A presentation of findings from a study performed for the California Department of Benefit Payments (DBP). A mail and telephone survey was conducted of all state welfare departments; in-depth data collection, including field visits, was made in six states; the suitability of other states' methods for California was analyzed; an extensive literature search was made; and 25 most promising methodologies were reviewed. Study conclusions include: (1) Promising methods exist that would enhance DBP's caseload and expenditure forecasting capability, particularly with regard to (a) anticipating the impact of changes in economic conditions, demographic trends, and behavioral patterns, and (b) explaining caseload movement during budget preparation. (2) DBP should first test on current California data the most promising techniques from other studies. (3) DBP should be sure that all input data needed to compute caseload forecasts are either directly available or can be reasonably well estimated by methods incorporated in the procedure selected. The survey package, summaries of state contracts and related literature, and an extensive bibliography are included.