District of Columbia: Taxes and Other Strategies to Reduce Alcohol Abuse
(Letter Report, 05/19/1998, GAO/GGD/HEHS-98-140).
The Fiscal Year 1998 District of Columbia Appropriations Act mandated
that GAO study the taxation and regulation of alcoholic beverages in the
District. This report compares the District's taxes on alcoholic
beverages with those of neighboring Maryland and Virginia; determines
whether the District's alcoholic beverage tax structure can be brought
into closer conformity with those in surrounding jurisdictions;
determines how much higher the Districts alcohol excise tax rates would
be if they had been indexed for inflation; determines whether existing
empirical research indicated that raising the District's taxes is likely
to curb alcohol abuse; identifies which states earmark their alcohol
taxes for specific purposes; and describes characteristics of effective
alcohol prevention programs and regulatory policies, especially with
regard to youth, that the District government could consider adopting.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: GGD/HEHS-98-140
TITLE: District of Columbia: Taxes and Other Strategies to Reduce
Alcohol Abuse
DATE: 05/19/1998
SUBJECT: Alcohol taxes
Alcohol or drug abuse problems
Teenagers
Municipal governments
Price regulation
Inflation
Comparative analysis
IDENTIFIER: District of Columbia
Consumer Price Index
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
GAO/GGD/HEHS-98-140
Cover
================================================================ COVER
Report to Congressional Committees
May 1998
DISTRICT OF COLUMBIA - TAXES AND
OTHER STRATEGIES TO REDUCE ALCOHOL
ABUSE
GAO/GGD/HEHS-98-140
Taxes and Other Strategies to Reduce Alcohol Abuse
(268833, 108355)
Abbreviations
=============================================================== ABBREV
ABC - Alcohol Beverage Control
APRA - Addiction Prevention and Recovery Administration
BAC - blood alcohol concentration
DISCUS - Distilled Spirits Council of the United States
MLDA - minimum legal drinking age
NIAAA - National Institute on Alcohol Abuse and Alcoholism
PRC - Prevention Research Center
SAMHSA - Substance Abuse and Mental Health Services Administration
SAPE - Substance Abuse Prevention Education
Letter
=============================================================== LETTER
B-279037
May 19, 1998
The Honorable Robert Byrd
Ranking Minority Member
Committee on Appropriations
United States Senate
The Honorable Lauch Faircloth
Chairman
The Honorable Barbara Boxer
Ranking Minority Member
Subcommittee on the District of Columbia
Committee on Appropriations
United States Senate
The Honorable Charles H. Taylor
Chairman
The Honorable James P. Moran
Ranking Minority Member
Subcommittee on the District of Columbia
Committee on Appropriations
House of Representatives
Alcohol abuse and alcohol dependence create major health problems and
social and economic consequences in the United States. They not only
cost the nation billions of dollars annually in health care costs and
reduced or lost productivity, they also contribute to homicides,
rapes, and other violent crimes; risky sexual behavior; traffic
crashes; injuries; and premature deaths. From 1986 to 1990, the
District of Columbia's average annual rate of alcohol-related deaths
was almost twice as high as the national rate. Alcohol use by
adolescents is of particular concern, because in addition to causing
immediate problems, it can lead to a lifelong pattern of alcohol
misuse. Over one-third of District of Columbia high school students
surveyed in 1995 reported they had drunk alcohol in the last 30 days,
and 13 percent reported episodic heavy drinking.
The Fiscal Year 1998 District of Columbia Appropriations Act mandated
us to study issues relating to the taxation and regulation of
alcoholic beverages in the District. Accordingly, in this report,
our objectives were to
(1) compare the District's taxes on alcoholic beverages with those of
surrounding jurisdictions and other states;
(2) determine whether the District's alcoholic beverage tax structure
can be brought into closer conformity with those in surrounding
jurisdictions;
(3) determine how much higher the District's alcohol excise tax rates
would be if they had been indexed for inflation;
(4) determine whether existing empirical research indicates that
raising the District's alcohol taxes is likely to reduce alcohol
abuse, particularly among youths, and related health problems;
(5) identify which states earmark their alcohol taxes for specific
purposes; and
(6) describe characteristics of effective alcohol prevention programs
and regulatory policies, especially with regard to youth, that the
District government could consider adopting.
