ArticlePDF Available

working paper BiNgE ThiNkiNg: A Look AT ThE SociAL impAcT of STATE Liquor coNTroLS

By Antony Davies and John Pulito
No. 10-70
November 2010
The ideas presented in this research are the authors’ and do not represent official positions
of the Mercatus Center at George Mason University.
Binge Thinking: A Look at the Social Impact of State Liquor Controls
1) How does the government regulate alcohol?
Studies on the efficacy of alcohol controls have focused on four broad categories of
market intervention:
Demand Regulation: Laws aimed at reducing demand target those who purchase and
consume alcoholic beverages. These laws place the restrictive burden on the alcohol
buyer and include age and visibly intoxicated person (VIP) restrictions.
Supply Regulation: Laws aimed at reducing the supply of alcohol target those who
sell alcoholic beverages. These laws include government ownership of retail or
wholesale outlets, restrictions on outlet density, caps on the number of retail licenses,
and limits on hours of operation, all of which place the restrictive burden on the
alcohol seller.
Price Regulation: These laws principally take the form of taxes and, in some cases,
price controls.
Social Regulation: These laws target behavior that is deemed socially undesirable or a
public health threatfor example, regulation of public consumption, and laws related
to driving under the influence (DUI) or driving while intoxicated (DWI). The
restrictive burden is placed on the alcohol consumer (who may or may not be the
alcohol buyer).
2) What is a “Control” State?
―Control‖ is somewhat of a misnomer as all states have authority to exercise control over
alcohol sales. The National Alcohol Beverage Control Association (NABCA) defines a
control state as one in which the state has a monopoly on the wholesale and/or retail sale
of one or more classifications of alcohol (e.g. beer, wine, and/or spirits).1,2 A non-control,
or ―license, state is one in which the state does not have a monopoly, but limits the
distribution and sale of alcohol by licensing vendors.
1 This definition is inferred from the details provided by the NABCA. Specifically, the NABCA classifies a state as
a control state if a controlled distribution system substitutes the state for the private marketplace in the retail and/or
wholesale of alcohol (NABCA, 2010a).
2 This classification ignores variations in state control. For instance, while Pennsylvania has a state monopoly on the
retail and wholesale sale of liquor and wine, Virginia has a state monopoly on the retail and wholesale sale of liquor
only. Regardless of these variations, Pennsylvania and Virginia are control states.
3) How many Control States remain in the United States?
NABCA classifies nineteen states as control states. It important to note, however, that the
state of Maryland itself is not a control state though several counties within Maryland
operate as local control jurisdictions. For the purposes of this paper, Maryland will
neither be depicted nor discussed as a control state. Therefore only eighteen states will be
considered control states.
Figure 1 Control States according to the NABCA in 2010 (NABCA, 2010b). As illustrated below, Maryland is not a
control state. However, Montgomery, Somerset, Wicomico, and Worcester counties operate as local control jurisdictions
in accordance with state law (Gansler et al., 2009).
4) Do all states regulate alcohol in the same way?
All states regulate alcohol in some form, though the exact form of this control varies from
state to state. For instance, of the eighteen control states that have monopolies over the
wholesale sale of liquor, twelve of those states also control liquor sales at the retail level.
Figure two provides more detail on the current variations.
Figure 2 - Monopolization of alcohol sales by state, alcohol type, and market in 2010 (NABCA, 2010b). As illustrated
below, Maryland is not a control state. However, Montgomery, Somerset, Wicomico and Worcester Counties operate as
local control jurisdictions in accordance with state law (Gansler et al., 2009).
5) How does a control state privatize its state alcohol retail stores?
The following three examples demonstrate three methods in which a state can privatize
its alcohol retail stores.
Pennsylvania: Pennsylvania is considering privatizing its monopoly on the retail and
wholesale sale of liquor and wine by auctioning off 750 retail and 100 wholesale licenses
(Pennsylvania House Bill 2350, 2010).
Virginia: Virginia is currently considering privatizing its monopoly on the retail and
wholesale sale of liquor by auctioning off 1,000 off-premises distilled licenses
(McDonnell, 2010).
