THE problem is as old as mankind. The Roman author Juvenal encapsulated it into a phrase “Quis custodiet ipsos custodes” or “Who guards the guards themselves?” It was neatly illustrated in the classic BBC series “I, Claudius”. The infirm Claudius wants the return of the Republic. But the Praetorian guard, set up by his relatives, needed an Emperor to ensure their special status. So on the murder of Caligula, they drag Claudius from his hiding place behind a curtain, and make him Emperor.
Throughout history, dictators have faced this problem. They can surround themselves with men with swords or guns. But it only takes one guard with a sword or gun to turn into an assassin or to seize power for himself. The Shah of Iran had a huge army in 1979 but it did him no good; the soldiers had more sympathy with the revolutionaries than with the Shah himself.
In business and finance, this is known as the “principal-agent” problem. Shareholders employ managers to run a company; investors use fund managers to look after their savings. That makes sense. It allows us to take advantage of the expertise of others, and of economies of scale in fund management (it costs little more to look after $10m than $1m). But it is extremely hard to align the interests of principals and agents exactly.
Before the 1980s, the worry was that business managers would worry more about expanding the company (and increasing their power, pay and perks) than in shareholder returns. So share options were dreamed up to align the interests of the two. However, share options are a one-way bet; very valuable if they get exercised but costless to the executive if they do not. (And the exercise prices are often rewritten in the latter case.) The overall effect has been to ratchet up the pay of executives, ultimately at the expense of shareholders. As the FT reports today, some efforts are being made to rein this back in the most egregious cases, but progress is slow; the cost of executive pay is spread very widely while the benefits go to a few.
In effect, this is a bit like the subsidies paid to raw materials producers in some economies. Those who receive them make out like bandits but the cost is spread among a large number of consumers. Indeed this is another example of the principal-agent problem. Legislation is left to our agents, the elected representatives. But they tend to respond to the concerns of those who lobby them. Those in receipt of million-dollar benefits are likely to lobby harder than consumers who pay a few pence more for a given shopping item.
In investment, fund managers are paid through ad valorem fees, a percentage of the funds invested. In a recent article, Paul Lewis, the BBC presenter, jokes that ad valorem is Latin for rip-off. Again the ad valorem approach sounds like it ought to work; if the asset value goes up, then so does the pay of the fund manager and the wealth of the client. But if the market stays flat, the fund manager still earns money while the client is worse off (to the extent of the fee). And if the market goes up, the fund manager’s fee will rise, even if he or she underperforms the index. Performance fees do exist in the hedge fund industry but those are on top of annual management fees. If managers were really confident of their skills, they would surely take all their earnings in performance fees; say 50% of everything above the benchmark but not a penny for anything less.
Investors are catching on to this problem; around half of all fund flows last year went to Vanguard, the low-cost index-tracker. But a lot of money is being earned by mediocre fund managers. As with executives,the problem over the last 30 years is that the agents are getting faster much more quickly than the principals.
A further problem with modern finance and business is that affairs are extremely complex; so there is an “information asymmetry” between the clients and the agent. The latter understands far more and thus knows which loopholes to exploit. Incentivise executives with an earnings per share target, for example, and it is relatively easy to run the business towards meeting that target rather than focus on things that create long-term value such as capex.
That suggests a more broader requirement such as the fiduciary principle, something which the Trump administration wants to water down (read the piece by Jack Bogle of Vanguard on this, headlined “Putting Clients Second“).
The problem that links business, finance and politics here is trust. Trust is easier to deal with in small communities where one can deal with each other face to face; this was even true of the Athenian democracy. Once our relationships with our agents are more remote, and our transactions more complex, we have to rely on incentive schemes and these are ripe for exploitation.
The trust problem is particularly difficult in politics where a large proportion of the public no longer trusts mainstream leaders. In part, this is because some have shown themselves untrustworthy. But it is also because the incentive structure of the system (particularly in the US, where funding is such an issue) teaches politicians to ally themselves with the powerful. And it is also because the problems they face are not amenable to easy solutions; as I’ve remarked before, important issues are global and complex but politicians get elected by focusing on simple and local answers. When those answers fail, cynicism increases.
The irony is that, as Edward Luce writes today, that Americans have elected a leader who is replte with conflicts of interest at every turn. When his promises disappoint, cynicism will only be given a further upward notch. Indeed, the issue of checks and balances seems all the more important; who guards a country against its elected leaders?