“All work deserves payment”, as the saying goes, and one of the key aspects of successful organizations is their ability to reward each member based on his or her actual contribution. This implies, on the one hand, a causal link between the objectives defined and remuneration, and therefore a constraint: if the objectives are achieved, then remuneration follows, with its amount depending on the proportion of objectives achieved; otherwise, a possible punishment can be put in place. On the other hand, this implies that the manager has perfect visibility of individual performance.
But is this type of performance-related remuneration actually effective? Or does constraint-free remuneration, which simply means giving the same salary to all individuals performing the same task, generate improved performance? Recent research in behavioral sciences suggests that this form of constraint-free managerial assessment, the simplest, is genuinely effective in terms of performance.
The power of “influence costs”
The main problem comes simply from what are known as “influence costs”. In order to be able to compensate employees based on their performance, the manager needs to have precise information on individual achievements, which is not always easy to obtain. The manager must therefore rely in part on employees’ accounts of their own contributions; however these accounts are likely to be biased.
Since presenting oneself in a positive light generally brings financial benefits, employees are likely to engage in “window-dressing” activities, to give the impression that they are key contributors. These activities, which have been widely documented in the financial and accounting literature, generally involve concealing relevant information, falsifying documents, or manipulating or blatantly influencing managers’ opinions.
Regardless of the strategy used to distort performance measures, the organization will be negatively affected by these manipulation activities, as they induce waste of time and distort incentive programs. This is what is referred to as influence cost. Manipulating performance will reduce the quality of information available to managers, which will ultimately weaken the correlation between compensation and performance measures.
Performance manipulation is therefore a possible explanation for the surprisingly limited strength of incentives in actual employment contracts. It may therefore be optimal for organizations to limit managers’ discretionary power over decisions affecting the distribution of resources.
This means that companies may adopt apparently unnecessary bureaucratic rules for reasons of efficiency. For example, companies can avoid giving bonuses to limit “influence activities”. The widespread use of equal pay could therefore be justified, despite its negative incentive effects, as a means of limiting influence costs.
The advantages of equal pay
Given that performance manipulation can only be effective if it remains hidden from management and that such activities may be severely punished, it is not surprising that no reliable archival data can be obtained on the topic. We therefore set ourselves the objective of collecting data to highlight the trade-off between the positive incentive effect of performance-related pay and its detrimental effect on promoting window dressing activities..
To produce data on performance manipulation in an organizational context and thus fill empirical gaps in the literature, we created a laboratory work environment in which workers could engage in a time-consuming activity designed to manipulate the way their performance was perceived, instead of actually making a real effort.
We show that in an organization with no manager, where window-dressing activities were not available, the work observed was equal to the actual contribution. On the other hand, when window-dressing activities were possible, employees made use of them and demanded a larger share of organizational output.
These activities were much more common with performance-related pay than with equal pay. As a result, organizational output was significantly higher with equal pay than with performance-related pay. This means that the beneficial effect of equal pay on deterring window-dressing activities more than offsets the negative incentive effect.
We also created a virtual organization in which individuals worked under the supervision of a manager in order to assess the extent to which managers could reduce performance manipulation by adjusting employee incentive programs. To this end, we created a laboratory work environment in which managers could both monitor employee performance and set employee compensation as they saw fit.
This experiment generated the same overall results: organizational output was much lower when window-dressing activities were available. However, employees adopted more sophisticated manipulation techniques: window dressing was minimal, with employees exaggerating their performance measurement as little as possible in order to avoid detection.
It is thus clear that intensive monitoring by the manager combined with weak incentive systems can effectively reduce the extent of performance manipulation activities. This in turn makes the company more efficient. But remember not to confuse equal pay with low pay! Employees must first and foremost understand that their work is valued.