Inequality and the Firm – To What Extent Are Corporate Pay Practices Contributing to the Increase in Inequality?
by Dr. Sandy Pepper
Thomas Piketty, in his book Capital in the Twenty-First Century, partly attributes the rise in inequality during the latter part of the 20th and early part of the 21st centuries to “the rise of the super-manager”.
My colleague Professor Paul Willman and I, with the support of the International Inequalities Institute at the London School of Economics, have been investigating why this might be the case, focusing on data from the UK.
First the data: during the period 2000-2014 (a period which, incidentally, includes the financial crisis of 2008-9) the annualised rate of increase of the total earnings of FTSE100 CEOs was 11.1% and for FTSE250 CEOs was 9.2%. During the same period the annualised increase in the retail prices index was 2.5% and the annualised increase in average nominal earnings was 3.3%. The median total pay of FTSE100 CEOs in 2014 was £4,284,000 compared with average wages of £25,029, a ratio of 171:1. In 2000 the equivalent statistics were £884,000 and £15,800, a ratio of 56:1 (Source: Income Data Services and Office for National Statistics; see figure 1)
While our focus has been on the UK, a similar story could be told with data from most other developed Western economies. So what is going on? Our research findings suggest that two factors in particular explain the data.
The demise of collective bargaining. In both USA and UK, unionisation increased throughout the middle part of the 20th century, peaking at just under 30% in USA in the early 1960s, but continuing to rise in UK until 1980, reaching nearly 55%. In the UK, trade union membership and collective bargaining coverage was substantial throughout the post-war period to 1980. Collective bargaining appears to have compressed differentials not just within bargaining units, but between them. There is evidence indicating the presence of unionism exerts a downward effect on senior management pay. Successive survey data shows that union membership and coverage of collective bargaining fell consistently from the 1980s. Between 1989 and 2014, union density fell overall from 38% to 25%. By 2013, collective bargaining coverage in the private sector in the UK stood at only 16%.
The growing prevalence of executive share plans. Two seminal articles published in the mid-1970s and early 1980s had a profound impact on academic thinking about executive compensation. Agency theory argued, inter alia, that in order to motivate executives (agents) to carry-out actions and select effort levels that are in the best interests of shareholders (principals), boards of directors, acting on behalf of shareholders, must design incentive contracts which make an agent’s compensation contingent on measurable firm performance outcomes. Tournament theory extended the agency model by proposing that principals structure a company’s management hierarchy as a rank-order tournament, thus ensuring that the highest-performing agents are selected for the most-senior management positions. Tournament theory postulated that executives compete for places in a company’s upper echelons via a sequential elimination tournament. It predicted that compensation is an increasing convex function of an agent’s position in the management hierarchy, with increases in remuneration between levels in the hierarchy varying inversely in proportion to the probability of being promoted to the next level. By implication, the compensation of the CEO, ranked highest in the tournament, would typically be substantially more than the compensation of executives at the next highest level.
The popularity of the two theories in academic circles was accompanied by changes in management practice, as an increasing proportion of senior executive pay was delivered in the form of stock options and other types of equity incentive. The pervasiveness of these instruments was not even affected when, in 1990, in a large scale empirical study in the US, Michael Jensen and Kevin Murphy were unable to find a strong connection between CEO pay and performance. Somewhat perversely, given that agency theory had been advanced as a positive theory (of what is) rather than a normative theory (of what should be) Jensen and Murphy concluded that the weak empirical connection between changes in CEO pay and changes in firm value meant that companies should be providing a greater proportion of compensation in the form of stock. That is indeed what subsequently happened in practice. For example, in the UK the proportion of FTSE 100 CEO pay delivered in the form of stock options or long-term share-based incentive plans increased from 12.7% to 48.8% between 2000 and 2015 (Source: Income Data Services).
Thus it can be seen that in the UK, from the mid-1980s onwards, senior managers were increasingly being rewarded on capital market measures, not the evaluation of jobs. At the same time, given the collapse of collective bargaining, many lower paid workers have had their income increases pegged. In these circumstances, administrative processes for the firm to regulate intra firm equity have given way to capital market and statutory influences in such a way that firms cannot easily control inequality within their boundaries.
Interview with Prof. Pepper
GBV: Your study deals with the UK but its conclusions appear to apply to most developed Western countries. Are there any exceptions to the rule?
Sandy Pepper: Indeed, our research is based on UK data but we do have anecdotal evidence that our findings would apply in the USA and in Northern Europe and we believe that most other Western countries where the same underlying forces are at play, would exhibit similar patterns. But this is not to say that this is a universal truth. Japan, for instance, would be very different, with much lower levels of inequality and no upward trend in the pay gap.
GBV: Resorting to stock-options and other long-term incentives that are granted after a significant delay seems to fly in the face of many insights from research in psychology, which state that the more immediate the reward, the more powerful it is. So why do shareholders grant long-term incentives in the first place?
SP: Certainly, there is an element of fashion – a management fad – at play. Agency theory became the foundation of many recommended management practices, despite its shortcomings and its negative side-effects. As a matter of fact, when it comes to pay-for-performance, executives require higher sums whenever the pay-out is deferred or uncertain. Share-based deferred compensation is both random (share prices fluctuate over time) and highly defend (three to five years seems to be the norm).
As a result of that, and somewhat perversely, executives have required and received very large share allocations which have resulted in the huge pay-outs that have been a major factor in the rise of pay inequality over the last 25 years. In other words, seeking to align executives’ interests with those of shareholders has both been costly to shareholders and widened the pay gap between executives and employees.
This fashion may go away and deferred share-based compensation may be replaced by cash bonuses with an option to get shares instead. Indeed, owning shares of one’s employer results in better alignment with the interests of other shareholders and should be encouraged.
GBV: Do large pay gaps in a firm matter at all?
SP: The evidence suggests that large pay gaps do have a detrimental effect on worker motivation unless a convincing explanation can be provided by executives to their teams. Absent rational accounts for big differences, I believe that firms need to take steps to reduce them.
GBV: Do employers bear the responsibility of keeping pay inequalities at a sustainable level or is it a matter for government?
SP: There are different views on that. Governments can reduce inequalities either by regulating compensation, either explicitly or through indirect measures such as enforced transparency and say-on-pay, or through the tax system, for example with progressive tax rates. But I think that employers have some responsibility to act and should take things in their hands.
GBV: What would be a “good” ratio between the highest and the lowest pay in a company ?
SP: Management expert Peter Drucker thought that a 1:20 ratio was as high as one should go, even though 1:50 ratios were already common in 1960s – still a far cry from what we observe today. I wonder if a ratio of around 1:100, such as we see in Switzerland, Germany and France, is the right sort of level. But one must also consider the fact that computing one firm’s ratio is not as easy as it seems. For example, should one use employees’ average salaries, or median salaries as the basis for comparison? What exactly is included in compensation? What about payroll taxes, withholding taxes, employee benefits?
In a nutshell, firms have in the last 25 years lost control of pay inequality as the power of organized labour has waned and executives’ variable pay has become both much higher and more random. High degrees of inequality being linked to lower levels of motivation at the firm level, firms should reach collective agreements with employees’ representatives and change the level and structure of executive compensation, notwithstanding societal issues as highlighted by economists such as Thomas Piketty.
Employers can and should do so on their own; government intervention is not needed unless no action is taken.