Executive Pay as a Collective Action Problem

Abstract

This chapter takes as its starting point Mancur Olson’s assertion in The Logic of Collective Action that his theory of group size and group behaviour has implications for the governance of companies. It explains why shareholders of public corporations are unlikely to solve executive pay problems because of a collective action problem, and how ideas about the governance of common pool resources have implications for the design of corporate governance mechanisms. A study of the FTSE 100 is used to illustrate the points raised.

Introduction

The UK is widely regarded as having one of the most robust company law and corporate governance regimes in the world,¹ and it was one of the first countries to introduce “say on pay” provisions for shareholders. Yet investors are often reluctant to vote against executive pay proposals—collective action problems mean it is difficult to bring together a shareholder alliance sufficient to reject a directors’ remuneration report. Government and the press urge boards to take more responsibility for moderating pay claims, yet non-executive directors may believe that the costs of challenging claims outweigh the benefits; they may wish to avoid conflict over the pay of executives who are fellow directors on unitary boards, and may feel compromised because they are or were themselves at one time executives of other large companies.

In his book, The Logic of Collective Action, Mancur Olson argues that his theory of group size and group behaviour has implications for the governance of companies. He writes,

The autonomy of management in the large modern corporation, with thousands of stockholders, and the subordination of management in the corporation owned by a small number of stockholders, may also illustrate the special difficulties of the large group. The fact that management tends to control the large corporation and is able, on occasion, to further its own interest at the expense of the stockholders, is surprising, since the common stockholders have the legal power to discharge the management at their pleasure, and since they have, as a group, also an incentive to do so, if the management is running the corporation partly or wholly in the interest of managers. Why, then, do not the stockholders exercise their power? They do not because, in a large corporation, with thousands of stockholders, any effort the typical stockholder makes to oust the management will probably be unsuccessful; and even if the stockholder should be successful, most of the returns in the form of higher dividends and stock prices will go to the rest of the stockholders, since the typical stockholder owns only a trifling percentage of outstanding stock. The income of the corporation is a collective good to the stockholders, and the stockholder who holds only a minute percentage of the total stock, like any member of a latent group, has no incentives to work in the group interest. Specifically, he has no incentive to challenge the management of the company.²

In this chapter I shall describe how Olson’s theory of groups and organisations provides insights that can be used to enhance the standard model and explains how excessive executive pay can be modelled as a collective action problem. In doing so I retain, for the time being, the standard economic assumptions of profit-seeking corporations, rational, rent-seeking principals and agents, and no non-pecuniary agent motivation. A study of public companies that make up the FTSE 100 index in the UK is used to help build the theory. For the time being, I also work within the parameters of the standard model, which assumes that shareholders have the rights to all residual profits, that is, those calculated after deducting all factor inputs. The chapter concludes by proposing various ways in which the standard model can be improved upon, including comments on how stakeholders other than shareholders might share in the common pool, building on the ideas put forward at the end of the previous chapter.

Modelling Agency Costs as a Collective Action Problem

Olson argues that the autonomy of managers in large modern corporations is a specific example of his general theory of groups and organisations. The fact that executives exercise management and control over large corporations and are able, on occasions, to further their own interests at the expense of shareholders might be recognised as a collective action problem.³ There is a sense in which the earnings of a corporation are a collective good to stockholders, so that a shareholder owning a small percentage of total stock is like any member of what Olson calls a “latent group”,⁴ with no incentive to challenge the management of the company as the costs in doing so are likely to outweigh the potential benefits. Russell Hardin points out that collective action problems often relate to the elimination of a cost, which constitutes a good to those who would otherwise bear that cost.⁵ Joseph Heath describes a large corporation as a “quasi-public good” to its members. They all derive benefits from the corporation, but individual self-interested action will not secure those benefits. In order to produce and sustain these quasi-public goods, as Heath says: “it is necessary to overcome a complex set of collective action problems.”⁶