Washington: On November 2, 2010, two ballot initiatives were voted onInitiatives
1100 and 1105. Although neither ballot received enough votes to become law, each
attempted to privatize Washington’s monopoly on the retail and wholesale sale of liquor.
Initiative 1100 called for all state liquor stores to be closed and replaced by a new
licensing system (Washington Initiative Measure No. 1100, 2010). Initiative 1105 took
this one step further by also requiring changes in the current liquor sales tax laws. For
instance, under this initiative, the liquor control board would have been required to
present a report to the Legislature by January 1, 2011, recommending a rate of taxation
(paid by liquor distributors) that would generate the same amount of revenue as is
generated under the current system plus an additional $100 million dollars (Washington
Initiative Measure No. 1105, 2010).
6) Would a control state suffer financially from privatization?
Pennsylvania: Pennsylvania is considering privatizing its monopoly on the retail and
wholesale sale of liquor and wine. A bill introduced by Representative Mike Turzai
would allow Pennsylvania to auction 750 retail and 100 wholesale licenses (Pennsylvania
House Bill 2350, 2010) which could raise approximately $1.7 billion (Segal and
Underwood, 2007). While this would only represent a one-time cash flow, Nathan
Benefield of the Commonwealth Foundation estimates that Pennsylvania would continue
to generate an estimated $500 million annually from alcohol sales taxes (Benefield,
2010). In the last fiscal year, Pennsylvania incurred the largest budget deficit in the
state’s history—the bulk of which was closed by federal money. The money potentially
raised by privatizing the state liquor stores would have closed this budget deficit.
Virginia: Governor Bob McDonnell estimates that Virginia could raise approximately
$400500 million as a result of privatization. With expected changes to the tax, markup,
and fee structures, the governor also expects that the state would continue to receive an
additional $324.2 million annually in taxes and fees (McDonnell 2010).
Washington: According to a state government performance review, state auditor Brian
Sonntag examined six possible options for privatizing Washington’s monopoly on the
retail and wholesale sale of liquor. From the six possible outcomes, Brian concluded that
Washington could increase revenues by as much as $277 million over five years
(Sonntag, 2010).
7) What happens to employees currently working in state-run liquor stores?
Initially, privatization could result in the loss of jobs by employees working in the state-
run stores. If state-run stores are simply shut down, some employees will experience
temporary unemployment as they shift from working for state-run stores to working for
private stores or other jobs. If state-run stores are sold, workers may not experience
unemployment at all but simply find themselves working under new management.
Noteworthy is that many of the jobs will not disappear, but will merely shift from the
public to private sector.
8) Assuming privatization leads to an increase in retail liquor stores, will this lead to a
corresponding increase in alcohol consumption?
An increase in retail liquor stores does not necessarily imply an increase in alcohol
consumption. Studies that point to a possible relationship between retail store density and
alcohol consumption fail to address causality.3 For example, one would expect to observe
a higher rate of alcohol consumption and a higher density of alcohol retailers near college
campuses. It would be incorrect to conclude that the high density of retailers causes the
greater rate of alcohol consumption. In fact, it is more likely that the causality is
reversedi.e. the density of retailers is caused by the concentrated population of alcohol
While there is a natural tendency to blame markets for people’s behaviors, markets are
merely the aggregation of people’s behaviors. In other words, markets do not cause
behavior; behavior causes markets. Liquor stores are more likely to choose to locate
where there are customers as opposed to customers choosing to live where liquor stores
are located. Finally, it is possible that, if the relationship between density and
consumption is causal, that the causality is bi-directional. It may be that increased density
causes increased consumption and increased consumption contributes to increased
density. For example, one might imagine college students choosing where to live based
on proximity to bars, whereas bar owners choose where to locate based on the proximity
to on- and off-campus student housing.
9) How will privatization affect alcohol consumption?
