The article reviews research on paid work programs in mental hospitals and reports a new investigation studying the effect of paying patients for work assignments in a short term mental hospital. Significant differences were found between paid experimentals and non-paid controls in the amount of assigned hours actually worked and the work ratings both favoring the experimentals. No differences were found at 6 month follow-up in involvement in training, number of weeks worked, earnings, or rehospitalization rate. In developing program budgets for paid work within hospitals, increases in number of hours worked and ratings when previously unpaid patients are paid, should be taken into account. The literature seems to indicate that workshops can be important additions to mental hospital rehabilitation programs. The meaning or status of work tasks may be significant in effecting changes. Hospitals should develop work programs according to their own unique circumstances.