Olson provides a typology of groups, which he describes as being “privileged”, “intermediate”, or “latent”. In a privileged group the benefits of action are likely to exceed the cost for at least some of members of the group so that, other things being equal, collective action is likely to succeed. In a latent group the cost of action is likely to exceed the benefits for all group members, so that, other things being equal, the action is likely to fail. Small groups are typically privileged; large groups are typically latent; intermediate groups may behave like privileged or latent groups depending on whether coordination, benefits-sharing, and cost-sharing are or are not possible in practice. Olson offers no numerical guidance as to what constitutes small, intermediate, or large group sizes. His main conclusion, that latent groups will fail, is modified in certain circumstances. According to his “by-product” theory, groups may selectively offer private goods on favourable terms to members who agree to combine in collective action. Olson postulates that this may explain why labour unions offer healthcare, insurance, and other financial services on exclusive terms to members. According to Olson’s “special interests” theory, large groups will sometimes form themselves into smaller special interest groups that are small enough to negotiate collective action arrangements among themselves. For example, business communities are typically divided into a series of industries, often containing only a small number of separate firms, in markets that tend towards oligopolistic competition. Trade associations are frequently established to represent collective industry interests in such a way that anti-trust considerations are not breached.⁷

The concepts of privileged, intermediate, and latent groups, and the by-product and special interest theories, can be used to model agency costs arising in a public corporation as a collective action problem. Formally, let E be the set of members of the executive committee of firm F, comprising n individuals indexed by e ∈ [1,…,n]; let D be the set of members of the board of directors of F, comprising n individuals indexed by d ∈ [1,…,n]; let S be the set of shareholders of F, comprising n individuals, funds, and companies, indexed by s ∈ [1,…,n]; and let x = x₁,…,x_n denote the vector of financial and non-financial benefits receivable by e, d, and s in respect of their involvement with F. The utility function of executive e ∈ [1,…,n] is given by

$$ {\mathrm{U}}_{\mathrm{e}}\left(\mathrm{x}\right)={\mathrm{w}}_{\mathrm{e}}-{\varepsilon}_{\mathrm{e}} $$

(4.1)

where w_e is the executive’s financial and non-financial reward received for working for F and ε_e is the executive’s cost in terms of effort. The utility function of shareholder s ∈ [1,…,n] is given by

$$ {\mathrm{U}}_{\mathrm{s}}\left(\mathrm{x}\right)={\mathrm{y}}_{\mathrm{s}}-{\mathrm{g}}_{\mathrm{s}} $$

(4.2)

where y_s is the financial return which s receives from F and g_s is the governance cost which s incurs in respect of the investment in F. The return y_s represents an individual stockholder’s proportionate share of the total profits Y of F. In the case of minority shareholders , y_s is delivered in the form of dividends and capital gains. Y is calculated after deducting governance costs borne by F as well any rents (i.e., excessive executive compensation) paid to executives. Executive rents are given by

$$ \mathrm{R}=\sum \limits_{\mathrm{e}=1}^{\mathrm{n}}\left({\mathrm{w}}_{\mathrm{e}}-{\mathrm{w}}_{\mathrm{e}}^{\ast}\right) $$

(4.3)

where ${\mathrm{w}}_{\mathrm{e}}^{\ast }$ represents the financial and non-financial rewards which would be payable to its senior executives by F in the absence of any agency problems.

Olson says that the income of a corporation is a collective good for stockholders. A corollary of this is that agency costs relating to F represent a potential asset to S to which, in Olson’s terminology, the logic of collective action applies. Executive rents represent a collective good to shareholders because a reduction in R would result in an increase in Y. To avoid confusion over signs I define this as |R|, representing executive rents embedded in F which are potentially recoverable. The proportionate share of |R| due to s is represented by p_s. Eq. (4.2) can therefore be rewritten:

$$ {\mathrm{U}}_{\mathrm{s}}\left(\mathrm{x}\right)={\mathrm{y}}_{\mathrm{s}}-{\mathrm{g}}_{\mathrm{s}}+{\mathrm{p}}_{\mathrm{s}}\mid \mathrm{R}\mid $$

