Never pay bonuses to CEOs – it makes them worse at their job

High CEO pay is back in the news. This is disappointing. We keep thinking that the tide is turning, and that top corporate executives and their self-serving remuneration committees have finally grasped that it is nuts to pay people enough to transform their finances for generations, to run a company that they didn't start and don't own.

We keep being wrong.

The Mail on Sunday this week reported on the tripling of the amount to be paid to Mike Tye, the chief executive of Spirit Group, and on the £9.5m to be paid to Carolyn McCall of Easyjet.

Subscribe to MoneyWeek

Subscribe to MoneyWeek today and get your first six magazine issues absolutely FREE

Get 6 issues free

Sign up to Money Morning

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Sign up

There was also a row over at BG Group earlier in the month when CEO Helge Lund was awarded a total package worth something in the region of £14m. This was later halved but £7m is still clearly an awful lot of money to earn in one year for taking over a going concern.

We've written here a great deal about the pointlessness of all this, about the talent myth', and about the silly long-term incentive plans that got us to this point. But there is one experiment, pointed out to me by David Gait of First State, that makes the case particularly well for dumping financial incentives and going back to fixed pay levels for all.

It is the candle task. Subjects are given a candle, a book of drawing pins, and a box of matches. They are asked to use them to find a way to fix the candle to the wall so that wax will not drip on to the floor when the candle is lit. The answer is to empty the matchbox and pin it to the wall. You then put the candle in the matchbox. Job done.

This sounds easy, but for most people the matchbox is not immediately obvious as a tool: it is seen purely as a container for matches. They can't do it. (But if you take the matches out, pile them next to the box and give a subject the same task, almost everyone can do it)

This experiment was refined in 1962 by Sam Glucksberg to see how financial incentives affected the outcome. Two groups were used. The first were not offered cash rewards, just told that it was pilot work on problem solving. The second were told that those who finished the task fastest would get $5 or $20, with the very fastest getting the latter.

Here's what happened: those incentivised by high cash payments performed worse than those not incentivised by payment at all. Yes you read that right: worse.

There are several explanations. It could be about justification people think that if they are getting paid that much to do something it must be really hard, so they overcomplicate it.

Or it could be that making payment dependent on performance creates stress, something that we know can shut down our creative and problem-solving skills (think fight or flight'). Glucksberg reckoned it was the latter.

The conclusion was that if you are hiring people to do what is essentially problem solving work (as most management is), tying their pay directly to the result of that work is a really bad idea. Something for shareholders to think about.