Fat Cat Tuesday: Adam Smith vs The High Pay Centre

Today is ‘Fat Cat Tuesday’. That is the name given by the UK High Pay Centre (HPC) – a think-tank focusing on issues of corporate governance and executive pay in the UK – to the day of the year when the average CEO of the FTSE 100 companies has earned as much as the rest of the UK population will earn all year. In 2016 that is the case already on the first Tuesday of the year! Indeed, the High Pay Centre has calculated that the average hourly wage of the average FTSE100 CEO is £1,260 while the median average yearly salary in the UK is £27,645. Assuming that CEOs start working on Monday 4th of January 2016 and work a 12 hour day, by Tuesday afternoon they will have earned more than the £27,645 median annual pay (see here).

This simple and admittedly rather coarse measure illustrates in a very effective way an important aspect of income inequality in the UK, which is closely related to the strong increase in top CEOs' salaries over the past decades. The average pay of the chief executives of the FTSE 100 companies has more than quadrupled between 1998 and 2011 from roughly £1m to £4.2m. [1]

Yet, the HPC’s idea of Fat Cat Tuesday is also heavily criticised. Some strongly worded reactions in the comments section of the HPC’s own web page are damning, accusing the HPC of ‘blatant statistical manipulation’, because the calculation compares the median average national annual wage to the mean average CEO wage.
This is hardly a convincing criticism of the methodology though. The difference between the median annual wage and the average annual wage in Britain in 2014 was, according to the ONS, £5,247. So, comparing the mean average CEO wage with the mean average national wage essentially means that Fact Cat Tuesday will be reached 4.16 hours later than if we used the median wage. So, at best, it implies that we should be talking about a Fat Cat Wednesday. The fundamental point that the HPC is making clearly remains valid.

However, more serious voices too criticises the HPC’s newly crafted ‘commemoration day’. The Adam Smith Institute (ASI) – a libertarian think tank based in London – condemns the HPC in no uncertain terms. In a press release, its Executive Director Sam Bowman derides the HPC analysis as ‘pub economics’.

It is worth quoting the ASI’s press release in full here:

‘Commenting on the High Pay Centre’s promotion of ‘Fat Cat Tuesday’, Executive Director of the Adam Smith Institute, Sam Bowman, said:

Despite consistent attacks on chief executive pay, the High Pay Centre has never told us how much they think CEOs are actually worth. Their complaints are the hand-waving of pub economics, not serious analysis – “Surely you don’t think executives can be this valuable to firms?”, or “Surely you don’t think executives are more important now than they were forty years ago?”.

None of these complaints are valid unless the High Pay Centre thinks it has a better way of estimating the value of executives to firms than those firms themselves. Can the High Pay Centre tell us how much CEOs are worth? If not, how can they say that they are overpaid?

Chief executives can be worth quite a lot to firms, as is shown by huge moves in company share prices when good CEOs are hired, or bad CEOs are fired. Steve Jobs can make a firm; Steve Ballmer can break a firm. The High Pay Commission’s complaints only make sense if you assume firms don’t actually care about making money – which is to say, they don’t make sense at all.’

This is an interesting statement, which unfortunately does not live up to their self-declared standards of ‘serious analysis’ either.

To be sure, it is indeed a very important and difficult question to know what exactly is ‘excessive’ CEO pay, or – in the words of the ASI – ‘how much is a CEO worth?’ This question should be discussed and would certainly merit more attention in the academic literature too. Nevertheless, the tautological argument that if ‘the firms’ are ready to pay their CEO as much as they do, it shows that they are actually worth as much, is incredibly naïve.

However, the argument that if firms are ready to pay such salaries they are justified is actually very wide-spread and influential among academics and practitioners alike. Indeed, it may be the dominant argument in the debate about executive pay. The probably most influential academic working on executive pay, Kevin J. Murphy, very much defends this point of view. This is not very surprising as Murphy was one of Gary Becker’s PhD students at Chicago and hence firmly rooted in the Chicago School of Economics, which in turn is closely associated with the efficient market hypothesis (EMH). Indeed, the argument that salaries are justified because firms are ready to pay them is – albeit implicitly – based on the belief that executive pay-levels are the result of a market equilibrium in the market for managerial talent. We could call this the efficient labour market explanation of executive pay.
This argument raises all sorts of fundamental questions about the validity of the efficient market assumption in general, and of the liquidity and efficiency of labour markets for highly specialised roles in particular.

More interestingly, however, is the other implicit argument made by the ASI, which consists in saying that firms somehow know how to estimate the ‘value’ of their CEO. This is interesting in several respects.

Thus, the fact that the statement refers to ‘firms’ in general shows that the ASI – in typical neo-classical fashion – treats the firm as a ‘black box’. On this account, it is the firm that estimates the CEO’s worth and that decides about the appropriate pay level. This, however, obscures the fact that firms are not unitary actors, but are composed of various parties or ‘stakeholders’ with various – at times conflicting – preferences. The simplistic neoclassical view of the firm also completely neglects the process by which executive pay is determined within the firm. Indeed, nowadays it is generally the case that executive pay is determined by the remuneration committee, which is a sub-committee of the board of directors. The UK corporate governance code (D.2.1) requires that the members of this committee be independent non-executive directors. The CEO should not be member of the committee. Nevertheless, a second influential theory of executive pay in the corporate governance literature argues that in spite of such provisions, executive have a great deal of influence over the determination of their own pay. This could be termed the ‘managerial power thesis’ of executive pay and is most commonly associated with the work of Lucian Bebchuk– a very influential Harvard Law professor and staunch advocate of minority shareholder interests. In a widely-cited paper with Jesse Fried, Bebchuk discusses a number of reasons why it is likely that boards of directors – even with a remuneration committees composed of independent directors – is likely to design pay packages for top executives that are overly generous and not related to performance. Their explanation essentially is that non-executive board members have more reasons to please the CEO than distant and anonymous minority shareholders.