(4.4)

The proportion p_s is calculated by dividing the number of shares held by s by the total number of equivalent shares issued by

$$ {\mathrm{F,\,so\,that}}\,{\mathrm{p}}_{\mathrm{s}}=\raisebox{1ex}{${\mathrm{s}}_\mathrm{n}$}\!\left/ \!\raisebox{-1ex}{${\mathrm{S}}_\mathrm{N}$}\right. $$

According to Olson’s logic, if S_N is large, then it is less likely that |R| will be recoverable, because it is more likely that governance costs will exceed the proportion of agency costs recoverable by any one shareholder, that is, formally, g_s > p_s|R|. S will be, in Olson’s terms, a “latent group”. A latent group is distinguished by the fact that, unless one member takes the lead in providing the collective good, no other member will be significantly affected and therefore have any reason to act. Nevertheless, as Olson and others have pointed out, collective action can still occur in latent groups. Three of the possible reasons for this are relevant here. First, if for any shareholder s, s_n increases at a faster rate than any increase in S_N, then for that shareholder it is possible that at some point p_s|R| > g_s, making it worthwhile for s to try to solve the collective action problem and to reduce excessive executive compensation. Other members of S would benefit as free-riders from the actions of this leading shareholder without directly incurring any further governance costs themselves. Secondly, an individual shareholder might obtain ancillary benefits from leading a collective action to recover executive rents, for example, by enhancing its reputation as an active investor, thus making it worthwhile for s to pursue an action on behalf of all members of S, despite that fact that on an individual basis g_s > p_s|R|. Thirdly, a group of shareholders might agree to work together so that on a collective action basis the potential reduction in their share of agency costs might exceed their direct governance costs. I describe the combination of these factors as the “principal force”⁸ or α. Therefore, Eq. (4.2) can be revised as follows:

$$ {\mathrm{U}}_{\mathrm{s}}\left(\mathrm{x}\right)={\mathrm{y}}_{\mathrm{s}}-{\mathrm{g}}_{\mathrm{s}}+\alpha\ {\mathrm{p}}_{\mathrm{s}}\mid \mathrm{R}\mid $$

(4.5)

where α > 1 if any combination of the three factors mentioned above applies; otherwise 0 ≤ α ≤ 1.

The last part of the model considers the position of the non-executive directors of F. The utility function of director d ∈ [1,…,n] is conventionally given by:

$$ {\mathrm{U}}_{\mathrm{d}}\left(\mathrm{x}\right)={\mathrm{w}}_{\mathrm{d}}-{\varepsilon}_{\mathrm{d}} $$

(4.6)

where w_d is the non-executive director’s financial compensation for serving F and ε_d is the director’s effort cost. Some directors have a powerful sense of their fiduciary responsibilities and professional ethics that defy conventional economic analysis, but can nevertheless be modelled by incorporating an additional factor into their utility functions. Directors are also subject to reputational effects. A director of good reputation can expect to obtain a portfolio of other high-status non-executive directorships.⁹ However, they are also subject to a set of onerous legal and regulatory obligations. In the model these moral, reputational, and legal effects are together combined in a factor that I call β, or the “fiduciary force”. This is a coefficient that is applied to the first part of the director’s utility function. Thus Eq. (4.6) is rewritten as follows:

$$ {\mathrm{U}}_{\mathrm{d}}\left(\mathrm{x}\right)=\beta\ {\mathrm{w}}_{\mathrm{d}}-{\varepsilon}_{\mathrm{d}} $$

(4.7)

where β can take various values. If β > 1, then the fiduciary force enhances the director’s utility function. If β < 1, then the director’s reputation is undermined and their utility function is correspondingly diminished.¹⁰

Study of UK FTSE 100

The Financial Times Stock Exchange 100 index is a share index of the 100 largest companies by market capitalisation listed on the London Stock Exchange. The index is maintained by the FTSE Group, a wholly owned subsidiary of the London Stock Exchange. Constituent companies must have a full listing on the London Stock Exchange, with sterling- or euro-denominated prices on the Stock Exchange’s electronic trading service. They must also meet certain requirements regarding a free float and the liquidity of their shares. On the date the case study was prepared (December 31, 2015) the largest company in the index was Royal Dutch Shell, with a market capitalisation of £160.1 billion. The total market capitalisation of companies in the index was £1.8 trillion, 73.2% of this value being represented by the top 35 companies and 85.1% by the top 50.

UK executive remuneration has escalated in the past two decades, with average chief executive pay in FTSE 100 companies reaching £4,284,000 in 2015, 171 times the average wage of employees (source: Income Data Services and Office for National Statistics)—see Fig. 4.1.

Fig. 4.1

Executive pay in the FTSE 100 in the period 2000–2015. (Based on data published by Income Data Services)

Section 439 of the UK Companies Act 2006 mandates an annual vote on directors’ pay, although these “say on pay provisions” (broadly correspond to equivalent provisions introduced in the US in 2010 by Section 951 of the Dodd-Frank Reform Act) are not binding on company.¹¹ Shareholders voted against FTSE 100 companies’ pay proposals on five occasions between 2009 and 2013, one of the largest revolts being in 2012 when nearly 60% rejected the £6.8m annual pay package of WPP chief executive Sir Martin Sorrell at the advertising agency’s annual general meeting. In April 2016 investors voted against BP’s remuneration report, with 59% of proxy votes cast going against the company’s decision to pay its CEO Bob Dudley nearly US$20m for 2015, a year in which the company ran up a US$5.2 billion loss. Andrew Tyrie, who was then chairman of the UK parliament’s influential treasury committee, urged investors to maintain their stand against excessive pay for corporate bosses following the BP vote. The Financial Times reported that BP’s board was facing pressure from large institutional shareholders to remove Dame Ann Dowling, who chaired its remuneration committee. The leader column in the FT urged company boards to take responsibility for limiting the quantum of executive pay.¹² Yet the typical structure of shareholdings in UK public companies makes this difficult to do for the reasons explained earlier in this chapter.

Under UK company law and stock exchange rules, any investor with a direct or indirect shareholding commanding 3% or more of the voting rights in a UK public listed company is required to disclose this to the company concerned. A company is required to identify in its annual report and accounts all investors owning 3% or more of its shares at the balance sheet date. Members with shareholdings representing at least 5% of a public listed company’s total voting rights can require the directors to call a general meeting and requisition the circulation of a statement regarding a proposed resolution. More than 50% of shareholders voting can pass an ordinary resolution at a general meeting, such resolutions being required, for example, to approve a related party or large transaction. Approval by 75% of shareholders voting is necessary for a special resolution, required, for example, by a public listed company to agree to a major transaction. More than 25% of shareholders voting can block a special resolution. Investors can appoint proxies to vote on their behalf. Proxies can often play an important part in corporate governance and shareholder activism by collecting mandates and voting en bloc. A number of proxy advisory groups, including Institutional Shareholder Services (ISS ), Glass, Lewis & Co., and Pensions Investment Research Consultants, issue general guidance on corporate governance and executive compensation; in some cases, they will do this by making specific voting recommendations.

A breakdown of holdings at various significant levels for each of the top 35, 50, and 100 companies in the FTSE 100 index on December 31, 2015 is set out in Table 4.1.¹³ Six companies have been excluded because they had a single dominant investor.¹⁴ On average 67.2% of shares held by the 100 largest shareholders in each company are held on behalf of retail investors by institutional shareholders (banks, insurance companies, mutual funds, pension funds, private equity firms, and other financial investors), 8.4% by trade investors, 9.0% by government (e.g., sovereign wealth funds), and only 6.1% directly by individual investors. The top 100 shareholders, on average, hold in aggregate 68.8% of the total share capital. In a typical company a small number of institutional shareholders (median = 5) have holdings of between 3% and 6% of company’s share capital, five investors might control more than 25% of the voting rights, 19 might control 50% of the votes, and 82 might control 75% or more.

