The effect of Sunday sales bans and excise taxes on drinking and cross-border shopping for alcoholic beverages.

INTRODUCTION

Excessive consumption of alcohol has been associated with adverse health effects, including cirrhosis of the liver, pancreatitis, gastritis, and various cancers (Friedman and Klatsky, 1993). Excessive drinking has also been charged with generating a myriad of negative outcomes stemming from behavioral change including highway fatalities (Ruhm, 1996), suicide (Carpenter, 2004), youth risky sexual behavior (Markowitz, Kaestner, and Grossman, 2005), transmission of STDs (Chesson, Harrison, and Kassler, 2000), child abuse (Markowitz and Grossman, 2000), nuisance crime (Carpenter, 2005), and workplace accidents (Ohsfeldt and Morrisey, 1997). Economic theory suggests that the market failure due to the presence of negative externalities may be corrected by the application of an excise tax, but Saffer and Chaloupka (1994) and Kenkel (1996) argue that excise taxes on alcoholic beverages are too low to correct for these externalities. In addition to concern for health and safety, many oppose consumption of alcoholic beverages because of moral or religious beliefs.

Against this backdrop, state governments have a variety of policies in place that intervene in the market for alcoholic beverages. Prominent among these have been laws that influence the prices of alcoholic beverages. At the end of 2002, for example, all states levied excise taxes on beer and in 32 "license" states the primary intervention in the spirits market was also an excise tax. In the remaining "control" states, spirits were sold only through state stores where prices were set by legislated markups on wholesale prices. In recent decades, no state has converted from "license" to "control" and conversions in the other direction have been rare. (1) In addition to being relatively low, many excise taxes have fallen when measured in real terms. From 1990 to 2003, for example, the average state beer tax fell from 17.1 to 13.6 cents per gallon measured in 1982 cents, and the average state spirits tax fell from $2.53 to $2.06. While the federal beer tax increased from 22.2 to 42.6 cents in 1991, inflation eroded it to 31.5 cents by 2003, and the real federal spirits tax decreased from $9.56 to $7.34 over the same time span. In fact, even when measured in 2003 prices, the combined excise tax on beer is below one cent per ounce. (2) Because of their relatively low levels, excise taxes on alcohol raise only modest revenues for state governments. In 2002, excise taxes on beer, wine, and spirits provided states with approximately 3.2 billion dollars in revenues, sales taxes on alcoholic beverages provided approximately six billion dollars of additional revenue, and control states collected nearly 800 million dollars in revenue net of expenses through operation of their state-owned spirits stores (U.S. Census Bureau, 2006). Summed together, these policies were responsible for approximately one percent of state government revenues, a contribution slightly lower than the roughly 12 billion dollars states earned from excise taxes on tobacco products for the same year (Orzechowski and Walker, 2003). Despite their relatively low levels, alcohol taxes are likely targets for increases because they are one of the few taxes palatable to the general public. A survey released by the American Medical Association Office of Alcohol and Other Drug Abuse reveals that while 65 percent of respondents favored increasing alcohol taxes, 75 percent or more were against raising either the sales or income taxes or reducing spending on social services, Medicaid, or education (American Medical Association, 2004).

In addition to tax and price policies, federal, state, and local governments have a long history of enacting laws that ban the sale of alcoholic beverages. Most famously, the eighteenth amendment to the U.S. constitution banned the sale of alcoholic beverages from 1920 until 1933 when the twenty-first amendment repealed it and explicitly gave states the right to restrict the sale of alcohol. Although Mississippi repealed the last statewide ban on the sale of alcohol in 1966, numerous local governments still ban the sale of alcohol, and many states have in place bans on the Sunday sale of alcohol. Sunday sales bans were originally an outgrowth of Blue Laws enacted before the American Revolution to regulate behavior on the Christian Sabbath. They were revived toward the end of the nineteenth century as part of the temperance movement and again after the end of Prohibition when states regained control of laws governing the sale of alcohol (Christian Science Monitor, 2003). At the end of 2002, 11 states had bans on Sunday sales of beer and 27 states plus the District of Columbia had such bans in place for spirits.

