A study to determine if state governments use 'rainy day' funds to prevent tax increases during fiscal downturns

State governments establish "rainy day" funds, primarily to smooth state finances during years of fiscal downturns. This dissertation tests the effect of these funds on state governments' ability to mitigate tax rate increases throughout the 1980s and 1990s. Previous studies tested for tax smoothing by state governments by primarily relying on the random walk proposition of the tax-smoothing hypothesis (TSH). However, this dissertation posits that the random walk proposition is not the sole criterion by which a conclusion can be drawn as to whether a state government engages in tax smoothing. In addition to the random walk proposition, three additional propositions derived from the TSH will be tested in this dissertation for each state government. These theoretical propositions are tested via unit root, granger-causality, vector autoregression and Wald tests. In essence, more statistical evidence will be sought in order to more confidently reject or fail to reject this study's principal null hypothesis of no tax-smoothing than has been used in the previous studies.;The results from this study indicate that the fiscal data from most states do not conform to the TSH. This study attributes the lack of strong explanatory power of the TSH for state finances to two factors. First, several assumptions in the theoretical model of the TSH are not directly applicable to conditions and circumstances facing state governments. And second, whereas the TSH was formulated to only account for savings to be used for smoothing purposes, state savings appear to have also been used for precautionary purposes that are unrelated to the business cycle. However, this state-by-state analysis reveals that the fiscal policies from ten state governments have been found to not be entirely independent of a desire to smooth their tax rates.