Editor’s note: ACCP is collaborating with LeaderPoint in bringing you a series of articles on popular topics in leadership and management. For information on the upcoming 2008 ACCP Leadership Experience in June, please visit http://www.leaderpoint.biz/accp.htm. Registration for the June course is open.

Can a manager really affect performance without being involved in a subordinate’s work? Thirty years ago, Arthur Elliott Carlisle, a business professor at the University of Massachusetts at Amherst, published "MacGregor,"¹ an article written in unconventional parable form, in the journal Organizational Dynamics. MacGregor (a pseudonym) was the head of an oil refinery who routinely ran the most profitable plant in the company without spending time in the plant, without solving subordinates’ problems, and without telling them how to meet their targets. And he didn’t spend time on formal employee performance evaluations.

In previous ACCP Report articles, we discussed the distinctions between performance and results and how feedback on results, not performance, is an important measurement tool for managers. Yet managers do rely on the performance of those who do the work. In this article, we explore answers to two questions: “Who evaluates performance?” and “How can managers support performance improvement?”

Because performance occurs as people work, usually only one party is privy to that performance and can effectively evaluate it: the individual doing the work. MacGregor never evaluated the actual performance of his subordinates. In fact, when they tried to get him to intervene in their work, he resisted. As Carlisle explains, “In all cases MacGregor left specifics on how agreed-upon results were to be achieved to the subordinates themselves.” MacGregor’s “performance evaluation” consisted of providing weekly feedback on pertinent results so that his refinery managers could reflect on their performance, make changes, and improve those results.

Consider the phenomenon of New United Motor Manufacturing, Inc. (NUMMI). A failing General Motors plant in Fremont, CA, was closed in 1982 because of the dismal performance of its unionized workforce. Shortly after the closing, NUMMI was formed as a joint venture with Toyota. In hiring back 85% of the same “failed” workforce, NUMMI went from 20% absenteeism to 3% – and significantly improved quality. The plant became the most productive plant for GM, and in 1998, it won the Award for Excellence from the National Association of Manufacturers.²

The NUMMI success, like MacGregor’s, can be attributed to an informal performance evaluation system that allowed those doing the work to assess their own performance. By routinely receiving and reviewing data on measurements such as attendance records and defect records (feedback on results), assembly teams not only improved their methods, but individual workers were also able to evaluate and modify their own performance – and see firsthand the effect on results.

The key, however, is to distinguish between a formal employee evaluation and an informal performance evaluation. A critical difference in NUMMI’s production system (and similarly lean production systems) is that any worker on the assembly line has the authority to stop the line to correct problems. Such a policy seems counterintuitive, given that every minute of downtime costs $15,000. But that downtime can signal learning on the part of the workers, enabling them to innovate in their methods and improve their performance. As one manager said, “When there’s no downtime, I know that my people are sending junk through or they’re trying to be superstars.” Although formal employee evaluations are often counterproductive, informal performance evaluation systems promote learning.

MacGregor admitted that he would allow relatively small mistakes by his subordinates because they would learn from them. Carlisle sums up his assessment of MacGregor’s performance system in terms of learning:

MacGregor’s overriding concern was with results: the results his subordinates achieved through methods they developed either by themselves or by working with their peers. He simply refused to do their work for them, even at the risk of incurring short-run costs. By refusing, he enabled them to grow in terms of their ability to make decisions even under conditions of uncertainty.

Managers such as MacGregor recognize that one of the best ways for workers to improve performance is to use their peers. MacGregor held a weekly meeting of his subordinates, which served as a platform for peers to give help to each other. Subordinates reported that if they went to MacGregor for help, he would refer them to their peers in the plant, who were “in touch with what goes on out there.”

Performance evaluation systems that promote learning are worth exploring and implementing. Ideally, as with MacGregor and NUMMI, individuals – with the help of their peers – evaluate and improve performance. Managers use this opportunity to create and maintain a system that pushes the responsibility for performance evaluation and improvement to the individual, encourages peer support, and provides constant feedback on results.

References:

1. Carlisle AE. MacGregor. Organ Dyn 1976;5:50–62.
2. O’Reilly CA, Pfeffer J. Hidden Value: How Great Companies Achieve Extraordinary Results with Ordinary People. Princeton, NJ: Harvard Business Press, 2000.