Although numerous studies have been conducted in this area, there is no clear evidence
that privatization of alcohol markets leads to either an increase or decrease in alcohol
consumption.4 Studies that show a positive relationship between privatization and alcohol
consumption are countered with studies, using the same data, which show no
While over-consumption of alcohol clearly has deleterious consequences, alcohol
consumption, per se, is not bad. In addition to being an activity that many people find
enjoyable, research has shown that moderate alcohol consumption reduces the risk of
heart disease, stroke, gallstones, and diabetes (Mayo Clinic Staff, 2010).
10) How will privatization affect underage drinking, underage binge drinking, and DUI
Although numerous studies have been conducted in this area, there is no clear evidence
that privatization of alcohol markets leads to either an increase or a decrease in underage
drinking, underage binge drinking, or DUI fatalities. Studies showing a positive
3 Refer to Gruenewald et al. (1996), Douglas et al. (1997), and Presley et al. (2002).
4 Refer to MacDonald (1986), Holder and Wagenaar (1990), Wagenaar and Holder (1991), Mulford, Ledolter, and
Fitzgerald (1992), Wagenaar and Holder (1995), Weitzman et al. (2003), Trolldal (2005a, 2005b), Miller et al.
(2006), Pulito and Davies (2009), and Stockwell et al. (2009).
relationship (e.g., Stockwell et al., 2009, Weitzman et al., 2003, Wagenaar and Holder,
1995) are counterbalanced by others showing an absent or ambiguous relationship (Pulito
and Davies, 2009, Trolldal, 2005a, Trolldal, 2005b).
Some studies that show relationships may suffer from unaddressed statistical anomalies
that bias the results in favor of finding relationships where none exist. For example,
Wagenaar and Holder (1995) include wine cooler sales in their study despite the fact that
wine coolers were excluded from the change in alcohol laws and during the period of
time studied, there was a coincidental surge in the popularity of wine coolers. The
analysis done by Stockwell et al.’s (2009) suffers from unaddressed statistical
complications that render their results meaningless.
Studies that do show relationships also suffer from unaddressed causality, making the
results useless for guiding policy makers. Implicit in many studies is the assumption that
alcohol consumption causes, rather than is caused by (or related to), social ills. If in fact
the causality is reversed or absent, one would expect that reducing alcohol consumption
would have no effect on social ills.
11) What type of control minimizes underage drinking, underage binge drinking, and
DUI fatalities?
Pulito and Davies (2009) test for differences in the incidence of underage drinking,
underage binge drinking, per-capita alcohol consumption, and DUI fatalities among
privatized versus non-privatized states. Unlike previous research, this study looks at 48
states over a period of 16 years and classifies the states according to the degree of
privatization. They categorize the possible degrees of privatization versus control as:
full control (sales of beer, wine, and liquor are monopolized at the retail and
wholesale levels).
moderate control (sales of beer, wine, and liquor are monopolized at the wholesale
level, and sales of only one type of alcohol are monopolized at the retail level).
light control (sales of beer, wine, and liquor are monopolized at the wholesale level,
and sales of all three types are privatized at the retail level).
license (sales of beer, wine, and liquor are privatized at the wholesale and retail
levels; the state only monopolizes alcohol sales via the licensing of private firms).
They compare the degree of state ownership to per-capita alcohol consumption, the
incidence of underage drinking, the incidence of underage binge drinking, and alcohol-
related traffic fatalities and find that alcohol consumption is significantly greater in
license states versus light-control states.
States with full, moderate, and light control are statistically identical. They find no
significant difference in the incidence of underage drinking or the incidence of underage
binge drinking among the four control classifications, and no difference in the rate of
DUI arrests among the four control classifications. The number of alcohol-related traffic
fatalities per DUI arrest is the same for full control and license states, and significantly
lower for moderate and light control states.
Table 1 Pulito and Davies (2009) results (changes are as compared with no privatization)
Benefield, N. (2010). ―Failing to understand liquor store revenues.‖ Commonwealth Foundation.
Douglas, K.A., Collins J.L., Warren, C., Kann, L., Gold, R., Clayton, S., Ross, J.G., and Kolbe,
L.J. (1997). ―Results from the 1995 national college health risk behavior survey.‖ Journal
of American College Health. 46(2): 55-66.