BACKGROUND
------------------------------------------------------------ Letter :1
Although about 60 percent of the U.S. population drinks alcoholic
beverages without serious consequences, the misuse of alcohol by
another 10 percent of the population has significant negative effects
on the social, economic, and health status of both those who abuse it
and society at large. According to the National Institute on
Alcoholism and Alcohol Abuse (NIAAA), about 14 million Americans meet
the medical diagnostic criteria for alcohol abuse or alcoholism, and
an estimated 100,000 alcohol-related deaths occur each year. About
half of the nation's high school students report current alcohol use,
a significant minority drink heavily, and few have difficulty
obtaining alcohol. The younger the age of drinking onset, the
greater the chance that an individual will develop a clinically
defined alcohol disorder at some point in life.
All states heavily regulate the sale of alcoholic beverages. Some
states take a more direct approach to controlling sales than others.
Eighteen states, including Virginia, are generally referred to as
"control" states, because the final sale to consumers of, typically,
liquor and in some cases wine and beer as well can occur only in
state-operated stores at prices established by state beverage control
boards.\1 State-operated stores are the exclusive retailers of all
legal liquor sold for off-premises consumption throughout Virginia.
Final prices are set by Virginia's Department of Alcohol Beverage
Control (ABC).\2
The District of Columbia and the remaining 32 states, including
Maryland, are referred to as "license" states, because the
distribution and sale of alcoholic beverages is carried out by
private license holders. Maryland, however, is not completely a
license state, because 1 of its 22 counties--Montgomery County, which
borders the District--is a control jurisdiction. The county
Department of Liquor Control is the exclusive wholesaler of all
alcoholic beverages sold within its boundaries. The Department is
also the exclusive retailer of liquor sold for off-premises
consumption.
To raise revenue and to help prevent alcohol misuse, most states and
the District of Columbia levy both excise taxes and retail sales
taxes on alcoholic beverages. Some county and city governments also
impose their own alcohol taxes. In order to determine whether the
District's taxes on alcohol are greater or less than those in
surrounding jurisdictions and other states, one needs to compare the
District's combined--sales and excise--tax structure with the tax
structures in the other jurisdictions. Such a comparison is
complicated by the fact that alcohol excise tax rates are generally
stated as fixed amounts per unit of volume, but sales tax rates are
almost always ad valorem (stated as percentages of product prices).
The combined tax burden within a given jurisdiction will vary by
type, size, and value of beverage and, in some cases, by type of
retail establishment.\3
It is also difficult to compare the alcohol tax systems in license
jurisdictions with those in control jurisdictions. In a license
state, private sector wholesalers determine wholesale prices, taking
into account their costs, including the excise taxes that they pay to
the state. Private sector retailers charge their own mark-ups on top
of those wholesale prices. For these states, one can readily
determine what share of the final price of the alcoholic beverage is
attributable to the excise tax.
In a control state such as Virginia, where the government acts as a
monopoly wholesaler and retailer of liquor, the government collects
revenue from the sale of alcohol in two ways: (1) by earning the
profits that private sector wholesalers and retailers would otherwise
have earned and (2) by imposing an excise tax on the alcohol. One
cannot easily determine how much of the combined revenue that the
state collects should be considered the profits of the state's
alcohol sales operations and how much should be considered an excise
tax. The state may make a distinction between the price mark-up that
it charges and the excise tax that it levies, but this distinction
has little meaning to consumers or taxpayers. There is a wide range
of mark-up/excise tax rate combinations that the state could have set
to yield the same amount of revenue and impose the same costs on
consumers.
One way to define an "effective" excise tax rate on alcohol in a
control state is to say that it equals the statutory excise tax rate,
plus any cost that the state control system imposes on consumers
above what those consumers would have borne if that system were not
in place.\4 For example, if a state-run store imposes a higher
mark-up than a private dealer would have, then this additional cost
to consumers can be considered part of the effective tax. The
effective tax rate defined in this manner provides a better basis for
comparison with the excise tax rates imposed in license states.
--------------------
\1 This classification of states as either "control" or "license"
states is made by the Distilled Spirits Council of the United States
in its publication, "Summary of State Laws and Regulations Relating
to Distilled Spirits," 29th edition, 1996. The main categories of
alcoholic beverages are beer, wine, and distilled spirits. The term
"liquor" is used in this report to mean distilled spirits.
\2 Beer and wine for off-premises consumption are sold by licensed
private stores. Private establishments that serve alcoholic
beverages for on-premises consumption must buy their liquor, but not
their beer or wine, from a state store. Appendix I provides details
on Virginia's and Montgomery County's pricing practices.
\3 We use the term "tax burden" to mean the share of tax in the final
cost of the beverage.
\4 By an "effective" tax rate we mean one that takes into account
other provisions of the law or other government policies, which are
not reflected in the statutory tax rates, that (1) impose an
additional cost on taxpayers and (2) raise revenue for the
government.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :2
Compared to the taxes levied in nearby Virginia jurisdictions, the
District's combined tax rates are higher for almost all beers and for
relatively high-priced wines, but the opposite is true for relatively
low-priced wines. Compared to the taxes levied in all Maryland
counties except Montgomery County, the District's combined sales and
excise tax rates are higher for all types of alcoholic beverages.
Effective tax rates on controlled beverages sold exclusively through
government-run stores in Virginia and Montgomery County, Maryland,
may differ from statutory tax rates, so it is not possible to make
precise comparisons with the District's statutory tax rates for these
items. Although the District's excise taxes on alcoholic beverages
are lower than those in most of the 50 states, the District's
combined tax rates on beer and wine are higher than those in most
states, because its sales tax is among the highest. The District's
combined taxes on liquor are higher, for at least some items, than in
most of the states for which we could make comparisons readily.
The District cannot conform its alcohol tax structure to those in all
surrounding jurisdictions at the same time, because the tax
structures among those neighboring jurisdictions differ
significantly. Moreover, the District would not be able to impose
exactly the same effective tax rates as those in either Virginia or
Montgomery County, because those rates are difficult to estimate
precisely, might change frequently, and are likely to vary across
beverage items.\5 The District also would have difficulty enforcing
an ad valorem liquor excise tax like Virginia's, which it would have
to collect from private retailers. Virginia does not face
enforcement problems, because it collects the tax from its state-run
stores.
Although the District's per-unit excise tax rates have declined in
inflation-adjusted terms since they were last changed, increases in
the District's ad valorem special sales tax rates on all alcoholic
beverages have, to date, more than compensated for the lack of
indexation of the excise tax rates. The District's alcohol excise
taxes will continue to decline gradually in inflation-adjusted terms
as long as inflation continues and excise taxes remain unchanged. In
order to keep the real value of its combined tax rates on alcohol
close to what they are currently, the District would have to increase
its excise or sales tax rates periodically, in step with inflation.
Taxes on alcohol are a means of reducing alcohol misuse while at the
same time raising revenue. Economic theory and empirical evidence
indicate that (1) higher alcohol taxes lead to higher consumer prices
for alcoholic beverages, and (2) higher prices reduce alcohol
consumption. Furthermore, higher alcohol taxes and prices have been
associated with declines in drunken driving, motor vehicle
fatalities, rapes, and robberies. Increases in the District's
alcohol taxes are thus likely to reduce alcohol use, particularly
among youths, but the special geographic circumstances of the
District--where all of its suburbs are in other jurisdictions--could
weaken these effects. As of 1993, 24 states earmarked at least a
portion of their alcohol excise tax revenues for specific purposes,
15 of them directing their use to alcohol treatment, substance abuse,
and/or mental health programs.
