2006

Editorial from UGFA Executive:

How Should We Recognize Merit?

Merit pay

The employees who work the hardest and who get the best results need to be rewarded, and the ideal way to reward them that is to pay them more. A merit system is good for the organization as a whole, because financial incentives encourage each employee to perform. These assumptions about merit pay are often taken for granted. But, as we examine them one by one, we find that they are not necessarily valid in the university environment.

Universities are not factories. It’s a fair bet that workers on the General Motors assembly line are motivated mainly by money. Although an auto worker can take some satisfaction in helping to build a nice car, any individual worker’s contribution to the overall success of GM is minuscule. Professors aren’t like that. If they had been motivated primarily by money, they would never have become academics in the first place. Their “outputs” – teaching, research, and service – are individual and intangible. Any professor would much rather receive a Nobel Prize or a national teaching award than a raise. Does it make sense to reward people whose major motivation is not monetary by giving them additional money?

Indeed, we now have major mechanisms for allocating non-monetary compensation to meritorious faculty. For example, the Canada Research Chairs program operates mainly by giving awardees better research opportunities (usually accompanied by some salary rewards). This program did not exist when the present merit system was established at the University of Guelph.

On the assembly line, workers may be hired with little track record or experience, and it is important for management to be able to discriminate between the most capable and the least capable, after they have been hired. But a university is not like that. Professors have already passed innumerable hurdles before they even get a job offer. We should not, and in practice do not, hire under-achieving professors at the University of Guelph. High rates of success in achieving tenure demonstrate this.

How do we evaluate a faculty member’s merit? In a garment factory, or on the GM assembly line, this evaluation may be straightforward. Many employees are performing almost identical assignments, such as sewing on buttons or painting auto bodies, and the tasks hardly change from year to year. One can apply quantitative measures of work output and quality. But how do we do this at a university? Faculty roles and responsibilities are tremendously diverse. Research may be so arcane that there few colleagues are in a suitable position to judge its quality, and the assessment of teaching excellence is notoriously difficult. These considerations don’t make it impossible to assess faculty merit, but they make it very challenging and time-consuming, and lead, at best, to equivocal outcomes.

Merit systems make most sense in organizations where there is a large difference in the performance levels of workers doing similar jobs. I would argue that this is less and less the case at the University of Guelph. We are an excellent university. Our hiring processes are rigorous. Our faculty members are highly qualified and very good at their jobs. As the university has matured, our standards of performance have increased and they have been applied more uniformly,so that the range of variation has narrowed considerably. (This phenomenon is similar to the trend towards more uniform batting averages in major league baseball, as noted by Stephen Jay Gould.)

What is the effect of a university merit system on the organization as a whole? Proponents of the system argue that the effect is positive, because it motivates professors to work harder and perform better; but the considerations noted above lead one to doubt this assumption. On the other hand, the negative effects of the merit system, although rarely acknowledged, are obvious. The evaluation of academic merit (through the TAPSI system) consumes countless hours of work as faculty prepare, read, and debate the quality of CVs and teaching dossiers. As argued above, almost all of our faculty members are highly meritorious, and yet only a fraction are allocated TAPSI awards. This “disconnect” between performance and reward engenders frustration and resentment. Assessments of university merit systems consistently demonstrate that they frustrate efforts to achieve equity: women and members of disadvantaged groups consistently receive, on average, disproportionately small shares of merit awards. Merit systems concentrate power in the hands of department chairs and deans, helping to justify their positions, since they are positioned as the crucial arbiters in evaluating merit.

Under the existing merit system at the University of Guelph, meritorious faculty take “extra steps” up the salary grid. Although the immediate salary increase resulting from a merit award may be relatively small, its cumulative effect can be large. Consider two faculty who begin their careers at grid step 4 and retire after working for 30 years. Prof. A receives single steps only. Prof. B receives an extra step twice, “skipping over” grid steps 7 and 10. At the time of retirement, Prof. A is on step 33; Prof. B has reached step 35 and has a salary 1.7% higher. But Prof. B’s career earnings have exceeded Prof. A’s by $73,179. (Calculation based on the present grid). Of course, with greater accumulated step differentials based on merit, and with increasing overall salaries, these differences could become very much larger.

And consider the following anomaly: Prof. C, like Prof. B, receives an extra step twice and retires at step 35. But Prof. C works in a field where scholarly careers develop more slowly; she is a “late-bloomer”; she skips over grid steps 17 and 20 rather than steps 7 and 10. Prof. C’s career earnings exceed Prof. A’s by only $32,524, because her merit became apparent a little later in her career.

Such accruing differentials in the salaries of faculty doing essentially the same jobs are known as “anomalies”, and some universities set up discrete “pools” of salary money used to reduce anomalies. Is it sensible up a “merit” system that creates anomalies alongside an “anomaly pool” system to mitigate them? This fundamental inequity results from the cumulative nature of the merit/grid system. Superior performance by a faculty member at the age of, say, 40 is rewarded over and over again until he or she retires. Ironically, this may promote a “resting on one’s laurels” attitude: “I’m already so high up on the grid, what’s the point of continuing to climb?”

So, should we eliminate the reward of merit? I would say No. Instead, we should adopt a system that breaks the rigid connection between merit steps and salaries. We should treat all faculty members equally, in terms of progress up the salary grid. Merit should be recompensed by flexible, appropriate, limited rewards. These could include one-time (non-base) salary bonuses. But, with some imagination, we could expand the “rewards” list to include, for example, earlier access to sabbatical leave; reductions in administrative assignments; stipends for graduate students or research assistants; enhanced Professional Development Reimbursements, etc. In this way, productivity would be encouraged and rewarded “in kind”, by providing faculty with the tools they need to enhance their scholarly activities, and without generating long-term substantial salary anomalies. Another desirable feature of a revamped merit system emphasizing one-time recognition of superior performance is that the rewards can be highly visible,
in contrast to the operation of the TAPSI system, which is shrouded in secrecy.

What are your thoughts on this situation? Please call the Association and let us know.

“Everybody assumes that 400 hitting disappeared because hitting has gotten worse. ... [But] the average batting average has ... always been around 260. ... 260 is a balance between hitting and pitching. It’s been maintained as everyone has gotten better. Hitting’s gotten better. Pitching’s gotten better. Everything’s gotten better. The balance remains the same. Now as everything gets better, the variation shrinks.”
(Stephen Jay Gould, 1996)