But even if directors and the members of the remuneration committee were sensitive to what shareholders want in terms of executive pay, it is by no means certain that increasing shareholder influence would lead to curbing executive pay. Indeed, different countries – including the UK – have recently introduced so called ‘say on pay’ rules, which give shareholders a right to vote during the Annual General Meeting (AGM) on executive remuneration packages. This was expected to lead to more scrutiny of executive pay and to a slowing down of CEO pay increases. Yet, anecdotal evidence and first empirical studies a couple of years after the introduction of these rules in the UK indicate that – despite some incidents of shareholder discontent – executive pay has continued rising essentially unhampered. That is puzzling from the mainstream view of corporate governance, which sees the conflict between managers and shareholders essentially as a zero sum game. One explanation for this puzzle is that shareholders do not really care how much managers are paid as long as they get their share of the profits. Indeed, it is an open empirical question at whose expenses managers have increased their salaries in recent years. More left-leaning academics have made the hypothesis that – rather than clashing over salaries – CEOs and other top executives form a coalition with shareholders and extract value from the company at the expenses of middle managers and shop-floor workers (see generally Krier 2005 who speaks of a ‘speculative coalition’, or a ‘speculative management teams’ composed of top managers and aggressive activist shareholders). This could be a counter argument to the ASI’s claim that executive salaries cannot be excessive if firms care about making money. Well, they certainly do want to make money, but the relevant question here is who actually gets the money that they are making. It is perfectly plausible that CEOs may be able to ‘buy’ pay levels above the minimum competitive equilibrium wage by essentially ‘bribing’ shareholders into accepting such levels in exchange for increasing pay-outs via dividends and share buybacks. That way, both shareholders and managers benefit, at the expense of other parties, e.g. employees, investment in future growth opportunities, and R&D activities.

In any case, the ‘managerial power thesis’ as well as the ‘speculative management team’ thesis suggests that the efficient executive labour market hypothesis, which by assumption implies that the salaries that are being paid constitute a competitive equilibrium, may be false.

Still, the thesis that the ever-increasing executive salaries are somehow a result of an increasingly competitive global market place for managers is omnipresent in the practitioner discourse (cf. the popular saying ‘If you pay peanuts, you get monkeys’). But even this seemingly very compelling and intuitive argument may be false. The argument would suggest that the more money you offer, the better the person you will be able to recruit will be. That’s a bold argument that fits in well with the neoclassical ‘homo oeconomicus’ theory of human motivation that sees money as the only motivator. It fundamentally contradicts more classical theories of motivation in organisational theory, which show that money is certainly part of what motivates people, but by no means the only factor (cf. Maslow’s pyramid of needs, Herzberg’s two-factor theory, and various process theories of motivation).
To use the ASI’s own example of former Apple CEO Steve Jobs and former Microsoft CEO Steven Ballmer: The ASI implies that Jobs was a good CEO, while Ballmer was not. Following their reasoning we would expect Jobs to have had a (much) higher salary than Ballmer. Yet, Jobs famously was one of a handful of CEOs who only had a symbolic yearly fixed salary of $1 and was not paid any bonuses in most years of his tenure at Apple between 1997 and 2011. Conversely, Ballmer – the ‘bad CEO’ in ASI’s example – took home around $1.3m every year (see here). In comparative terms, that is not a massive salary for the CEO of Microsoft and it is clear that Jobs received other services and perks from Apple in lieu of a salary (e.g. Apple would cover his annual travel expenses of around a quarter of a million USD). Nevertheless, even this crude comparison does not seem to support the view that the level of the salary directly reflects the quality of the CEO.

When Ballmer left Microsoft, he was replaced by Satya Nadella who was awarded $84m in his first year at the helm of the technology company. A massive increase over the Ballmer’s pay, which was justified by the chairman John Thompson with the usual argument that the company wanted to “attract and motivate a world-class CEO”.
In his second year, this ‘world class CEO’ had managed to half the net income of the company from $22.1bn to $12.2bn. His base salary still continued to increase compared to his predecessor (who had earned around $600,000 in base salary), to $918,917 in 2014, and $1.2m in 2015. He also still received 120 per cent of his target bonus ($4.3m) for 2015 in spite of the rather disastrous income figures (see here).

Many more examples could be added that show that the case for a close and direct link between the level of pay and the quality of the CEO is weak to say the least.

To be sure pay packages are complex things and in particular the contribution of a CEO to the company’s performance (good or bad) is extremely difficult to isolate from other factors that affect the business (such as the global market environment and pure chance) in any given year. Nevertheless, the empirical evidence and the various theories mentioned above clearly show that the Adam Smith Institute’s simplistic arguments are hardly any better than ‘pub economics’.


[1] Pickard, J., Groom, B, Masters, B. ‘Cable plans binding votes on executive pay’, Financial Times, June 20, 2012.