Performance Related Pay: Does It Stack Up?

I’ve only once been involved in the design of a Performance Related Pay (PRP) scheme. I’d joined a new organisation as a director and discovered that there was no incentive scheme for the sales team. It’s received wisdom that all sales teams have some sort of bonus scheme isn’t it, so I pulled out my old MBA notes and got to work.

Vroom’s expectancy theory seemed relevant:

Motivation = E x I x V

where E = expectation that effort will result in achievement of performance goal;
I = expectation that goal achievement will be rewarded;
V = value of reward to worker.

I carefully designed the scheme around this, and waited for the enhanced performance to kick in.

Nothing. In fact if anything, a slight dip.

At the time, I never quite got my head around why the scheme wasn’t effective but now the book ‘Predictably Irrational’ [1] by Dan Ariely, Professor of Behavioural Economics at Duke University, has made sense of it for me.

He says “we live in two worlds: one characterised by social exchanges and the other characterized by market exchanges. And we apply different norms to these two kinds of relationships. Moreover, introducing market norms into social exchanges… violates the social norms and damages the relationship.”

In one research project, Ariely and a colleague investigated the productivity of students dragging a circle across a screen with a mouse for a five-minute period. Some of the participants received five dollars for participating in the experiment, and others 50 cents. They introduced the task to the third group as a social request, a favour. They found that those who received five dollars dragged on average 159 circles, those who received 50 cents 101 circles.

So far, so predictable. But it turns out that those who worked under the non-monetary social norms dragged 168. Other similar experiments demonstrated that when participants were primed to think about money (by putting a value on the gift they were to receive at the end of an experiment), they were less productive.

When Performance Related Pay Doesn’t Pay

Back to my sales team. They were embedded within an organisation that had a strong social culture: good relationships, trust, authenticity, laughter, a lack of hierarchy. There was strong social contract: employees knew they could rely on the company to step up when issues arose around childcare, or healthcare.

In those circumstances, people are able to think clearly. They are less prone to the high emotions that saturate the limbic system and switch off our higher order thinking, but instead are more likely to enter a state of flow, a feeling of energised focus. Incidentally, Ariely also found that the stress of performing in front of others can also reduce productivity on tasks that are complex.

When I introducing market norms into the relationship, it only served to undermine the social contract. Ariely compares it to offering to pay for the Christmas dinner that your mother-in-law has just cooked for you.

If you work within an organisation where there is a strong social contract, it makes more sense to deploy benefits that serve to reinforce the culture, such as employee wellbeing days, child care, pensions, and flexible working. (Interestingly these are precisely the type of benefits offered by many leading tech companies.)

When Does Performance Related Pay Pay?

Obviously, it’s worth thinking twice before introducing PRP where there is a strong social contract, but does it ever make sense?

There are some environments where the relationship beween employer and employee is more transactional, say in a large warehouse for an online retailer or a merchant bank. If there is no social contract, surely money then becomes more useful as a motivator?

Though Ariely doesn’t cite any research on the topic, a consideration of the underlying psychology suggests that this is probably the case for mechanical, repetitive tasks: if higher order thinking is not required, productivity is less likely to be impaired by stress. And this is maybe where Vroom’s expectancy theorem has some merit, though it ignores the morality of subjecting your staff to incessant high stress, and also the impact of such stress on health and staff turnover in the long term.

Ariely conducted another experiment offering different levels of bonus for an array of tasks that demanded attention, memory, concentration, and creativity. He discovered that those that the productivity of those receiving a high bonus ($600) for a 4-minute task was lower than those receiving a medium bonus ($60) for the same task. He concluded that the stress associated with the prospect of a higher bonus can actually impair the ability of people to think clearly and execute the task efficiently.

Not surprisingly, Ariely did not get a particularly positive response from a group of banking executives when he offered to investigate the effectiveness of sky high bonuses on performance in the banking sector. However, his research suggests that it is a bad idea, even before the moral hazard that it often gives rise to is taken into account.

Summary

I’d offer the following matrix as a summary the appropriateness of PRP as a motivator in different circumstances.

In a culture increasingly defined by market norms — “everything and everyone has their price” — PRP seems to be making a comeback. Ariely’s research suggests that in adopting it, companies are forgetting that it originated in the workplace of the Victorian era, where the majority of tasks were mechanical and the employer-employee relationship transactional.

Even then it’s effectiveness is limited. It may have harmful long-term effects, and where tasks demand attention, memory, concentration, and creativity, it can actually impair performance beyond a certain level.

Vroom’s expectancy theory is simply not adequate for the majority of today’s industries. Insights from neuroscience and the research of academics such as Ariely tell us that if we need independent thinking from our colleagues then a convivial, compassionate culture is the way to go.

References

Ariely D (2009), Predictably Irrational, Expanded Edition, London: Thorsons