Gansler, D.F., Darsie, J.L., and McDonald, R.N. (2009). ―Liquor control boards—whether liquor
control boards are subject to state or county ethic laws.‖ Letter to the State Ethics
Gruenewald, P.J., Millar, A.B., and Roeper, P.J. (1996). ―Access to alcohol: Geography and
prevention for local communities.‖ Alcohol Health and Research World. 20(4): 244-251.
Holder, H.D. and Wagenaar, A.C. (1990). ―Effects of the elimination of a state monopoly on
distilled spirits’ retail sales: A time-series analysis of Iowa.‖ British Journal of Addition.
85(12): 1615-1625.
Pennsylvania House Bill 2350. (2010). ―Privatizing retail and wholesale wine and spirits sales in
Macdonald, S. (1986). ―The impact of increased availability of wine in grocery stores on
consumption: Four case histories.‖ British Journal of Addiction. 81(3): 381-387.
Mayo Clinic Staff. (2010). ―Nutrition and healthy eating.‖ Mayo Foundation for Medical
Education and Research.
McDonnell, B. (2010, September 09). McDonnell: End Virginia’s outdated government
monopoly.‖ Richmond Times-Dispatch.
Underage Binge
Total Sales
Partial privatization of retail stores
No change
No change
No change
Full privatization of retail stores
No change
No change
Full privatization of retail and
wholesale stores
No change
No change
Miller, T., Snowden, C., Brickmayer, J., and Hendrie, D. (2006). ―Retail alcohol monopolies,
underage drinking, and youth impaired driving deaths.‖ Accident Analysis and
Prevention. 38(6): 1162-1167.
Mulford, H.A., Ledolter, J., Fitzgerald, J.L. (1992). ―Alcohol availability and consumption: Iowa
sales data revisited.‖ Journal of Studies on Alcohol. 53(5): 487-494.
National Alcohol Beverage Control Association (NABCA). (2010a). ―Historical overview.‖
National Alcohol Beverage Control Association (NABCA). (2010b). The effects of
privatization of alcohol control systems.
Presley, C.A., Meilman, P.W., and Leichliter, J.S. (2002). College factors that influence
drinking. Journal of Studies on Alcohol. Supplement 14: 82-90.
Pulito, J. and Davies, A. (2009). Government-run alcohol stores: The social impact of
privatization. Commonwealth Foundation Policy Brief. 21(3): 1-16.
Segal, G. and Underwood, G. (2007). Divesting the Pennsylvania Liquor Control Board.
Reason Foundation.
Sonntag, B. (2010). ―State government performance review: Opportunities for Washington.‖
Stockwell, T., Zhao, J., Macdonald, S., Pakula, B., Gruenewald, P., and Holder, H. (2009).
―Changes in per capita alcohol sales during the partial privatization of British Columbia’s
retail alcohol monopoly 2003-2008: A multi-level local area analysis. Addiction.
104(11): 1827-1836.
Trolldal, B. (2005a). An investigation of the effect of privatization of retail sales of alcohol on
consumption and traffic accidents in Alberta, Canada. Addiction. 100(5): 662-671.
Trolldal, B. (2005b). ―The privatization of wine sales in Quebec in 1978 and 1983 to 1994.‖
Alcoholism: Clinical and Experimental Research. 29(3): 410-415.
Wagenaar, A.C. and Holder, H.D. (1991). ―A change from public to private sale of wine: Results
from natural experiments in Iowa and West Virginia.‖ Journal of Studies on Alcohol and
Drugs. 52(2): 162-173.
Wagenaar, A.C. and Holder, H.D. (1995). Changes in alcohol consumption resulting from the
elimination of retail wine monopolies: Results from five U.S. states. Journal of Studies
on Alcohol and Drugs. 56(5): 566-572.
Washington Initiative Measure No. 1100. (2010).
Washington Initiative Measure No. 1105. (2010).
