By IESE Insight
“If it ain’t broke, don’t fix it,” you may have heard.
But if it is “broke”? Of course, you fix it. Any good manager knows that. So, how do you tell if your company’s strategy needs fixing? First you need to measure accurately whether it is working or not.
Accurately assessing performance is, therefore, critical to good management. So, how to evaluate it? One way is to compare current and past performance: the historical approach. Another is to compare it to others’ (often competitors’) performance: the social approach. While comparisons between past and current performance are fairly straightforward to measure, social evaluations are actually quite complex and prone to bias.
In “Self-Assessment, Self-Enhancement, and the Choice of Comparison Organizations for Evaluating Organizational Performance,” Pino G. Audia, Sebastien Brion and Henrich R. Greve unpack a few key motives for choosing certain comparisons in social evaluations. This research reveals that managers may unwittingly have biases in their evaluations: they may veer to self-enhancement, instead of self-assessment, when choosing comparisons, especially when performance is low.
Recognizing the potential to be biased in evaluations can help managers avoid the self-enhancement trap. That is to say, it can help them see, without rose-colored glasses, when a change in strategy is truly necessary.
Social Evaluations: Apples to Apples?
There’s nothing inherently wrong with social evaluations. They help managers gain insight into how well or poorly their companies are doing. For example, the CEO of General Motors may compare her company’s results to Ford, Volkswagen and Toyota for reasonably objective and useful comparisons of car sales and profits in today’s market conditions. Comparing to direct competitors is most like comparing apples-to-apples: it makes sense for granular evaluations.
Naturally enough, most managers take an apples-to-apples approach for social evaluations. But what prompts CEOs and other managers to deviate from this objective standard? Why might a car company CEO look further afield, to household appliances or computer hardware, to make more apples-and-oranges comparisons?
One answer is the self-enhancement motive. Managers may unwittingly select companies for comparisons that put their own performance in a more flattering light. That exposes them to the potentially dangerous side effect of self-flattery: They may then fail to recognize the need for change.
Self-assessment vs. Self-enhancement
In assessing the potential for bias in social evaluations, Audia, Brion and Greve’s research published in Advances in Strategic Management provides support for four points relevant to performance feedback theory — and managerial practice. The four points are:
1. Self-assessment generally prevails over self-enhancement as a motive for managers in performance evaluations. In other words, given the choice of comparing (a) more similar and better performing organizations and (b) less similar and worse performing organizations, decision makers generally choose (a) for more accurate evaluations.
2. But self-enhancement becomes more tempting when performance is low. That is, given the choice of (a) more similar and better performing organizations and (b) less similar and worse performing organizations, decision makers become more likely to go with (b), even though the comparisons are less useful, in the face of low performance.
3. When self-enhancement is a motive (and evaluations are more apples-and-oranges), there’s the potentially dangerous side effect: Managers are less likely to recognize the need to change their current strategy.
4. Finally, mindset matters. If decision makers are primed, unbeknownst to them, to be in a “self-enhancement” mindset with a leading questionnaire and exercise, they are more likely to opt for the more flattering but less useful (b) comparisons. That priming also made them more likely to ignore evidence that their strategy needed to change.
In two role-playing studies, participants were given hypothetical companies to run in a few different circumstances. Specifically, participants took on the role of CEO of the fictitious Allied Waste Industries. They were given background and performance information for Allied Waste and asked a series of questions.
The answers confirmed the four points above. That is to say, participants were more likely to opt for accurate self-assessments, given the choice. But those who received evidence that their performance was low were more likely to veer toward self-enhancing evaluations. These self-enhancers were still in the minority, but the shift was statistically significantly. Critically, the result of these self-enhancing evaluations was that they were less likely to recognize the need for a new corporate strategy in the face of poor results.
The Dangers of Rose-Colored Glasses
“When performance is low, organizational members’ greater reliance on favorability than on similarity when selecting comparison organizations may induce them to form rosier evaluations of low performance than is often assumed,” write Audia, Brion and Greve.
Interestingly, participants were not asked to report their evaluations to an audience. Other studies have pointed to “externally directed impression management” as a motive for flattering evaluations, but that does not appear to be a factor here. Independent of the pressures that individuals may face from important constituents, it appears that the self-enhancement motive may drive biased assessments of performance.
Summing up, the authors emphasize that accurate self-assessment remains most managers’ goal. Yet, other motives do come into play in evaluations and they should be recognized. They write: “Here we take a step toward recognizing these complex effects by focusing on the tension generated by the competing influences of the self-assessment and self-enhancement motives.” Awareness of these biases should help managers better assess the need for change as a result of declining performance.
Extending this research outside of the laboratory, Brion, Audia and IESE’s Horacio Rousseau have begun to examine this phenomenon in the field. In an analysis of archival data from annual reports, they find evidence that the self-enhancement bias is particularly pronounced in powerful CEOs. CEOs who dominate their firms appear to be especially prone to selecting favorable social comparisons. The results are in their working paper titled “Disregarding Legitimate Reference Groups: CEO Power and the Propensity to Use Customized Peer Groups.”