Sunday sales bans, and to a lesser extent excise taxes, have been the focus of much recent legislative and judicial activity. The Distilled Spirits Council of the United States (DISCUS) records 28 excise tax changes for beer between 1990 and 2004 and over 30 changes in state excise taxes and markups for spirits; in 2005 at least twelve states considered bills to increase their taxes on alcoholic beverages (Center for Science in the Public Interest, 2005). Table 1 summarizes recent changes in Sunday sales bans for beer and spirits. New Mexico and Oregon repealed their bans on Sunday sales in 1995 and 2002, respectively; Delaware, Kansas, Massachusetts, New York, and Pennsylvania repealed or relaxed restrictions on Sunday sales of alcohol in 2003, five other states adopted similar measures in 2004 and Washington followed suit in 2005. No states have imposed new Sunday sales bans in recent decades. Because these interventions raise the full price of alcohol, some consumers may choose to purchase substitute products or to circumvent them by traveling to jurisdictions where laws or enforcement are more lenient. Higher prices, for example, may cause some consumers to travel to other states where prices are lower and some consumers may shop in other states when faced with a Sunday sales ban at home. Such activity is of concern because it undermines the ability of states to curb drinking, influences the amount of revenue states collect through taxation, and wastes resources by imposing unnecessary time and transportation costs on consumers.

The literature contains several papers that consider commodity tax competition and cross-border shopping from a theoretical perspective. Mintz and Tulkens (1986) examine commodity taxation in a general equilibrium model of two countries trading two goods. Although their model is quite general, this generality comes with a price; the paper reaches few general conclusions regarding the welfare effects of tax competition. More to the point, it does not consider the effect of country size on strategic interaction, an element likely to be important when explaining cross-border traffic between U.S. states. Kanbur and Keen (1993) develop a partial equilibrium model of two countries with one taxed good where population sizes are allowed to differ. Two predictions from their model are of particular relevance to this study. First, they find that the smaller country always undercuts the tax of the larger country. Second, they find that per-capita revenue is larger in the smaller country even though the smaller country has a lower tax rate, implying that it must be exporting sales to the larger country. The prediction for the U.S. is, thus, that smaller states will undercut the taxes of their larger neighbors and earn revenues from cross-border shoppers.

Empirical studies in this literature fall into two categories: those showing the interdependence of policies across jurisdictions and those showing that policy differences can result in cross-border shopping or migration. Conway and Rork (2004) find evidence that states' inheritance, gift, and estate taxes are influenced by the reliance on those taxes of competitor states. Rork (2003) shows that tobacco, gasoline, personal income, sales, and corporate income taxes are affected by the rates of those taxes in neighboring states. Although the implication in these papers is that competition is driven by cross-border shopping or migration, the papers do not directly test whether migration or sales in a state are influenced by the policies of neighboring states. Cross-border shopping has, however, been well documented in other studies. Using data from Tennessee, Luna (2004) shows the interdependence of county sales tax rates and that county tax bases are influenced by the sales taxes of neighboring counties. (3) Numerous authors in the tobacco literature have shown that interstate differences in cigarette taxation have led to cross-border shopping or long-distance cigarette smuggling (Baltagi and Levin, 1986; Coats, 1995; Saba, Beard, Ekelund, and Ressler, 1995; Thursby and Thursby, 2000; Yurekli and Zhang, 2000; and Stehr, 2005).

If tax or policy differences lead consumers to cross state lines to purchase alcoholic beverages, then this behavior should be reflected by unexpected variation in sales across states. Table 2 presents average state per--capita annual sales of beer and spirits for the years 1990-2004 from DISCUS. Sales of spirits in New Hampshire, Washington DC, Delaware, and Nevada are very high, while sales in Utah and Pennsylvania are very low, relative to other states. Dispersion in per-capita beer sales is not as great but is also evident. While the data are consistent with cross-border traffic, they are also consistent with other hypotheses. Low sales in Utah, for example, are likely due to the strong religious influences in that state, and high sales in Nevada are likely due to the strong tourism and gambling industries.