Research on other alcohol prevention strategies has not adequately
demonstrated the effectiveness of many common techniques and
programs. Most programs to educate individuals and build their
resistance skills have not been evaluated rigorously or over a
sufficient length of time to determine their effectiveness. Although
evaluations have shown that some recent efforts to combine those
programs with strategies that focus on legal and regulatory controls
have been effective, evaluations of other similar studies are still
under way. More rigorous evaluations of prevention strategies and
programs could provide better information about which approaches are
most likely to succeed.
The best current evidence suggests that several legal and regulatory
strategies, when enforced, can reduce illegal drinking and
alcohol-related problems. According to researchers and local
government officials, visible enforcement coupled with education
about these laws and regulations should create a stronger deterrent
effect. Officials in the District of Columbia and other local
jurisdictions have tried innovative enforcement techniques they
believe were successful, although formal evaluations have not been
done. Budget and staffing constraints have stymied full
implementation of these strategies. The District also funds and
operates a variety of alcohol abuse prevention and treatment
programs.
--------------------
\5 By a "beverage item" we mean a specific beverage type (beer, wine,
Scotch, gin, etc.); brand; and container size.
OBJECTIVES, SCOPE, AND
METHODOLOGY
------------------------------------------------------------ Letter :3
To meet our first two objectives of (1) comparing the District's
taxes on alcoholic beverages with those of surrounding jurisdictions
and other states and (2) determining whether the District's tax
structure can be brought into closer conformity with the tax
structures in surrounding jurisdictions, we reviewed the relevant
laws and regulations of the various jurisdictions. We also
interviewed officials from the District's Office of Tax and Revenue
and from the alcohol beverage control commissions in the District;
Montgomery County, Maryland; and Virginia. We also obtained alcohol
tax information for the 50 states from the Federation of Tax
Administrators and the Distilled Spirits Council of the United States
(DISCUS). In an effort to determine whether the pricing policies of
Montgomery County and Virginia result in effective tax rates that are
higher than the statutory tax rates for certain alcoholic beverages,
we obtained lists of the wholesale prices that Montgomery County
charges on all of the alcoholic beverages that it sells to retailers
in the county and lists of the wholesale prices for wine and liquor
that private sector wholesalers charge retailers in the remainder of
Maryland. The latter prices are published in the Maryland Beverage
Journal. We compared Montgomery County's prices (effective during
January 1998) to those of the private wholesalers for (1) 14 liquor
items identified by the Virginia ABC Commission or DISCUS as top
sellers, (2) a random sample of an additional 28 liquor items from
Montgomery County's price list, and (3) a random sample of 29 wine
items from the county's price list. We could not make a comparison
of beer wholesale prices, because the state of Maryland requires
publication of only liquor and wine wholesale prices, not beer
prices.
As one possible way to determine whether Virginia's controlled prices
for liquor are, on average, higher than the competitive market prices
in the same region, we considered doing a survey comparing prices in
Virginia with those in the District and its Maryland suburbs for the
top-selling liquor items. However, representatives of the District's
Alcohol Retailers Association told us that a common retail pricing
practice is to sell selected popular items at a large discount,
possibly even at a loss, in order to attract customers who will then
also buy items that have higher price mark-ups. For this reason, a
survey limited to just the best-selling items would probably be
misleading. The retailers' representatives also indicated that it
would be very difficult to obtain good estimates of the average
prices in each jurisdiction, because there are so many different
brands and bottle sizes, and the prices for each item are likely to
vary considerably across different types of retail outlets in each
jurisdiction. It was beyond the scope of this study to undertake the
extensive retail price survey and analysis needed to make such
estimates.
To determine how much higher the District's alcohol excise tax rates
would be if they had been indexed for inflation, we raised these
rates by the same percentage increase as the average increase in
alcoholic beverage prices since the last time each rate was changed.
