Lane, and an editor and the referees of this Journal for their comments and suggestions. We are responsible for all remaining errors.

This paper examines why local governments rely heavily on the property tax, even when they have access to another revenue source, using data from Ohio’s recent experience of permitting local school districts to use both property taxes and residence-based income taxes. Nechyba’s (1997) theory that local governments’ reliance on the property tax instead of the income tax is due to fiscal competition for relatively high-income residents is tested using data from 610 Ohio school districts. The Ohio residence-based school district income tax is used by only 119 school districts, at low tax rates, to supplement the traditional property tax. The use of a local income tax declines sharply as fiscal competition increases, as measured by the number of nearby school districts. School districts with greater opportunities to export the burden of the property tax to non-residential property owners are less likely to adopt a local income tax.

We study the implicit tax incidence of raising state revenue through a monopoly state-run lottery using a new dataset on individual Minnesota lottery game sales by zip code. We use the bootstrap to compute SEs and construct confidence intervals for Suits Indices of seven lottery products. We conclude that the implicit tax on each product is regressive, and find statistically significant differences in regressivity between some products. Minnesota's newly introduced G3 instant scratch product, printed at time and place of purchase, is also the most regressive lottery game.

This paper examines the tax base elasticity of the regulated casino industry in Illinois to help estimate state-level revenue impacts of casino tax rate changes. Illinois' shift to a graduated rate schedule increased the highest marginal tax rate on casino adjusted gross receipts (AGR) from 20 percent to 70 percent before reverting to a 50 percent rate. We construct a state-level casino tax rate variable, which is a statewide average for each month of the marginal casino tax rate facing each casino. We find that a 1 percent increase in this state-level casino tax rate decreases overall Illinois casino AGR by around 1.1 percent.

We analyze the increase in sales of daily numbers lottery games -Pick 3 and Pick 4 games -after Ohio introduced midday drawings in August 1999. Using a 36 month panel data set of Ohio lottery sales by zip code we find that midday drawings increased Pick 3 sales by 12.1% per adult and increased Pick 4 sales by 16.6% per adult. The increase in both Pick 3 and Pick 4 sales after midday drawings began was greater in zip codes with a greater percentage of households receiving public assistance and zip codes with a higher percentage of black residents. The Pick 3 Red Ball promotions during this period, which increased the payouts to winning Pick 3 tickets, were successful in raising Ohio Lottery sales and profits. Pick 3 sales per adult would have declined 15.4% and Pick 4 sales per adult would have dropped 2.4% between 1998 and 2000 without any marketing innovations by the Ohio Lottery. This trend of falling lottery sales over time was more severe in areas near casinos, many of which opened in the mid 1990s.