abstract: In this study, I examine the extent to which firms rely on relative performance evaluation (RPE) when setting executive compensation. In particular, I examine whether firms use information about peer performance to determine compensation at the end of the year, i.e. after both firm and peer performance are observed. I find that RPE is most pronounced for firms that allow little or no scope for ex post subjective adjustments to annual bonuses. Conversely, firms that rely mainly on subjectivity in determining bonus exhibit little use of RPE. These findings suggest that information about peer performance is not used at the end of the year. Instead, peer performance seems to be incorporated in performance targets at the beginning of the year, at least among firms primarily using objective performance measurements. In addition, I provide new evidence on the determinants of the use of subjectivity. / Dissertation/Thesis / Ph.D. Accountancy 2013

This study examines whether eliciting self-evaluations increases or decreases the propensity of a productive agent to retaliate against an employer for paying compensation that the agent perceives to be too low for the work performed. Specifically, I consider a setting in which a principal knows more about the agent’s production than even the agent can observe. In such a setting, an agent might perceive that s/he is being underpaid if the principal pays less than the agent believes s/he deserves, especially if the agent is overconfident about his/her own productive ability. Such an agent could take retaliatory actions against the principal that would be costly to both parties. Self-evaluations could mitigate such tendencies if they result in compensation that is more aligned with agent self-perceptions. Alternatively, self-evaluations could worsen such tendencies if they reinforce the perceived inequity of compensation that does not match agent self-perceptions. I present experimental evidence from comparing a control condition without self-evaluations to three different forms of self-evaluation reports, finding evidence consistent with the premise that self-evaluations increase retaliatory actions and lower welfare. My findings show a cost to self-evaluations that thus far has not been sufficiently considered in the literature. / text