Industrial sociology and studies in the social-psychology of organizations have long been concerned with the effects of work setting on mental health. Despite the familiarity of the expression “My job is driving me to drink,” very little research has focused on the effects of work-setting induced stress on increased alcohol consumption. In this paper we test two hypotheses linking the perception of various work-setting problems to (1) the frequency of drinking, and (2) reasons proffered for drinking. Data on a national sample of workers are used to test these hypotheses, controlling for job status, education, age and sex. To test the first hypothesis a five-way analysis of variance was performed for each of eight possible work-setting problems; for the second hypothesis a series of log-linear models were estimated to determine whether attitudes towards drinking differ by the presence or absence of work-setting problems. Results on the first hypothesis were mixed, but the log-linear analysis indicated that a worker is much more likely to state a particular reason for drinking is important if he/she experiences any one of eight different work-setting problems.