Weitzman, E.R., Folkman, A., Folkman, K.L., and Wechsler, H. (2003). The relationship of
alcohol outlet density to heavy and frequent drinking and drinking-related problems
among college students at eight universities. Health and Place. 9(1): 1-6.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Recent changes in Iowa liquor control laws ended the 51-year-old state alcohol monopoly distribution system and turned the sales of bottled wine and spirits over to the private sector. The resulting increase in the availability of these beverages provided a unique opportunity to study the relationship between increased wine and spirits availability and changes in their consumption. Time series analyses of monthly sales (apparent consumption) trends showed that the increased availability had no lasting impact on consumption.
Full-text available
Several proposals to change the existing structure of state alcoholic beverage control authorities have been advanced over the past 10 years in the United States. Most call for eliminating or substantially reducing the state's role in retail sales of alcohol, particularly distilled spirits and wines. In recent years the states of Iowa and West Virginia eliminated state monopolies for retail sales of wine, and now allow such sales by privately-owned licensed establishments. Using time-series methods, we assessed the effects of these policy changes on the alcoholic beverage market in each state. Privatization was associated with statistically significant increases in wine sales in Iowa and West Virginia, after controlling for an initial stocking effect and broader nationwide trends in alcohol sales in the 1980s. In addition, there was a net increase in absolute alcohol consumed in both states across all beverages (beer, wine and distilled spirits) associated with privatization.
Full-text available
Since the end of prohibition in the USA, only one state, Iowa, has eliminated its state retail monopoly of distilled spirits sales for off-premise consumption. In March 1987 all state retail stores were closed, and private establishments were licensed to sell distilled spirits. Interrupted time-series analyses of apparent spirits consumption in Iowa, controlling for nationwide sales trends over the past two decades, identified a statistically significant 9.5% increase in spirits sales following the policy change. While there was a corresponding 13.7% decline in wine sales, and no change in beer sales, privatization of spirits retail sales yielded a net increase in total alcohol consumption in Iowa. No changes were found in spirits sales in states bordering Iowa.
Full-text available
The role of publicly owned and operated retail alcohol monopolies is currently under debate in many Western and former Eastern-bloc countries. We studied the effects of privatizing wine sales in five U.S. states. Data on monthly sales of alcoholic beverages were collected for each of the five states, for all states bordering each of the five states and for the U.S. as a whole for the period from 1968 through 1991. Beer and spirits sales data were collected for comparison with wine sales. A quasiexperimental interrupted time-series design was used, including comparison groups consisting of border states, all other U.S. states and beer and spirits sales within the focal state. Box-Jenkins time-series statistical modeling was used to control for intra- and cross-series dependencies and to estimate the net effect of privatization on wine sales. After controlling for both nationwide and state-specific trends, we found significant increases in wine sales after privatization of 42% in Alabama, 150% in Idaho, 137% in Maine, 75% in Montana and 15% in New Hampshire. The increases in liters of pure ethanol per year in the form of wine were 621,000 in Alabama, 432,000 in Idaho, 364,000 in Maine, 363,000 in Montana and 171,000 in New Hampshire. The structure of the retail alcohol distribution system has a significant effect on alcohol sales. We recommend that the social costs associated with increased alcohol use be carefully considered before such major policy changes are contemplated.
Full-text available
The purpose of this article is to examine the aspects of collegiate environments, rather than student characteristics, that influence drinking. Unfortunately, the existing literature is scant on this topic. A literature review of articles primarily published within the last 10 years, along with some earlier "landmark" studies of collegiate drinking in the United States, was conducted to determine institutional factors that influence the consumption of alcohol. In addition, a demonstration analysis of Core Alcohol and Drug Survey research findings was conducted to further elucidate the issues. Several factors have been shown to relate to drinking: (1) organizational property variables of campuses, including affiliations (historically black institutions, women's institutions), presence of a Greek system, athletics and 2- or 4-year designation; (2) physical and behavioral property variables of campuses, including type of residence, institution size, location and quantity of heavy episodic drinking; and (3) campus community property variables, including pricing and availability and outlet density. Studies, however, tend to look at individual variables one at a time rather than in combination (multivariate analyses). Some new analyses, using Core Alcohol and Drug Survey data sets, are presented as examples of promising approaches to future research. Given the complexities of campus environments, it continues to be a challenge to the field to firmly establish the most compelling institutional and environmental factors relating to high-risk collegiate drinking.
