A simple starting point for analyses of state mental health policy is to view state policy decisions as being the result of choices by a social planner seeking to coordinate mental health services for the poor. State policy makers have two major instruments: direct state spending on services, and the insurance-like 66 Medicaid program. Michael (1980) and Frank (1985) regard the “state” as a single decisionmaker with an objective function containing: welfare for the poor and state budget costs. The choices of the state planner are constrained by factors such as the income in the state, the size of the public mental hospital system, the availability of alternatives to state funded providers (e.g. nursing homes), federal rules governing Medicaid especially the federal matching rate on spending,31 and the amount of private insurance coverage in a state. Direct state spending on mental health care will be reduced, according to this approach, by generous federal matching provisions,32 the availability of care in settings funded by Medicaid (nursing homes), and expansion of private insurance coverage for mental health care. These all point to rational cost shifting responses to exogenous changes in regulation and market structure. Although few formal analyses of mental health financing for countries other than the U.S. have appeared in the literature, similar observations have been made about the relation of central government financing to local funds. For example, Britain has experimented with central government matching grants to local authorities (Knapp, 1990; Yellowlees, 1990).