//
you're reading...
Uncategorized

The usefulness of performance measures over time

What is the value of performance measures over time?

As time goes by, measures improve but performance doesn’t. People get better at improving the measures but no performance. Why is that?

We had a discussion about this paradoxical phenomenon recently at the CBP. Monica Franco-Santos, one of the members of the CBP, neatly summarizes the reasons in a book she co-authored with Luis Gomez-Mejia & Pascual Berrone.

Here’s her explanation:

“The usefulness of performance measures decays over time due to their loss in variability (Meyer and Gupta, 1994). In other words, the diminished variance of performance measures reduces their ability to differentiate “good” from “bad” performance as time goes by, which in turn reduces its validity (Franco-Santos, 2008). For instance, Earnings per Share (EPS) was a performance measure commonly used in executive incentive systems in the early 1970s. This measure soon became irrelevant as it was no longer effective in driving the expected behavior. Executives found a way of getting the expected EPS results. High levels of EPS were achieved but this measure no longer differentiated good performing companies from bad performing companies. Although the measure was appropriate when first used, it became distorted as time went by (Meyer and Gupta, 1994).

Meyer and Gupta (1994) suggest that the usefulness of performance measures decays over time due to four causes: positive learning, perverse learning, selection or suppression. Positive learning refers to the fact that, over time, people learn how to improve their performance in whatever they are being measured on. Perverse learning or gaming refers to the idea that people learn how to meet the specific requirements of a performance measure without improving the performance that is sought. Selection refers to the fact that organizations learn over time, either positively or perversely, how their employees perform and decide to replace low performers with high performers. Finally, suppression refers to the indication that performance measures are suppressed when performance differences cannot be diminished by either improvement, the appearance of improvement, or selection.”

From the book “Compensation and Organizational Performance: Theory, Research, and Practice”, by Luis Gomez-Mejia, Pascual Berrone, & Monica Franco-Santos, written in 2010.

Advertisements

Discussion

3 thoughts on “The usefulness of performance measures over time

  1. This is a phenomena lots of us forget, or wish to forget about. The Royal Statistical Society’s 2005 report on Performance Measures – the good the bad and the ugly has some good examples of how this happens in practice.

    Posted by Mike Bourne | June 10, 2013, 2:52 pm
  2. On Radio 4 this morning we heard the Minister discussing the new schools curriculum with the head of the teaching union. One of the statistics that came out was that the percentage of students getting A and A* grades has risen from 8% to 22% over the years. The Minister was arguing that this was grade inflation and was arguing for more rigour. From the teachers’ perspective, the union was arguing that teachers had become better at how they taught the curriculum.

    An example of a performance measure (student grades) losing their usefulness over time?

    The question that should be asked is “what are student grades for?” Are they to allow us to differentiate between the different abilities of the students, show who is better and who is not as good? Are they to show how well students are doing over time against some objective measure of learning? Are they to assess improvements in the quality of teaching?

    My comment would be that they can’t achieve all three objectives at the same time. So if we want to differentiate between students we will need to develop the measure differently to what we would use to measure the improvement in teaching.

    Posted by Prof Mike Bourne | June 11, 2013, 12:48 pm

Trackbacks/Pingbacks

  1. Pingback: JUHTIMISIDEED: Tulemusmõõdikute head ja vead | Pärnu Konverentsid - September 17, 2013

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: