Relying on existing research to predict the effects of such afundamental reform on the currently uninsured is potentially prob-lematic, because existing work generally focuses on the decisionto enroll in employer-sponsored health insurance. The currentlyuninsured are rarely offered the opportunity to purchase insurancethrough an employer (Kaiser Family Foundation, 2004), calling intoquestion the utility of existing estimates for understanding insur-ance demand of this population. Not only are uninsured individualssubstantially poorer than the average worker offered employerinsurance, but the decision to, say, purchase subsidized insurancefrom a state exchange might fundamentally differ from the deci-sion to enroll in an employer-sponsored health plan, which takesplace in the context of co-workers, an employer and potentiallybenefits counselors.