Profit Sharing: Does It Make a Difference?
Kruse's dual focus is the influence of profit sharing on productivity and on employment stability. He concludes that although previous research and his own new findings suggest that profit sharing is associated with higher productivity and greater employment stability, a causal link has not yet been demonstrated. Indeed, throughout the book, Kruse is careful to distinguish between what previous research and his own new findings do and do not tell us about the causal effects of profit sharing.
What are the factors that make causal inferences difficult here? The problems are familiar to anyone who has attempted to draw causal inferences from research that is not experimental, and Kruse does a nice job of discussing how they relate to his study. Consider the issues Kruse encounters when trying to determine whether profit sharing causes increased organization productivity. Potential selection bias exists because organizations cannot be randomly assigned to either the profit sharing or no profit sharing conditions. Therefore, more productive organizations may be more likely to adopt profit sharing because they are in a better position to share profits. Selection bias will also occur to the extent that those organizations that tried profit sharing but later discontinued it because it did not meet their objectives are less likely to be included in the sample. The organizations with "successful" and continuing plans are more likely to find their way into samples. Another threat to accurate causal inference is possible omitted variable bias. For example, unmeasured human resource management changes implemented along with profit sharing may be what really drive apparent profit sharing effects. These possibilities make it difficult to know whether any observed association between profit sharing and productivity is causal.
In the theory section, Kruse discusses the conditions under which profit sharing is expected to increase company performance. Profit sharing is thought to be most useful where the costs of monitoring employee productivity are high. In such cases, piece rates, deferred compensation, efficiency wages, or profit sharing schemes may be used to reduce the need for monitoring of specific behaviors. In agency theory terms, profit sharing is a contracting scheme that links pay to an outcome (profit) rather than specific behaviors (which are costly to monitor). Further, profit sharing is hypothesized to be more successful in companies where employee cooperation is important and where the number of employees covered by the plan is not so large as to reduce the motivation of individual employees, who may begin to see little connection between their effort and overall company performance (or may be tempted to become free riders).
The theory predicting increased employment stability under profit sharing rests on Weitzman's share economy idea. The key prediction is that companies ignore the profit sharing payment part of employee compensation when making short-run employment decisions. Instead, they base such decisions on the fixed wage. If the fixed wage in profit sharing companies is lower than in other companies, profit sharing companies will find it less necessary to reduce employment when profits decrease, because the profit sharing protion is variable and permits them to reduce labor costs without necessarily reducing employment. Profit sharing companies are also expected to be able to increase employment sooner than other companies when profits begin to increase.
Turning to the results on productivity first, Kruse finds that productivity growth in profit sharing companies is 3.5--5.0% higher than in companies not using profit sharing. However, these effects are almost exclusively found in companies using current cash plans, as opposed to companies using deferred plans (as a retirement income vehicle, for example). Another contingency factor is company size. Annual productivity growth in small companies (less than 775 employees) was far larger than in larger companies--11.1--17.2% versus 0--6.9%. This finding is consistent with theory that goes under various names in different disciplines (social loafing, the free rider problem, the 1/N problem). Finally, an important non-finding in the productivity analyses is that no specific human resource practice (such as information sharing) moderated the relationship between profit sharing and productivity. Therefore, the research does not tell companies what other human resource practices fit best with profit sharing and make its success more likely.
The results on employment effects are consistent with much of the previous research and theory on Weitzman's share economy idea and the consequences of flexible pay. Kruse finds that profit sharing is associated with less variability in employment in response to changes in aggregate and company-specific product demand. This finding is strengthened, however, when two contingency factors are taken into account. First, companies that adopted profit sharing after 1975 experienced a larger reduction in employment variability than companies that adopted profit sharing before 1975. Prior to adoption, employment in the "new" profit sharing companies was highly responsive to shifts in demand; afterward, employment changes were significantly less responsive to demand changes. Second, profit sharing led to greater employment stability in companies where profit sharing payments substituted for base pay (as opposed to being "gravy" on top of base pay).
Kruse's study provides some interesting findings, some of them new and some not. The attempt to study contingency factors such as other human resource practices and company size is commendable. For example, the finding that profit sharing is more successful, on average, in small companies than in large companies is an important finding that should be considered by large companies moving to profit sharing. The main drawback of the book, especially for public and private policy decisions, is the uncertainty about whether profit sharing has a causal impact on productivity and employment stability. Obviously, this problem is not specific to Kruse's book and it would be rare for a single study to put such questions to rest. Indeed, Kruse's caution in interpreting his own results is very welcome. The question, then, is how future research should build on Kruse's fine work.
One possibility is to focus on collecting more process information. If profit sharing is theorized to influence company performance through its effects on employee level outcomes such as effort and cooperation, we need to measure effort and cooperation. If the hypothesized linkages between profit sharing, employee level outcomes, and company level outcomes are found, we will have more confidence of a causal link between profit sharing and company performance. A second possibility is to focus more attention on companies that once used profit sharing, but later discontinued the program. What are the reasons? Did it "fail?" If so, why? Perhaps it did not fail. Perhaps it served its purpose for a time and was then replaced by some other program or strategy that was a better fit to changing circumstances. The answers to such questions should be helpful in better understanding whether profit sharing makes a difference.
|Printer friendly Cite/link Email Feedback|
|Article Type:||Book Review|
|Date:||Jan 1, 1995|
|Previous Article:||Emotion in Organizations.|
|Next Article:||Bet on Cowboys, Not Horses: A Technological Breakthrough for Employee Selection.|