Every day brings fresh evidence that the most dangerous word in the management lexicon is ‘incentives’. Incentives are a star instance of management’s besetting sin, its urge to impose simple solutions on complex issues. It’s hard to overestimate the consequences of their resulting abuse. To quote Mihir Desai in that well-known radical organ HBR, today’s incentives and rewards have ‘contributed significantly to the twin crises of modern American capitalism: repeated governance failures... and rising income inequality’. Others would add at least some blame for the current lack of global growth and companies’ dismal record of job creation to the charge sheet.
Oversimplification in incentives is endemic and fractal, distorting outomes at every level.
The flaw inherent in all outcomes-based management is that history runs forward, not back. While it seems attractive to manage back from incentives attached to desired outcomes, it can only be done by sleight of hand and optical illusion, as in the alluring but impossible constructions of M.C. Escher. In the real world results are always context-dependent and often hard to measure (over what time frame? At what cost to other parts of the organisation – eg VW?). What’s more, since they are the emergent product of a complex system with lots of moving parts, it’s also near impossible to establish a linear chain of cause and effect. Hence a kind of social science uncertainty principle: the more we know about the outcome the more complex it becomes and the less we are able to attribute it to a particular cause or person. In this situation incentives simply make no sense.
In such a context, the problem is not that incentives don’t work. It’s that they do – just in a way that was entirely foreseen. As systems guru Russ Ackoff never tired of pointing out, organisations being systems, their problems ‘are almost always the product of interactions of parts, never the action of a single part. Treating a single part destabilises the whole and demands more fruitless management intervention; management becomes a consumer of energy, rather than a creator.’ Incentives are a classic example of treating a part as a whole. Then, like league tables and targets, incentives further step down multidimensional performance into a few simplistic targets. Unfortuately, 'focusing on a small number of these dimensions as targets directs attention on these at the expense of others of equal importance,’ notes John Kay, ‘The effect of the target is to focus attention on the target rather than its purposes. The target is met; the underlying objective remains elusive.’
But even if incentives didn't fail conceptually, they don't work on so many practical, empirical scores that it would be comical if the consequences weren’t so dangerous.
Innumerable studies show that money doesn’t motivate. Incentives based on them doubly don’t motivate, except to make people demand incentives. Their effect is only temporary, so even if they did work they would have to be increased at a prohibitively accelerating rate (which is exactly what has happened to CEO pay) to be any use.
Financial incentives can’t make people cleverer nor (by definition) more ethical. Not only do they not result in better or more work except in the very simplest piecework situations – the relationship is often negative. One powerful reason is hubris: the higher the pay and the incentive, the greater the reinforcement of narcissist and overconfident tendencies, leading to overinvestment particularly in value-destroying mergers and acquisitions (think RBS in the run-up to the Great Financial Crash). As Berkshire Hathaway’s Charlie Munger once put it, ‘II’ve been in the top five per cent of my age cohort all my life in understanding the power of incentives, and all my life I’ve underestimated it. And never a year passes but I get some surprise that pushes my limit a little farther.’ That’s a good reason, says Nicholas Nassim Taleb, author of The Black Swan, for not offering incentives ‘to those who manage risky establishments such as nuclear plants and banks… Society gives its greatest risk-management task to the military, but soldiers don't get bonuses’.
Despite all the refuting evidence the incentive sledgehammer is still routinely used to bash almost any performance management issue, nut or not. This is partly for self-serving and psychological reasons, but also because incentives are obstinately seen as a technical issue that can be ‘solved’ if only they are done better, whereas in fact the problems are intrinsic to the thing itself. In other words, incentives are the problem, which can therefore only be resolved at a different level.
How? It has been known for aeons that while pay doesn’t motivate, it most reliably demotivates – the most powerful emotions relating to money being anger and anxiety, according to psychologist Adrian Furnham. So the best thing to do is as far as possible to take money off the table, allowing workers to focus on the job rather than the reward and managers to spend their time and energy improving the system rather pacifying the discontented (see Ackoff above). Fire remuneration consultants (another saving) and devise a scheme based on pay ratios, say 50 from top to bottom, that will satisfy the strong human desire for fairness, not to mention the empirical finding that less widely dispersed pay is associated with better performance, and connect the measures they use to purpose – in other words, give them a good job to do. James Treybig, the founder of Tandem Computers in the US, carefully recruited people by attitude with the promise only of a ‘competitive’ reward package on selection. If they insisted on knowing the exact salary they didn't get the job. If they came for money, reasoned Treybig, they'd leave for it, too.
In a list of warning maxims for management, incentives perfectly illustrate the top two: 'be careful what you wish for', and 'you can’t solve a problem by doing better the thing that was at the root of the problem in the first place'. As Peter Drucker earily used to say, 'There is nothing so useless as doing efficiently that which should not be done at all.’