At least that is the case for the healthcare obligations of most states and local areas, though Robert Pozen points out that this is likely to change soon (available here). This is something I discussed when Detroit went through bankruptcy proceedings. Pension and healthcare obligations of the city made a contribution to the bankruptcy, but only because the broader electorate allow it. If we get accurate accounting and compel elected officials to use realistic discount rates there would be a better sense of the amount of these obligations. The proposed accounting changes are a step in the right direction, requiring more transparency and better assumptions.
For anybody paying even a modicum of attention pension math has been a source of significant debate in the last decade. There are many aspects of pension math worthy of discussion and this recent New York Times article points out many of these issues in light of the Detroit bankruptcy filing and the public sector employee pensions. To put it bluntly, most pensions assume their result.