The debate on executive pay became more intense in the UK last weekend, with politicians of all hues arguing that it is too high and needs to be contained. In an interview, Prime Minister David Cameron told the BBC that there had been a "market failure", with some bosses getting huge rises despite firms not improving their performance.
I won’t argue with the comment that many executives receive too much pay. I won’t argue with the statement that the pay is apparently increasing regardless of performance (although I would refer you to Manifest’s blog which gives some numbers on this). My position in this article is that it’s not a market failure, because there is no real ‘market’ in top executives.
I’ve been saying this for a while. I first pointed it out in my PhD, in which I interviewed members of remuneration committees, executives and advisors. I went on to write it in academic articles such as The Platonic Remuneration Committee. But on the basis that few people will have the time or inclination to read any of that, I set out some thoughts below.
1. The UK Corporate Governance Code does not require that pay be benchmarked against a market. It says, more subtly, that committees should “judge where to position their company relative to other companies”. The word ‘market’ is not mentioned, and so the debate really centres around defining appropriate comparators.
Each company selects its own comparators (generally, research suggests, from companies where pay is high). Therefore, each company sees a different ‘market’. It could be argued that what is being benchmarked is not a market but instead a sampling population: companies define their market as a limited population of elite salaries, sampled 100%. (Not surprisingly, nobody ever says that – the term ‘market’ carries a lot more legitimacy than ‘sample of elites’!)
2. Even if companies attempt to define their market, data become anomalous because companies cannot benchmark like against like. In my ‘Platonic’ paper I give the example of a utility that unintentionally ended up benchmarking against high tech companies through a hidden chain of comparators.
3. Individual executives are not fungible. They have different qualifications, talents, backgrounds and reputations. These human capital attributes mean that any ‘market-determined’ price may need to be tailored to individual circumstances. Companies too vary widely - two organisations are not similar just because they have the same turnover or market capitalisation. Simple benchmarks ignore pay that has been awarded for individual skills or circumstances.
4. Executive pay will increase for three reasons: pay inflation, new entrants to jobs; and merit awards for individuals progressing in their roles. Lumping all pay rises together without differentiating their cause gives a distorted view of the ‘market’ rise, which should just be that inflationary element.
In summary, executive pay is not a market in the generally-accepted sense. When economists speak of markets they assume a competitive market in which prices settle at the intersect of supply and demand curves to arrive at a market-clearing price. The goods being traded are identical, or at least closely similar. If a merchant sets his prices at a premium to the market-clearing price he will find that there are few buyers; likewise, setting below-equilibrium prices should increase the demand for his goods. That isn’t in any way what we have in executive pay.
Putting it another way, the ‘market’ surveys on which we rely do not themselves represent a price at which individuals could, to put it crudely, be acquired. They represent the prices currently being paid to other elite individuals. They are not an ‘offer for sale’ and will never be tested as such. The market data reflect what others are reporting as earnings, not what a market-clearing price would be. For example, if the median rate for CEOs in a certain sector were £300,000 one could probably find a qualified individual who would take the job for, say, his reserve price of £200,000. Paying the median gives that individual a surplus of £100,000. Put in terms of the UK Governance Code, the money is certainly sufficient to “attract and retain”, but it fails on the criterion of “not excessive”. Indeed, taking this analysis a stage further, we can see that individuals who see their worth as being higher than the median get paid as such, but others get paid at the median, rarely below it. Thus, the median is an inadequate proxy for the true market-clearing rate.
Does it matter that we refer to executive pay as a ‘market’? Yes, I think it does. We should not say that the market has failed, because it conveys the wrong impression - the market does not exist. Trying to correct that ‘market’ is perhaps a waste of time, and other approaches may be more useful. We need a radical approach to executive pay, and re-defining and clarifying our terms might help us to start to achieve this.
No comments:
Post a Comment