This paper examines the impact of diabetes on the employment of Canadian males and females ages 15–64. Using data from the National Population Health Survey (1998), it utilizes a recursive bivariate probit approach to take into account the potential endogeneity of diabetes in employment outcomes. The results suggest that treating diabetes as exogenous yields an over-estimation of its impact on male employment. The study finds that diabetes has a significant negative impact on female employment probability, but has no significant impact on that of non-white Canadians. An implication thus is that policy-makers should take endogeneity into account in estimating labor market costs of chronic diseases such as diabetes. Copyright © 2008 John Wiley & Sons, Ltd.