This chapter, from a book project on the politics of mental health, compares processes of psychiatric deinstitutionalization across countries. Despite the movement's international momentum and ongoing support, individual countries have pursued very different approaches to this reform. To demonstrate, I first use the tools of concept formation in political science to develop a portable definition of psychiatric deinstitutionalization; then I use its indicators to systematically measure cross-national trends in mental health service provision. The data set includes observations from 16 affluent democracies from 1935 to the present. These countries represent the first movers, societies whose early industrialization prompted the rise of the asylum and whose extraordinary economic prosperity in the twentieth century prompted its decline. The experiences of these cases framed global mental health policy norms. The data are drawn from these countries' national statistical yearbooks, yearly health reports, and other primary sources.

The data set reveals several features of deinstitutionalization and mental health policy, which to date that have been overlooked in the existing literature. First, not all countries deinstitutionalized in the same way. Although all western societies reduced the population of hospital residents, only some subsequently closed those hospitals. Others re-purposed them. A second and related finding shows that this latter group of countries expanded non-hospital care to a greater extent than the former group. Contrary to the expectations of the deinstitutionalization movement, the expansion of non-hospital, community-based care does not require the wholesale closure of mental hospitals. In fact, the opposite may be true: hospital and non-hospital psychiatric care appear to be complements, not substitutes. Third, the extent of public financing is correlated with the extent of mental health care services. The countries that retained mental hospitals and expanded community care also devote more government funds to this policy area, compared to those that reduced the hospital supply and lack non-hospital replacements. In this way, the mental health care market is different from the general (somatic) health care market, insofar as the extent of government spending does not determine the supply of mental health care. Public financing matters much more for mental health care provision than it does for general health care provision. I explain the theoretical reasons why this may be the case, and with what implications for scholarship on deinstitutionalization more broadly.