The study was prompted by the observation that while many countries' official crime statistics reveal that unemployed people have high crime rates and that communities with a lot of unemployment experience a lot of crime, this cross-sectional relationship often does not occur in time-series studies of unemployment and crime. In Australia, no individual-level or cross-sectional studies of unemployment and adult crime have failed to produce a positive relationship, whereas no time-series studies have supported a positive relationship. The present research involved a time series of homicide from 1921 to 1987 in Australia. Results revealed no significant unemployment effect. A theoretical resolution of this apparent was proposed in terms of the effect of female employment on crime in a patriarchal society. Crime was posited as a function of both total unemployment and female employment. When female employment was added to the model, it had a strong positive effect on homicide; unemployment also had a strong positive effect. Thus, the effect of either male or female unemployment rates became larger and more significant after adding the female employment rate. Therefore, the female employment rate changed across time in a way that masked the positive effect of female unemployment and male unemployment rates on crime. Figures; tables; footnotes; appended data sources and definitions; and 95 references (Author abstract modified)