To compute the average price increases, we used the U.S. Department
of Labor's quarterly Consumer Price Index alcoholic beverage
component for the Washington metropolitan area. We also compared the
combined--excise and sales--tax burdens on typical alcoholic
beverages over time.
To determine whether existing empirical research indicates that
raising the District's alcohol taxes is likely to reduce alcohol
abuse, particularly among youths, and related health problems, we
reviewed and summarized several surveys of the relevant academic
literature that have been published in recent years. We then
obtained comments on our summary from five academic and government
experts in this field of research and modified our summary to reflect
those comments.
To identify which states earmarked their alcoholic taxes for specific
purposes, we used the latest summarization of state tax earmarking
prepared by the National Conference of State Legislatures. The
latest available earmarking information is for fiscal year 1993. We
did not independently verify the accuracy of this information.
To describe the characteristics of alcohol prevention programs and
legal and regulatory strategies that researchers have deemed
effective, we identified and reviewed selected literature on alcohol
prevention research and evaluated syntheses of research literature.
Given the vast literature on this subject and the time available to
us, we relied heavily on information presented in NIAAA's 1997
publication, Alcohol and Health.\6 We also interviewed key officials
responsible for overseeing alcohol prevention research for youth and
other populations at NIAAA. To obtain information on the District of
Columbia's alcohol laws and regulations, we interviewed the Executive
Director of the D.C. Alcoholic Beverages Control Board, reviewed
relevant provisions of the District of Columbia's codes and municipal
regulations, and examined reports provided by the Board. To obtain
information on how the District's alcohol laws are being enforced, we
interviewed responsible officials of the D.C. Metropolitan Police
Department. We obtained similar information from officials
responsible for managing and enforcing alcoholic beverage programs in
Montgomery and Prince George's Counties in Maryland and the City of
Alexandria and Arlington County in Virginia. To obtain information
on the District's health and education prevention programs, we
interviewed officials at the Department of Health's Addiction
Prevention and Recovery Administration and the District of Columbia
Public Schools and reviewed key agency documents.
We conducted our review in Washington, D.C.; Virginia; and Maryland
between December 1997 and April 1998 in accordance with generally
accepted government auditing standards. We requested comments on a
draft of this report from the District of Columbia's Office of Tax
and Revenue, Department of Consumer and Regulatory Affairs, and
Department of Health; the District of Columbia Financial
Responsibility and Management Assistance Authority; the Virginia
Department of Alcohol Beverage Control; NIAAA; the Substance Abuse
and Mental Health Services Administration (SAMHSA); and the
Prevention Research Center (PRC). These comments are summarized and
discussed near the end of this letter. We also requested comments
from the Montgomery County Department of Liquor Control but did not
receive any in time to include them in this report.
--------------------
\6 NIAAA, "Prevention of Alcohol Problems," Chapter 9 of Alcohol and
Health: Ninth Special Report to the U.S. Congress, U.S. Department
of Health and Human Services, June 1997, pp. 301-327.
THE DISTRICT'S COMBINED TAX
RATES ARE HIGHER THAN ITS
NEIGHBORS' RATES FOR SOME
BEVERAGES BUT NOT FOR OTHERS
------------------------------------------------------------ Letter :4
The interaction of the sales taxes and the excise taxes within each
jurisdiction results in combined tax burdens that vary by type, size,
and value of beverage and, in some cases, by type of retail
establishment. We computed the combined statutory taxes paid in each
jurisdiction on a range of different beverage items. In comparison
to the taxes levied in all but one Maryland county, the District's
combined statutory tax rates are higher for all types of alcoholic
beverages. In comparison to the taxes levied in adjacent Virginia
jurisdictions, the District's combined statutory tax rates are higher
for almost all beers and for relatively high-priced wines, but the
opposite is true for relatively low-priced wines. The effective tax
rates on liquor in Virginia and on all alcoholic beverages in
Montgomery County may differ from the statutory tax rates as a result
of government controls over prices. Consequently, precise
comparisons of the taxes on those controlled items cannot be made.
