SummaryIt has been established that relatively high income families receive a disproportionately large share of UI benefits. Little information on the distribution of costs by income class is available, but it is quite possible that the upper income groups may also bear a disproportionately large share of the costs. More empirical research is needed before the redistributive effects of UI can be known with any degree of confidence.There is little argument about the desirabilityon income mainlenance groundsof extending Unemployment insurance benefits in a recessionary period, all other things being equal. U1 is quite efficient as an income maintenance tool for covered workers. With respect to theunemploymentproblem, though, it is feared by some that the program goes too far in maintaining the incomes of covered workers, and as a result any extension of UI benefits will aggravate the unemployment problem. The evidence presented here suggests that fears of amajorincrease in unemployment as a result of such changes are largely unwarranted. However, extended UI benefits would appear to have some small, nontrivial adverse impact on unemployment. Finally, it seems that UI'sincome distributionalimpact has been portrayed unjustly as pro‐rich. It may be that income is transferred through the UI system from the relatively well‐to‐do to the relatively poor.There is no scientific way of bringing these considerations to bear in deciding whether or not to restructure the unemployment insurance system. This question can be answered only by invoking fundamental value judgments to decide how sociely would be better off, and I would not presume here to offer my own. This paper will have served its purpose if it has provided the basis for a more informed dec