Several possible linkages exist between business conditions and crime. High unemployment rates may lower the opportunity cost of time spent in criminal activity. On the other hand, recession could discourage crime by reducing the quality of criminal opportunities or the use of criminogenic commodities such as alcohol or guns. Reduced government budgets in a recession may diminish the system's capacity to contain crime. This list demonstrates that there are no clear predictions about whether crime is likely to be procyclical, countercyclical, or both. A nonparametric analysis of the nine complete business cycles as defined by the National Bureau of Economic Research between 1933 and 1981 provides strong evidence that robbery and burglary are countercyclical, and auto theft is procyclical. The procedure, however, does not produce an estimate of the magnitude of the effects and lacks statistical power; therefore, regression analysis of burglary, robbery, auto theft, and homicide was conducted, using the unemployment rates and employment-population ratio as indicators of economic conditions. The results support the conclusions of the nonparametric analysis. Tables and 24 footnotes are supplied.