For several decades the dominance of a rather simplistic, reductionist and pessimistic 'medical model' has, especially in relation to 'schizophrenia', relegated poverty and its attendant disadvantages (child neglect and abuse, overcrowding, dysfunctional families, etc.) to the role of mere triggers of a supposed, but unproven, genetic predisposition. For seventy years, however, research has repeatedly demonstrated not only that poverty is a powerful predictor of who develops psychosis, and who is diagnosed 'schizophrenic' (with or without a family history of psychosis), but that poverty is more strongly related to 'schizophrenia' than to other mental health problems. This paper argues that an evidence-based resolution to the longstanding debate between 'social causation' and 'social drift' explanations is that the former perspective explains how poverty is a major cause of psychosis and the latter explains how poverty is involved in its maintenance. Poverty is also a predictor of diagnosis and treatment selection, sometimes regardless of actual symptomatology. Evidence is also presented demonstrating that relative poverty may be an even stronger predictor of mental health problems, including 'schizophrenia', than poverty per se. Psychologists are encouraged to pay more attention to the psycho-social causes of their clients' difficulties, to the role of the pharmaceutical industry in perpetuating a narrow 'medical model' and, most importantly in the long run, to the need for primary prevention programmes.