Finkelstein, Hambrick and Cannella.9

Rebitzer and Taylor10 provide a general examination of behaviouralapproaches to agency and labour markets in the fourth edition ofAshenfelter and Card’s influential handbook on labour economics.However, a settled theory and agreed terminology for the behaviouralagency model does not yet exist In contrast to the standard agencyframework, which focuses on monitoring costs and incentive align-ment, behavioural agency theory places agent performance and workmotivation at the centre of the agency model, arguing that the inter-ests of shareholders and their agents are most likely to be aligned

if executives are motivated to perform to the best of their abilities,given the available opportunities Behavioural agency theory builds

on four constructs which have been identified as key factors affectingbehaviour by behavioural economists and economic psychologists.These are:

• loss aversion and reference dependence;

• preferences relating to risky and uncertain outcomes;

• temporal discounting;

• fairness and inequity aversion

It incorporates a theory (crowding-out) relating to the trade-offbetween intrinsic and extrinsic motivation It also introduces goal-setting theory to the agency model, on the basis that it represents apragmatic way of contracting between principal and agent.11

This chapter proceeds as follows: it begins by describing agencytheory’s main elements and underlying assumptions, before review-ing the limitations of positive agency theory as an explanation of therelationship between senior executives and shareholders, and recon-ceptualising what is meant by economic man (i.e., homo economicus

8 Sanders & Carpenter (2003).

9Finkelstein et al (2009).

10 Rebitzer & Taylor (2011).

11See Camerer et al (2004), Frey & Jegen (2001), Sliwka (2007) and Locke &

Latham (1984, 1990).

of neoclassical economics) It continues with an explanation of thebehavioural agency model, describing the main component systemsand commenting in some detail on the significance of motivation,risk, time discounting, inequity aversion and goal setting It exam-ines the relationship between job performance and firm performance,discusses ways in which behavioural agency theory departs from stan-dard agency theory, and considers the implications of behaviouralagency theory for compensation design, before concluding.

Positive agency theory

Positive agency theory, the standard model of agency which I sider in this chapter, has been extensively used as a basis for theo-retical and empirical work by management scholars and organisationtheorists, as well as being widely applied in examining research ques-tions relating to executive compensation.12 It argues that the firm

con-is a special case of the theory of agency, that a firm provides anexus for a complex set of contracts, both written and unwritten,between various parties, and that agency costs are generated as aresult of the different interests and contractual arrangements betweenowners and top managers.13 The underlying assumptions are thatorganisations are profit seeking, that agents are both rational andrent seeking, and that there is no non-pecuniary agent motivation.14

It is further assumed that principals are risk neutral, because theycan balance their portfolios, that agents are risk averse, because the

12 Jensen (1983) identifies two different strands in the literature on agency ory He calls these the “positive theory of agency” and the “principal–agent” literature Eisenhardt (1989) describes the latter as a “general theory of the principal–agent relationship”, while Wiseman and Gomez-Mejia (1998) call it

the-“normative agency theory” Positive agency theory focuses on the special case

of the principal–agent relationship between owners and managers of large porations (Jensen 1983, Eisenhardt 1989, Charreaux 2002) Normative agency theory aims to provide a formal theory of the principal–agent relationship

cor-in all its guises, cor-includcor-ing employer–employee, lawyer–client, buyer–supplier, etc (Eisenhardt 1989).

13 See Alchian & Demsetz (1972), Jensen (1983) and Jensen & Meckling (1976).

14 Eisenhardt (1989) states that positive agency theory also assumes bounded rationality, but I can find no other reference to this in the agency theory lit- erature After the first, formative papers on agency theory, Jensen & Meckling (1994) later develop the resourceful, evaluative, maximising model of man,