We explore reporting nonrecurring gains and losses (extraordinary and special items) among municipalities. We begin by documenting the nature, type, and frequency of reporting, finding that surprisingly few municipalities report items, including those with FEMA disasters and despite GASB standards. Results suggest municipalities with CPA-finance directors are significantly more likely to report items, while less likely reported among those led by unelected bureaucrats (council-manager political governance). Moreover, nonrecurring items are systematically associated with surpluses and deficits in a manner suggesting strategic reporting. Specifically, our evidence suggests officials report income-increasing nonrecurring items to reduce or avoid reporting deficits, and income-decreasing items to reduce surpluses. Strategic reporting is generally magnified when state laws allow direct voter initiatives and reduced when state laws mandate GAAP accounting or external audits. Overall, we conclude that accounting expertise and political governance structures are significant determinants of reporting nonrecurring items, and that some officials strategically report them.

We explore whether municipalities with public sector unions exploit aspects of governmental (or “fund”) accounting to obscure the availability of discretionary resources in fund balance accounts, relative to municipalities without public sector unions. We first investigate whether governments with unions report higher proportions of discretionary resources outside of the general fund, a primary measure of financial health, and instead within less prominent fund types. Second, we explore whether governments with unions report lower ratios within accessible general fund balance account categories – that is, report lower proportions of unreserved fund balance. Primary findings are consistent with both hypotheses. Although somewhat mixed, cross‐sectional analyses reveal that effects are magnified when unions have more bargaining power, as proxied by the ability to strike or the absence of state right‐to‐work laws. Further analysis corroborates cross‐sectional findings by examining difference‐in‐differences specifications surrounding the quasi‐exogenous shock of Wisconsin's 2011 weakening of state public sector union laws and Ohio's time‐varying union contract negotiations. Overall, the evidence suggests that governments with unions shelter resources to avoid the appearance of large discretionary amounts available.