THE DISTRICT'S COMBINED
STATUTORY TAX RATES ON BEER
AND SOME WINES ARE THE
HIGHEST IN THE AREA
---------------------------------------------------------- Letter :4.1
The statutory alcohol excise tax rates levied by the District and the
state of Maryland are very similar, as shown in table 1.\7 Virginia's
excise taxes on beer and wine are significantly higher than those in
the District and Maryland. Virginia's liquor tax is not directly
comparable to taxes in the other two jurisdictions�both because it
has an ad valorem rate and because the state sets the price upon
which the tax is computed. There are no local government alcohol
excise taxes in either Maryland or Virginia.\8
Table 1
Statutory Taxes on Alcoholic Beverages
Imposed by the District of Columbia,
Maryland, and Virginia
Sales Sales
tax tax
Beer excise Wine excise Liquor excise (off- (on-
Jurisdictions rates rates rates site) site)
-------------- -------------- -------------- ---------------- ------ ------
District of $.09/gal. $.30/gal. $1.50/gal. 8% 10%
Columbia (alcohol 14%
or less)
$.40/gal.
(alcohol over
14%)
$.45/gal.
(sparkling
wine)
Maryland $.09/gal. $.40/gal. $1.50/gal.\a 5% 5%
Virginia $.283/gal. $1.51/gal 20% tax on the 4.5%\c 4.5%\d
(in 12 oz. sum of supplier
bottles)\b price, including
a handling
charge, plus a
mark-up
--------------------------------------------------------------------------------
\a For liquor over 100 proof, $.015 is added for each proof point
over 100.
\b Virginia tax rate varies by container size. For example, the rate
per gallon for beer in barrels is $.2565.
\c The statewide rate is 3.5 percent, but local governments can
choose to have the state collect an additional 1 percent, which is to
be returned to those local governments. Arlington and Fairfax
Counties and the City of Alexandria have all chosen this option.
\d Does not include local "meal taxes" (on alcoholic beverage served
separately or with meals) of 4 percent in Arlington County and 3
percent in the City of Alexandria. Fairfax County has no such tax.
Sources: Excise and sales taxes from District of Columbia and state
tax codes. "Meal tax" rates from telephone communications with
Treasurers' offices of the several jurisdictions.
The District imposes an 8 percent sales tax on alcoholic beverages
sold for off-premises consumption; it levies a 10-percent tax on
alcoholic beverages and food sold for on-premises consumption. These
rates are higher than the 5.75-percent rate the District imposes on
the sales of most goods. They are also significantly higher than
Maryland's and Virginia's sales taxes.\9
The interaction of the sales taxes and the excise taxes within each
jurisdiction results in combined tax burdens that vary by type, size,
and value of beverage and, in some cases, by type of retail
establishment. In order to compare the combined statutory tax
burdens on alcohol across jurisdictions, we have computed the taxes
that would be paid on three beverage items sold for off-premises
consumption and on three items sold for on-premises consumption. In
order to demonstrate the importance of prices in the calculation of
combined tax burdens, we used a wide range of prices for each
beverage item. (See tables 2 and 3.)
Table 2
Combined Taxes Paid on Selected
Alcoholic Beverage Items in the District
and Surrounding Jurisdictions for Off-
Premises Consumption
Beer (six- Wine (750-ml Liquor (1-
pack) bottle) liter bottle)
-------------- -------------- --------------
Pre-sales tax price (a low-
price and a high-price example
for each beverage) $4.00 $7.00 $4.00 $12.00 $8.00 $30.00
-------------------------------- ------ ------ ------ ------ ------ ------
District of Columbia
excise tax .05 .05 .06 .06 .40 .40
sales tax .32 .56 .32 .96 .64 2.40
================================================================================
Combined tax .37 .61 .38 1.02 1.04 2.80
Maryland
excise tax .05 .05 .08 .08 .39 .39
sales tax .20 .35 .20 .60 .40 1.50
================================================================================
Combined tax .25 .40 .28 .68 .79 1.89
Virginia
excise tax .16 .16 .30 .30 1.33 5.00
sales tax .18 .32 .18 .54 .36 1.35
================================================================================
Combined tax .34 .48 .48 .84 1.69 6.35
--------------------------------------------------------------------------------
Note: The price shown at the top of each column is the pre-sales
tax, or "shelf," price for the item. This price already reflects
federal and state or District excise taxes. The first price for each
beverage items is meant to represent the price for a relatively
inexpensive brand; the second price represents that of a relatively
expensive brand. These prices are not meant to represent the
absolute lowest and highest prices likely to be found in the
Washington metropolitan area.