Alcohol availability, measured in terms of geographic density of alcohol-sales outlets, is linked to specific patterns of alcohol-related motor vehicle crashes in communities. To curb alcohol-related problems such as violence, traffic crashes, and drinking and driving, community advocates often focus on reducing alcohol availability through modifications of zoning laws and licensing requirements. In developing interventions and policy activities to reduce alcohol availability, community policymakers should assess the effects of outlet density using an adequate community "biogeography" of the interrelationships among drinkers, their drinking environments, the locations of alcohol outlets, and the locations of alcohol problems. Communities also should consider that the economic development of downtown areas, which is often accompanied by the rapid growth of alcohol outlets, is likely to affect the incidence and prevalence of alcohol-related problems in surrounding areas.
To investigate the independent effects on liquor sales of an increase in (a) the density of liquor outlets and (b) the proportion of liquor stores in private rather than government ownership in British Columbia between 2003/4 and 2007/8. The British Columbia Liquor Distribution Branch provided data on litres of ethanol sold through different types of outlets in 89 local health areas of the province by beverage type. Multi-level regression models were used to examine the relationship between per capita alcohol sales and outlet densities for different types of liquor outlet after adjusting for potential confounding social, economic and demographic factors as well as spatial and temporal autocorrelation. Liquor outlets in 89 local health areas of British Columbia, Canada. The number of private stores per 10,000 residents was associated significantly and positively with per capita sales of ethanol in beer, coolers, spirits and wine, while the reverse held for government liquor stores. Significant positive effects were also identified for the number of bars and restaurants per head of population. The percentage of liquor stores in private versus government ownership was also associated significantly with per capita alcohol sales when controlling for density of liquor stores and of on-premise outlets (P < 0.01). The trend towards privatisation of liquor outlets between 2003/04 and 2007/08 in British Columbia has contributed to increased per capita sales of alcohol and hence possibly also to increased alcohol-related harm.
Regression analyses were performed on per capita wine consumption in four states of the U.S.A. where the availability of wines was increased. In three of four states, wine consumption was found to be significantly greater in the years of the changes than one would expect, based on the regression trends of the previous eight to thirteen years. Differences in findings were probably a reflection of differences in the nature and extent of the increased availability among the four states. Consumption of beer and spirits was also analyzed for the years in question, and substitution from beer or spirits to wine was not found. Finally, alternative explanations for the patterns of increases were examined and considered unlikely.
Results from the 1995 National College Health Risk Behavior Survey, which monitored health risk behaviors among US college and university undergraduates, suggest that many students' behaviors increase their likelihood of adverse health outcomes. During the 30 days preceding the survey, 34% of the participants had consumed five or more alcoholic drinks on at least one occasion, and 27% had drunk alcohol and driven a car. Thirty-one percent had smoked cigarettes regularly during their lifetimes, 49% had ever used marijuana, 30% had used a condom during their last sexual intercourse, 21% were overweight, and 38% had participated in vigorous physical activity on 3 or more of the 7 days preceding the survey. These data were analyzed by gender, age group, race and ethnicity, and institution type. They can be used by those responsible for the health and education of college students to reduce risks associated with the leading causes of mortality and morbidity.
To determine whether alcohol outlet density was correlated with heavy and frequent drinking and drinking-related problems, we compared ecological measures of outlet density with survey measures of drinking using a geographic information system and the Harvard School of Public Health College Alcohol Study (n=3,421, site n=8). We identified 966 outlets within 8 2-mile study areas. Densities/site ranged from 32 to 185. Density was correlated with heavy drinking (r=0.82, p=0.01), frequent drinking (r=0.73, p=0.04) and drinking-related problems (r=0.79, p=0.02). Women, underage students and students who picked up binge drinking in college were affected. Implications for prevention and research are discussed.