Table 4.1 Analysis of shareholdings in FTSE 100 companies on December 31, 2015

What is apparent from this analysis is that, except in the small number of cases that have been identified, there is no single dominant shareholder. Even though, on average, five institutional investors control 26.5% of a company’s shares, sufficient to block a special resolution, this does not constitute a significant level of control, even if the five could be persuaded to act in concert. It takes on average 19 institutional investors to control 50% of the votes and 82 to control 75%. This is too large a number to make it likely that coalitions to vote down executive pay proposals will naturally emerge. Furthermore, many of these institutional investors are themselves listed companies—Table 4.2 provides an analysis. This means that there is an additional disincentive for those companies to vote against executive pay proposals because of the risk of reciprocal action.

Table 4.2 Investors with 3% holdings in FTSE 100 companies on December 31, 2015

The most commonly represented institutional shareholders were Blackrock (a US listed company), Legal and General (a UK listed company), and Capital Group (a US private company).¹⁵ Blackrock has itself been criticised for overpaying its CEO and for being “too soft” on “excessive executive remuneration” at companies in which it invests. In May 2016 75,000 people signed an online petition urging Blackrock to overhaul its approach towards executive pay at the companies in which it invests.¹⁶

Commentators have identified a number of possible solutions to the collective action problem among investors. An article published in The Review of Financial Studies in 2008 described in detail the activities of the Hermes UK Focus Fund (HUKFF) based on private information made available by Hermes, the fund manager owned by the British Telecom Pension Scheme.¹⁷ HUKFF generated above-average returns by engaging in private interventions with companies on various matters of performance and corporate governance, including executive remuneration. HUKFF was set up by Alastair Ross-Goobey, a well-known figure in the investment industry and a notable advocate of the need for shareholders to engage with boards and push for corporate governance reforms. The fund was created “as a response to the problem of free riding in institutional activism as perceived by the BT pension fund trustees. The trustees felt that the cost of higher intensity activism could not be sufficiently internalised through the core engagement, and it was therefore necessary to overweight the fund’s position in underperforming stocks that were to be engaged more intensively”.¹⁸ However, the case of HUKFF is an example of an exception which is the rule—HUKFF was unique among institutional investors when it was founded in 1998. Its significance declined after 2009 and it was eventually sold to RWC Partners, a London-based hedge fund, in 2012.¹⁹

Some hedge funds also engage in shareholder activism because of concerns about corporate governance. For example, Elliott Associates successfully fought a prolonged battle with the board of Alliance Trust, a FTSE350 investment company, over financial performance and corporate governance issues, including the pay of its chief executive Katherine Garrett-Cox. They eventually secured organisational changes, including the ousting of Garrett-Cox, before selling their stake in 2017. However, activist hedge funds typically have bigger fish to fry than executive remuneration. Interventions tend to focus on changing business strategy, especially where divestment or demerger has the possibility of realising substantial short-term capital gains.

A third possibility is that proxy advisory firms such as ISS help to coordinate the actions of disparate shareholders. PwC has examined the outcomes of advisory votes on remuneration reports for FTSE 100 companies for the period 2015–2017. They found evidence that ISS voting recommendations do have an impact on voting outcomes, increasing a negative vote by 10–15 percentage points when advising against a resolution.²⁰ However, this is a relatively marginal effect and will only occasionally cause remuneration reports to be voted down. PwC also points out companies often complain that ISS follows a mechanistic approach to voting recommendations, advising against atypical remuneration plans which depart from established norms, rather than carefully analysing the remuneration committee’s detailed proposals.

