Perverse Incentives: The Limits of Metric-Based Evaluation

As a manager, how should you evaluate performance? The recent trend is toward ever-more-sophisticated formal models. However, in the wake of several scandals, IESE's Natàlia Cugueró-Escofet and Josep M. Rosanas point to the dangers of modelling and rigid metrics. Instead, they argue, companies desperately need to rediscover informal methods of control.

A not-unfamiliar situation: A mortgage salesman strives for bonuses and recognition at work based on the dollar value of mortgages he sells. He sells as many mortgages as he can. The bank is happy with its revenues. The salesman is well rewarded.

But when the bottom falls out of the industry, it turns out that many of those mortgages should never have been approved. The clients cannot repay and the bank is soon in trouble.

Where does the fault lie? With the salesman, who has incentives to approve risky mortgages? Or with management, for relying on rigid performance metrics that encourage risky behavior? In an article published in the Journal of Business Ethics, IESE's Natàlia Cugueró-Escofet and Josep M. Rosanas weigh in and question the ethics of metrics.

Manage by numbers

Staff need to be appraised, but how? Over the past few decades, companies have increasingly looked to rigid performance measurements, such as key performance indicators (KPIs) or the balanced scorecard (BSC). But these metrics don't account for factors such as trust and common sense, and, as a result, managerial discretion and professionalism are routinely undervalued. This can create serious problems in the long-run.

In their article, Cugueró-Escofet and Rosanas investigate the effects of over-reliance on formal evaluation criteria. They explain how it tends to lead to "gaming the system," to the company's detriment.

And the solution? More flexibility. The performance metrics need room to breathe, and managers must be able to look to "informal justice."

A damaging shift

Once upon a time, performance was evaluated using a mix of formal targets and informal controls (company culture, managerial discretion, personal relationships and so on). In recent years however, management control literature has promoted increasingly more sophisticated, agency-based methods.

This means rating managerial performance against a number of objectively measurable indicators tied to the corporation's strategic agenda, often described as the balanced-scorecard approach.

Informal evaluation, meanwhile, has been largely overlooked. In the case of mortgage lenders at a mid-sized Spanish bank, a manager who opposed the risky over-lending practices was "promoted" sideways, to a better office with a higher salary, but out of a decision-making role. In a similar case at a large bank, a board member with a strong grasp of accounting felt that he couldn't speak up about over-valued assets, as the compensation of the whole board depended upon profitability, and he did not want to have everybody against him.

The results were catastrophic for the banks, which had to be bailed out from economic collapse in 2008. The very rewards system they had installed had incentivized workers to act contrary to the companies' best interests.

Rediscovering flexibility

It's not just a case of the wrong metrics being applied, Cugueró-Escofet and Rosanas argue. No set of metrics can anticipate every possible contingency. And rigid evaluation policies can cause even those workers who do have the company's best interests at heart to lose faith: they may leave the company, or shift priorities for personal betterment when they see others rewarded.

Instead, it's high time to bring informal methods back into management control. This means re-engaging with less regimented forms of motivation and evaluation, investing in staff so that they care about the company's success, and increasing the role of discretion and common sense in evaluations, while recognizing the limits of metrics. Informal forces can include concepts like "work ethic," as well as company culture and management attitudes toward subordinates.

These soft management practices have to be applied with what the authors call "informal justice," whereby managers instinctively correct for deficiencies in the formal metrics and feel secure making suggestions to help improve them.

Two types of justice

The co-authors briefly present the formal management control systems of four selected case studies regarding Spanish banks, an IT company and fashion retailers. They then define and distinguish "formal justice" and "informal justice" and look for the presence (or absence) of both in their analysis. For short and long term ethical development, they posit that informal justice is the key to improve an unjust situation.

In conclusion, they argue that shifting to exclusively formal evaluations has been a mistake, and many companies and countries are still paying for it. "Formal justice", derived from formal systems, may be absent; in which case it is "informal justice" that can help in two ways: by working as a corrective when metrics go wrong and by putting pressure to change the formal system for the future to obtain better alignment between company and staff interests, thereby preventing future crises.