Alcohol-Related Problems and Retail Density

In addition to Michigan’s monopoly over the wholesaling of spirituous liquor, the state regulates the number of retail liquor stores according to population density. By controlling the number of liquor stores, the state hopes to control the availability of alcohol and thereby reduce the rate of problematic alcohol consumption. As a MAPPHS spokesman told a Detroit News reporter, “We know that the more alcohol outlets there are, the more alcohol-related problems and harm there is [sic].”[24]

There is research, however, that casts doubt on this view. Tenaya Marie Sunbury, a University of Michigan Ph.D. candidate, studied alcohol-related problems in Michigan in her 2010 dissertation, titled “Urban-Rural Influences on Driving Behaviors and Driving Outcomes Among Michigan Young Adults: An Investigation of Roadway Characteristics, Alcohol Establishments, and Social Influences.”[25] Her research is unique in investigating the issue of retail density in Michigan at a “micro-level” — that is, it included individual characteristics of the Michigan residents who were part of the study. These characteristics included age, sex, marital status, education, personal income, vehicle type and miles driven.[26] Sunbury also incorporates “psychosocial” characteristics measured by the individuals’ responses to survey questions about, for instance, whether they “like to live dangerously,” think it is wrong “to shoplift something of value from a store” or “hit back” if “people push me around.” These characteristics, which described individuals’ tolerance of deviance, risk-taking and verbal and physical hostility, were selected as relevant to the respondents’ “driving behaviors.”[27]

Among other findings, she concluded, “For both men and women, higher density of alcohol establishments was related to lower alcohol consumption (quantity/frequency), binge drinking, and drink/driving [sic].”[28] Sunbury further found, “[P]eople who reside near fewer alcohol establishments (e.g. rural areas) are at greater risk of alcohol misuse and alcohol-related crashes.”[29] Sunbury hypothesizes this may be due to greater driving distances, which increase the risk of a crash.[*]

Another intriguing paper, titled “Alcohol-related crashes and alcohol availability in grass-roots communities” was published in 2003 by economist Patrick McCarthy.”[30] His research involved data from an eight-year period during the 1980s and examined the alcohol outlet density in 111 California nonmetropolitan cities and the density of alcohol-related establishments and alcohol-related auto accidents.[31]

McCarthy considered two main types of outlets: “off-site” and “on-site.” An off-site establishment sells alcohol for consumption off the premises. An on-site establishment permits alcohol consumption on the business’s property. He also looked at two types of licenses: general alcohol licenses, which permit an establishment to sell all types of alcohol (including hard liquor), and beer-and-wine licenses, which do not permit liquor sales.[32]

McCarthy found that an increase in the density of off-site general spirits establishments was associated with decreases in both fatal and nonfatal alcohol-related auto accidents.[33] He found that the density of off-site outlets with beer-and-wine licenses had no effect on fatal crashes, but was again associated with decreases in nonfatal crashes and total crashes.[34]

McCarthy also found that an increase in the density of on-site beer-and-wine establishments had no effect on alcohol-related crashes.35 He did, however, observe that the density of on-site sellers with general alcohol licenses was associated with higher nonfatal car crashes, though it was not associated with higher fatal crashes.

On the whole, McCarthy’s study suggests that increasing the density of all four types of establishments — onsite and offsite with either general alcohol or beer-and-wine licenses — had either no effect or a beneficial effect on alcohol-related crashes, with the exception of on-site general alcohol businesses, which had a negative effect on nonfatal alcohol-related crashes. On balance, McCarthy notes, the positive effects outweighed the negative result, and, “[A] uniform increase in general alcohol licenses (on- and off-site) will be safety-enhancing.”36

It is worth noting that McCarthy’s study did not review the economic costs of this retail density regulation on consumers and the retail economy. Both of these costs could themselves reduce the population’s well-being.

These are only two studies.[†] Nevertheless, each involved extensive data — in one case, from Michigan itself — and both suggest that retail density regulation does not generally have positive effects on driving safety and may even have negative effects.

[*] Sunbury, “Urban-Rural Influences on Driving Behaviors and Driving Outcomes Among Michigan Young Adults: An Investigation of Roadway Characteristics, Alcohol Establishments, and Social Influences,” (University of Michigan, 2010), 79, (accessed March 29, 2012). Sunbury also speculates that the finding may be a function of a possible “drinking culture” that may make drinking and driving more accepted in rural areas. Ibid.

[†] These are not the only studies to find that regulation of retail density does not produce the intended public health effects. Research by Gabriel Picone et al. used data from 1985 to 2001 for four major U.S. cities to investigate the relationship between alcohol consumption and the proximity of bars to people’s residences. The authors found, “When person-specific fixed effects are included, the relationship between alcohol consumption and the number of bars within a 0.5 km radius of the person’s place of residence disappears. … We conclude that bar density in the area surrounding the individuals’ homes has at most a very small positive effect on alcohol consumption.” Gabriel Picone et al., “The Effects of Residential Proximity to Bars on Alcohol Consumption,” International Journal of Health Care Finance and Economics Vol. 10, no. 4 (2010): 347.