Among top-level defense managers, there is a general belief that a better information system should be able to resolve the endemic problem of lapsed funding, that is, funding that goes unspent at the end of a fiscal year. We argue against this point of view by modeling the decision process for activity expenditures as a two-stage process where some of the cost uncertainty is resolved before the end of the year. Our main result is that an improved information system cannot eliminate lapsed funding, and may in fact have very little impact at all.