Most performance measurement systems use multiple measures. Target-setting research, such as the budget-ratcheting literature, however, tends to focus only on summary financial measures of performance. This study extends that research to examine the target-setting behaviors used for measures of different types—specifically financial vs. nonfinancial and common vs. unique. We hypothesize that the target variances of the different types of measures are differentially informative about an agent’s effort level and this, in turn, affects a principal’s target revision behavior. The analysis of a proprietary dataset regarding a multi-measure short-term incentive plan used in a conglomerate shows that the conglomerate puts a higher weighting on the information of an agent’s effort level that is revealed in past performance in more informative measures (financial or common). They thus choose not to fully incorporate the information provided by both positive and negative in new targets to avoid the “ratchet effect,” which would cause agents to withhold effort. In contrast, for less informative measures (nonfinancial or unique), it eliminates the effects of positive variances with full upward ratcheting of targets. Negative variances in these measures are penalized because downward adjustments are incomplete.
Performance Target Revisions under Annual Bonus Plans with Multiple Performance Measures
KIM, JONGHWAN;
2015
Abstract
Most performance measurement systems use multiple measures. Target-setting research, such as the budget-ratcheting literature, however, tends to focus only on summary financial measures of performance. This study extends that research to examine the target-setting behaviors used for measures of different types—specifically financial vs. nonfinancial and common vs. unique. We hypothesize that the target variances of the different types of measures are differentially informative about an agent’s effort level and this, in turn, affects a principal’s target revision behavior. The analysis of a proprietary dataset regarding a multi-measure short-term incentive plan used in a conglomerate shows that the conglomerate puts a higher weighting on the information of an agent’s effort level that is revealed in past performance in more informative measures (financial or common). They thus choose not to fully incorporate the information provided by both positive and negative in new targets to avoid the “ratchet effect,” which would cause agents to withhold effort. In contrast, for less informative measures (nonfinancial or unique), it eliminates the effects of positive variances with full upward ratcheting of targets. Negative variances in these measures are penalized because downward adjustments are incomplete.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.