Source: GAO computations based on information from the state and
District of Columbia tax codes.
Table 3
Combined Taxes Paid on Selected
Alcoholic Beverage Items In the District
and Surrounding Jurisdictions for On-
Premises Consumption
Beer (12-oz Wine (6-oz Liquor (2-oz
bottle) glass) shot)
-------------- -------------- --------------
Pre-sales tax price (a low-
price and a high-price example
for each beverage) $2.00 $4.00 $2.00 $4.00 $2.00 $7.50
-------------------------------- ------ ------ ------ ------ ------ ------
District of Columbia
Excise tax .01 .01 .01 .01 .02 .02
Sales tax .20 .40 .20 .40 .20 .75
================================================================================
Combined tax .21 .41 .21 .41 .22 .77
Maryland
Excise tax .01 .01 .02 .02 .02 .02
Sales tax .10 .20 .10 .20 .10 .38
================================================================================
Combined tax .11 .21 .12 .22 .12 .40
Virginia
Excise tax .03 .03 .07 .07 .08 .30
Sales tax\a .17 .34 .17 .34 .17 .64
================================================================================
Combined tax .20 .37 .24 .41 .25 .94
--------------------------------------------------------------------------------
Note: The price shown at the top of each column is the pre-sales
tax, or "shelf," price for the item. This price already reflects
federal and state or District excise taxes. The first price for each
beverage items is meant to represent the price for a relatively
inexpensive brand; the second price represents that of a relatively
expensive brand. These prices are not meant to represent the
absolute lowest and highest prices likely to be found in the
Washington metropolitan area.
\a The Virginia sales tax rate that we use in this table is 8.5
percent, which includes the 4.5 percent state tax plus the 4 percent
meals tax in Arlington County. In the City of Alexandria the total
sales tax rate is 7.5 percent, and in Fairfax County it is 4.5
percent.
Sources: GAO computations based on data from the state and District
of Columbia tax codes and Treasurers' offices of surrounding
jurisdictions. The District's combined statutory taxes on alcohol
are higher than those levied in adjacent Maryland counties for all
types of beverages and across all the price ranges that we examined.
The District's combined taxes on beer are also higher than those in
adjacent Virginia jurisdictions. In contrast, Virginia's combined
statutory taxes on liquor are higher than those in the District
across the full range of prices we examined. For wine, Virginia's
combined taxes are higher than the District's, except at the high end
of the price range.\10
--------------------
\7 The statutory tax rates in table 1 and the combined tax burdens in
tables 2 and 3 do not reflect any effects that the alcohol control
systems in Montgomery County or Virginia may have on the final prices
of alcoholic beverages.
\8 The Maryland Code allows one exception for Garrett County to tax
beer.
\9 Alcoholic beverages in those two states are subject to the basic
sales tax rates. Some Virginia jurisdictions neighboring the
District levy their own "meal taxes," which cover alcoholic beverages
consumed on-premises. Arlington County's meal tax is 4 percent and
the City of Alexandria's tax is 3 percent, so the total on-premises
sales tax rates in those two jurisdictions are 8.5 percent and 7.5
percent.
\10 In table 3, the combined taxes that we show for Virginia are
those in effect in Arlington County, the jurisdiction having the
longest border with the District. The combined taxes in the City of
Alexandria and in Fairfax County are lower than those in Arlington
County.