The data provided in this chapter, along with the case study of AstraZeneca in Chap. 2, illustrate the complex web of agency and fiduciary relationships which exist between shareholders and directors, directors and managers, institutional investors and company boards, retail investors and investment managers, and so on. Some of these are “strong” principal-agent relationships, where an agent has been appointed by a principal under the terms of a contract that specifies the terms and conditions governing the relationship. Others are “weak” fiduciary relationships, where there is no direct contractual relationship between the two parties and the connection is more akin to that of trustee and beneficiary. The case also demonstrates that another set of agency problems arises at the funds level, where the collective action problems are even greater because holdings in retail funds are more widely dispersed and because there is far less transparency about governance and pay than in public quoted companies: the pay of investment executives is not widely publicised; some investment firms are private companies or partnerships which are not subject to the same degree of scrutiny as public corporations; in any case, the pay of executives who are not also company directors does not have to be disclosed in the detail required of public company directors. The study provides evidence of the difficulties in limiting excessive executive pay when shareholdings are widely dispersed, as predicted by the formal theory.

α and β Factors

This chapter has identified two factors that are critical in determining whether corporate governance will be successful in moderating excessive executive compensation costs. The first of these is the α factor or “principal force” which determines whether shareholders will combine together to take collective action to address agency costs. The second of these is the β factor or “fiduciary force” which determines how probable it is that non-executive directors will carry out their fiduciary responsibilities to the fullest extent possible, thus having a moderating influence on agency costs. The formal theory set out in the second section of this chapter predicts that the principal force will be at its strongest if, first, a single institutional investor’s holding is sufficiently large as to make the proportionate benefits of reducing executive rents greater than the additional individual governance costs incurred in securing the reduction; secondly, an institutional investor obtains ancillary benefits from leading a collective action to recover executive rents, for example, by enhancing its reputation as an active investor; or thirdly, a group of shareholders agrees to work together to reduce executive rents and is able to spread the additional governance costs incurred in such a way that the benefits outweigh the costs in every case. The fiduciary force will be at its strongest if an individual non-executive director, for example, the chair of the remuneration committee has a powerful enough sense of their fiduciary responsibilities and professional ethics or expects to gain sufficiently valuable reputational benefits from taking a hard line on excessive compensation costs. These two factors, the principal force and the fiduciary force, are independent of each other but may operate in combination: corporate governance will be at its most effective when both α and β forces are at their strongest.

The conclusion stated here regarding the α factor is consistent with previous research on shareholder power; shareholder power has been described as a continuum extending from the relatively powerless (passive retail funds with small holdings in widely spread investment portfolios who rely, if anything, on soft activism) to the powerful (hedge funds and private equity firms who take large stakes in a small number of companies and follow a path of concentrated activism). In the middle are a number of active funds who rely on both soft activism and coordinated action.²¹ It is worth noting that shareholders in UK public companies possess more legal powers and participation rights than their counterparts in the US. The UK Financial Reporting Council has promoted the concept of “stewardship” by adopting a Stewardship Code in 2010 in response to a recommendation made by the Walker Review of Corporate Governance in the UK Banking Industry. The objective is to encourage investors to exercise their powers more actively, by engaging in debate with companies on their business strategies, financial performance, corporate governance and executive remuneration, as well as by voting and monitoring.²² Nevertheless, even during the “shareholder spring” of 2012 and its mini-revival during the season of company annual general meetings in April and May 2016, UK shareholders have only succeeded in overturning executive pay proposals in a relatively small number of cases.²³

The β factor illustrates the underlying paradox in standard agency theory of relying on the ethical motives of directors to solve agency problems. It reinforces the need, as set out in the previous chapter, to devise a more sophisticated model of economic man that recognises the significance of moral sentiments as well as economic impulses. It also gives force to the importance of developing normative models of executive and director behaviour that incorporate high deontic expectations of company directors and senior executives. By deontic, I mean expectations relating to duty and obligations as moral concepts. In other words, we need an agency theory that focuses on the professional ethics of corporate managers and company directors, not just on material incentives.

