It is common for organizations to offer awards to motivate individual behavior, yet few empirical studies evaluate their effectiveness in the field. We report a randomized field experiment (N = 15,329) that tests the impact of two common types of symbolic awards: pre-announced awards (prospective) and surprise awards (retrospective). The context is U.S. schools, where we explore how awards motivate student attendance. Contrary to our pre-registered hypotheses and organizational leaders' expectations, the prospective awards did not on average improve behavior, and the retrospective awards "decreased" subsequent attendance. Moreover, we find a significant negative effect on attendance after prospective incentives were removed, which points to a crowding-out effect. Survey experiments probing the mechanisms suggest that awards may cause these unintended effects by inadvertently signaling that the target behavior (perfect attendance) is neither the social norm nor institutionally expected. In addition, receiving the retrospective award suggests to recipients that they have already outperformed the norm and what was expected of them, hence licensing them to miss school. Exploratory analyses shed further light on differential effects of awards by age and performance. [This is the in press version of an article published in "Organizational Behavior and Human Decision Processes" (ISSN 0749-5978).]