Friday, November 26, 2010

Pay for Performance

As I perused the morning newspaper, I came across an interesting article. In it, the author claims that Charlotte-area companies are now beginning to link year-end bonuses to company or individual performance. Gasp . . . what a concept. Linking pay to performance in the private sector. What will they think of next? The author states:
As the unsteady economy continues to pressure businesses in Charlotte and across the country, many year-end bonuses have disappeared, going the way of pay raises, 401(k) matches and lavish holiday parties. The ones that remain are often based on performance, paid only if the worker or company does well.
This year, there are hints that bonuses are returning. And experts say those performance-based incentives will be most common, as companies look for ways to keep employees engaged while keeping the company afloat.
The article drew me back to the recent yakking in the paper and elsewhere about this concept of pay for performance in the education field—paying teachers more only if they perform well—emulating, of course, that paragon of virtue, the private sector. But, you mean the private sector hasn’t always linked pay to performance? You mean, they just gave out bonuses, regardless of performance??? Like paying huge bonuses to auto CEOs, even when their company was teetering on the edge of bankruptcy. Or paying out large bonuses to banking and investment rogue CEOs, even after their firms had collapsed because of their crazed performance, and required the public (that’s you and me) to bail them out, lest the world should implode. Yeah, that private sector. The paragon of virtue that is apparently without virtue, and where all the CEOs wear no clothes . . . ya’know, like the Emperor.
It also drew me back to days of yore, when I worked for the Government during the Carter & Reagan administrations—when they were implementing Civil Service Reform and Pay for Performance for public sector employees. I was remembering that the experiment didn’t turn out too well. Oh, they put it into practice, and bonuses were paid out to the senior executives alright. But, as it turns out, few can discern the difference between high performance and average performance.
The entire concept of performance based pay, or even the broader topic of performance appraisal, is a topic worthy of study, but often, it seems, the systems are being implemented before any serious study has been accomplished. It turns out that pay for performance would require us to understand performance in more sophisticated ways than we seem capable. While an assembly line factory worker performing an entirely repetitive task could arguably be appraised on the basis of how well he/she carried out the task (that’s what industrial engineers have been doing for perhaps 100 years), how is one to assess the teacher of a mixed group of 30 fourth graders? Or how would one assess the performance of a biomedical researcher working on a new vaccine? Or how might one assess the performance of a design engineer working in a group of 150 other design engineers all working to produce a new design for a missile system? How might we assess the performance of a “journalist” who writes a weekly column for the Washington Post? Or how about the performance of an investment banking CEO when his company nears financial collapse?
It is true that there are relatively mechanical ways to assess performance of people such as the above. The teacher could be assessed on the basis of the performance of the students on standardized tests; the biomedical researcher on the basis of how well the vaccine eventually produced works to eliminate a particular virus (perhaps delaying his/her payment for ten years until the vaccine is developed and tested). Or the engineer could be assessed if he produced "high quality" design drawings according to an arbitrary schedule. We could, of course, assess the banker based on how well the overall bank portfolio performs for his clients, but, hahahaha, that would much too sensible.
All of the obvious approaches present problems because of external variables outside the control of the person being assessed. When we used to carry out performance evaluations of social or economic development programs, we used a thing called a “logical framework” to lay out the essential elements of the design and the evaluation schema. The framework laid out the series of cause and effect means and ends that would lead from a person’s specific work input (brains being applied to a problem) to an intended outcome. But after laying out that logic, we would add an element at each stage of the input to outcome logic. That addition was the question: “suppose you do everything correctly, what other external factors might weigh in to frustrate the achievement of the desired outcome?" The reason for adding that question was to see whether it might be possible to bring those external factors within the control of the program? If not, then at a minimum, one would want to measure those external factors to determine how they were affecting the achievement levels expected. In teaching that group of fourth graders, for example, the home situation of the kids might be such a factor. If the kid’s life at home is chaotic, the learning process will be negatively affected, and the teacher might not be able to work around that problem.
But in real life today, we don’t work that hard to develop our plans for performance pay. We measure what is easy to measure, as distinct from measuring that which is important. Or, in the case of the world of commerce—that highly vaunted private sector, we don’t bother to measure at all. We just put the bonuses into play, regardless of performance.
So, in the midst of all the chaos of today’s mixed up economy, it’s nice to see the private sector rethinking its bonus system. Maybe there’s light at the end of this very long and very dark tunnel after all. And perhaps we might envision the possibility of some bright soul rethinking the pay for performance concept as applied to our teachers. Couldn't hurt.
Post a Comment