The Corporation as Commons

So far in this chapter I have worked within the parameters of neoclassical economics, seeking to demonstrate, through theory and empirical analysis, that a major shortcoming of the standard model is the fact that it overlooks the collective action problem at the heart of the public corporation. Working along similar lines, Simon Deakin, Professor of Law at Cambridge, has argued in a paper entitled The corporation as commons: rethinking property rights, governance and sustainability in the business enterprise that the commons might provide a better foundational model for theorising about public corporations than the current combination of the standard model and legal fiction theory.²⁴ He draws a similar distinction to the one that I have drawn between “the firm” and “the corporation”, quoting with approval the French jurist Jean-Philippe Robé, who says,

The firm and the corporation are very often confused in the literature on the theory of the firm. The two words are often used as synonyms. They correspond, however, to totally different concepts: a corporation is a legal instrument, with a separate legal personality, which is used to legally structure the firm; a firm is an organized economic activity, corporations being used to legally structure most firms of some significance.²⁵

Deakin argues that corporate law has a more central role to play in determining the nature of the corporation than the standard model envisages. He sees company law as an emergent phenomenon that has co-evolved with the emergence of corporations in industrial societies.²⁶ He calls for economic and legal theories of the corporation to be more empirically grounded in actual observation than the eviscerated view of legal fiction theory.²⁷ He concurs with my view that shareholders are not “owners” of corporations, saying,

Shareholders have many rights, ranging from voice and voting rights to rights in relation to distributions, which stem from the property they have in their shares. However, none of these rights either derives from or confers a right to property in the firm itself, or its assets, nor do any property claims which shareholders might have given them a right to manage the assets of the firm.²⁸

The “agency” responsibilities of directors and executives are determined partly by company law and partly by employment law. The fiduciary responsibilities of corporate managers to the corporation, derived from common law, are more substantial than an agency perspective might imply. Other employees also have rights and responsibilities determined by employment law, and management’s authority over them is conditioned by their responsibility for the physical, economic, and psychological well-being of workers.²⁹

Deakin’s conceptualisation of the collective action problem at the heart of the public corporation is, however, much more widely drawn than the picture I have painted in the previous section. He puts it like this:

The firm as such cannot be owned, but in the context of the modern business enterprise, there are multiple, overlapping and often conflicting property rights or property-type claims which the legal system is meant to adjust and reconcile. As we have seen, corporate law is largely concerned with one set of such rights, those of shareholders, but this by no means exhausts the set of claims on the firm’s assets. Employment law, insolvency law and fiscal law also identify claims of this kind. Each of these areas of law has a dual function: specifying the conditions under which various contributors of inputs (or, as they are sometimes called…“stakeholders”) can draw on the resources of the firm while at the same time preserving and sustaining the firm’s asset pool as a source of productive value. This is the sense in which the business enterprise is a “commons”.³⁰

In other words, in addition to shareholders, certain other persons, most notably employees (through obligations enshrined in employment law), also have rights in respect of the commons that management must respect. Corporate managers must arbitrate between these various “overlapping and conflicting” rights at the same time as they exercise their responsibility for maximising total firm value over the long term and “sustaining the firm’s asset pool as a source of productive value”. This is a much more holistic view of the agency responsibilities of executives.

One of the benefits of the standard model, according to its proponents, is the way that it resolves value claims between different stakeholders. Michael Jensen has argued that stakeholder theory is flawed because it violates the principle that a single value objective is a prerequisite for rational corporate strategic decision-making. He goes on to say: “a firm that adopts stakeholder theory will be handicapped in the competition for survival because, as a basis for action, stakeholder theory politicises corporations and leaves its management empowered to exercise their own preferences in spending the firm’s resources.”³¹ The standard model tries to resolve these difficulties by allocating property rights and specifying that the primary objective of the corporation is to maximise shareholder value. This principle should be used, supporters of the standard model say, as the decision criterion for all major corporate decisions, including, for example, whether to acquiesce to a hostile takeover bid, whether to outsource a major part of a corporation’s activities in the interests of cost savings, but at the expense of direct employment opportunities, whether to forgo current investment opportunities in order to benefit short-term profits, but at the expense of long-term value creation, and so on.

However, Deakin points out that there is an extensive literature describing an empirical research programme conducted over two decades, principally led by the Noble prize winner, Elinor Ostrom, which shows that the apparent contradictions and conflicts in the collective use of valuable resources can be overcome if appropriate governance and management regimes are put in place. The research on common pool resources is summarised in a collection entitled Working Together – Collective Action, the Commons, and Multiple Methods in Practice by Amy Poteete, Marco Janssen, and Elinor Ostrom.³² They explain how eight design principles for the governance and management of common pool resources can be inducted from the empirical work. These design principles—summarised in Table 4.3³³—are a rich source of ideas about the effective governance of public corporations.

Table 4.3 Managing the commons as a source of corporate governance design principles (after Deakin, 2012)

The most critical of these eight design principles are discussed further below along with the implications for the governance of public corporations.

Proportionality Between Benefits and Costs

Poteete , Janssen, and Ostrom emphasise that, for the effective management of common pool resources, benefits should be allocated in proportion to inputs, for example, of capital and labour. Rules that respect proportionality are more likely to be regarded as equitable. Rules that disproportionately benefit elites will be perceived as inequitable. Perceived inequity undermines trust. Perceived fair pay is an important characteristic of high-trust organisations. If shareholders, and employees generally, perceive that senior executive pay is excessive, then their confidence in top management will be undermined. If excessive executive compensation is seen as a collective action problem, and public companies are in effect quasi-public goods, then extracting high pay for managerial elites or excessive special dividends in the short term for shareholders is like “overharvesting” in a common pool situation.

Collective Choice Arrangements

There should be broad participation in governance arrangements—individuals who are affected by resource allocation rules should have representation rights in governance systems. Voice mechanisms, such as works councils, employee advisory panels, and worker representation on company boards or major committees, can be important ways of building trust.

Monitoring

Individuals charged with monitoring should be broadly accountable, as reliable monitoring raises confidence among users of common pool resources. The board of directors must recognise its accountability to a wide range of stakeholders, including minor as well as major shareholders, employees generally, the communities in which the corporation operates, and so on.

Conflict Resolution Mechanisms

There should be rapid, local conflict resolution arrangements. Local mechanisms that allow conflicts to be aired quickly help to build trust. Some conflicts arise simply because users interpret rules differently. Sanctions for violations of rules should be graduated. Graduated sanctions signal that infractions are notices while allowing for misunderstandings, mistakes, and exceptional circumstances. Companies should recognise that conflicts with stakeholders will inevitably arise. It is important to ensure that there are mechanisms for resolving conflicts quickly, when they do arise, and that management’s mistakes are acknowledged.

Local Governance Arrangements

In much the same way that Neil Fligstein describes organisational fields as “embedded in other fields like a Russian doll”,³⁴ so Poteete, Janssen, and Ostrom talk about “nested enterprises”. They advise that, in complex common pool structures, users should be encouraged to devise their own governance arrangements as these will be best suited to local conditions. The role of governmental authorities is to enable and support local governance. This principle is relevant to the governance of multinational firms. It is consistent with ideas about self-determination and self-regulation, underpinned by the legal system, with government intervention only when it is clear that self-determination is not working effectively.

Conclusion

In this chapter I have explained why shareholders of public corporations are unlikely to resolve executive pay dilemmas because of collective action problems, and how ideas about the governance of common pool resources have implications for the design of effective corporate governance mechanisms. I shall return to effective corporate governance architecture in the final chapter. In the meantime, I turn in Chap. 5 to the design of executives’ incentives, and to the lessons that can be drawn from behavioural science.

Further Reading

A number of the essays in Joseph Heath’s book deal with collective action problems in public corporations. A good general text, which summarises Mancur Olson’s ideas and also covers the prisoners’ dilemma, is Hardin, R. (1982|2013). Collective Action. Routledge. Readers may also like to refer to Olson’s own seminal work, especially Parts I and II, Olson, M. (1965|1971). The Logic of Collective Action – Public Goods and the Theory of Groups. Harvard University Press – the situation of public corporations is addressed on pages 55–57.