Advertisement

Why bonuses don’t work

They're falling one by one. First RBS, now National Rail. This year's bonus season could finally be the moment the tide turns against these incentives, at least where public ownership is a factor. Labour's opposition day debate wants to limit them to just "genuinely exceptional performance". But even that is too much.

Bonuses should be scrapped for all but the simplest mechanical tasks. This is not for moral or political reasons but purely financial ones. They do not work. Bankers and City lobbyists tell you they must reward highly-talented individuals or else they will flee the company. The argument is self-serving, but more importantly, it is demonstrably false.

In 2005, the Federal Reserve Bank of Boston published a study by economists at MIT. Researchers had collected a group of students and offered them cash rewards of around $50-$60 for completing a set of tasks, which ranged from memorising strings of digits to throwing balls at a wall. There were three levels of reward available. The researchers found that for tasks which involved mechanical skill bonuses worked very well and led to higher performance. If you use them for picking strawberries and factory assembly lines, they will pay off.

But where the tasks required even rudimentary cognitive skill, something interesting happened: the bonuses led to poorer performance. "Tasks that involve only effort are likely to benefit from increased incentives," the economists concluded. "While for tasks that include an element of thinking, there seems to be a level of incentive beyond which further increases can have detrimental effects on performance".

Being economists, these finding unnerved the researchers. They took their research to Mundurai, India, where the financial incentive was much stronger than in the US. Here, low performance led to two weeks' wages, medium performance led to a month's and excellent performance led to two months' wages.

The results? The medium group performed no better than the low performance group, and the top reward group did even worse. The tests have been replicated over and over again — by economists, sociologists and psychologists. We get the same results, every time.

Worse, most research on performance-based incentives suggests that it is impossible to evaluate. Wendelin Schnedler's empirical exploration of bonus pay found raised a series of problems with financial incentives for individuals: tasks are too complex to include all measures, unforeseen contingencies emerge which demand responses which don't appear in contract and the need for cooperation and coordination is undermined by purely individual measures. The research come to the conclusion: "Complex tasks will be negatively related to bonus pay."

Most current research suggests that the optimum financial incentive for employees is to pay them enough so that money is off the table as a concern. If they're paid too little it's actually a demotivator. But once they feel respected through the salary, you will struggle to improve performance by offering more.

This reflects something fundamental about human beings and money under capitalism: the confusion between financial reward and relative social status. This is especially apparent in high-paying sectors. Premier Leaguefootballers are known to routinely mock each other for earning more each week. For bankers, bonuses aren't just a payment, but also an appraisal of their career prospects and work reputation. This phenomenon explains why a man who is already a millionaire would risk widespread public hatred for a few more quid. It explains why a well-loved footballer would risk the hatred of the terraces for a few extra thousand a week.

It's not money to them anymore. It's a social marker. It will always drive ever upwards as the individuals in high-paying industries seek to outdo each other. We know where this goes: football clubs that are ready to collapse despite being at the peak of their international popularity. A banking sector which took unacceptably risks and brought the world economy to its knees.

The research all points away from financial incentives. Instead it suggests that employees need three things: autonomy, mastery and purpose. The need to self-improve is what motivates really talented people. If you don't believe that, take a look at the free-to-use software made by engineers. Look at the success of Linux, or Apache, or Wikipedia, as highly-skilled workers use their free time to make something which is then given away.

Bonuses do not improve performance. They worsen it. The banking sector's track record is the proof which defeats its own arguments.