The study examines longitudinal suicide data for the United States between 1910 and 1933. Marshall found that national wars decrease suicide not because they produce greater political integration in society, but because they decrease the national unemployment rate, thereby producing greater economic integration. This work argues that national wars also decrease suicide because they diminish alcohol availability in society, which reduces suicide triggered by alcohol consumption. In order to test the hypothesis the paper examines aggregate data for the United States in the World War I and prohibition era. Since the death registration area is incomplete in the previous time period, a control variable is introduced into the model to minimize measurement error. A pooled cross-sectional analysis of the data for 20 individual states indicates no lag effect for the alcohol consumption variable. A longitudinal analysis of the total national data between 1910 and 1933 shows that the levels of alcohol consumption and changes in the national business cycle are significantly related to suicide, while participation in World War I did not significantly influence national suicide.