GOVERNMENT PRICE MARK-UPS
MAY ALTER THE "EFFECTIVE"
TAX RATES ON ALCOHOL
---------------------------------------------------------- Letter :4.2
The monopoly power that the governments of Virginia and Montgomery
County have over the sale of some alcoholic beverages within their
boundaries provides them with an opportunity to set prices for those
beverages to achieve objectives, such as discouraging alcohol
consumption or maximizing the government's monopoly profits, that
either would not be considered by private sector businesses, or would
not be achievable in a competitive market.\11 If the government's
pricing policy results in final prices to consumers that are higher
than those that private sector businesses would charge, then the
effective tax rates on those beverage items exceed the statutory tax
rates. Conversely, if the government's pricing policy results in
lower final prices, then the effective tax rates are lower than the
statutory tax rates.
The explanations of Virginia's and Montgomery County's pricing
practices given to us by government officials did not provide a
sufficient basis to allow us to say whether the controlled prices in
those jurisdictions are likely to be higher or lower than those that
would have existed without government controls. Officials from the
Virginia Department of Alcohol Beverage Control told us that the goal
of their pricing policy is to generate a reasonable rate of return
for the state. The prices for liquor items sold in state stores are
standard across the state. The officials told us that these
statewide prices are not greatly influenced by price competition from
the District. An official from Montgomery County's Department of
Liquor Control told us that it was his understanding that the
county's wholesale price mark-ups for alcoholic beverages are
intended to reflect the prevailing mark-up practices in the industry.
Information provided by Virginia and Montgomery County officials and
by representatives of the District's Alcohol Retailers Association,
whom we interviewed, suggests that the relationship between
controlled and free market prices is likely to vary across beverage
items. Our limited comparison of liquor and wine wholesale prices in
Montgomery County and the remainder of Maryland supports this idea.
We do not have sufficient data to determine whether the controlled
prices in Virginia and Montgomery County are, on average, higher or
lower than those that would have existed without the controls.
Consequently, we are not able to say whether Virginia's effective tax
rates on liquor are higher or lower than the statutory rates, or
whether the effective tax rates on all alcoholic beverages in
Montgomery County are higher or lower than the statutory rates.\12
--------------------
\11 Virginia sets the retail prices for all liquor items sold in the
state for off-premises consumption. Bars and restaurants in the
state must also purchase their liquor from the state at these same
prices. Montgomery County sets the wholesale prices for all
alcoholic beverages sold in the county.
\12 Appendix I provides information on Virginia's and Montgomery
County's pricing practices and the results of our wholesale price
comparison.
THE DISTRICT'S COMBINED TAXES
ON BEER AND WINE ARE HIGHER
THAN THOSE IN MOST STATES
------------------------------------------------------------ Letter :5
Although the District's excise taxes on alcoholic beverages are lower
than those in most of the 50 states, the District's combined tax
rates on beer and wine are higher than those in most states because
its sales tax is among the highest. The District's combined taxes on
liquor are higher, for at least some items, than in most of the
states for which we could readily make comparisons.
Our analysis of data published by the Federation of Tax
Administrators indicates that the District's excise tax rate on beer
is lower than the rates in 40 states; its excise tax rate on liquor
is tied with that of Maryland as the lowest among the 32 states that
do not control the sale of liquor; and its excise tax rate on wine is
lower than the rates in 38 out of the 46 states that do not control
the sale of wine (see app. II). It was beyond the scope of this
study to estimate effective excise tax rates for those states that
control the sales of liquor and/or wine.
In contrast to the excise taxes, the District's sales taxes on
alcoholic beverages are higher than those in 45 states and equal to
those in another state. For the remaining four states, the
comparison is mixed--depending on whether the sales are for
on-premises or off-premises consumption. The tax rates shown in
appendix II do not include any taxes that may be levied by
governments below the state level.
We computed combined alcohol tax burdens for most of the states in a
manner similar to our computations for the District, Maryland, and
Virginia in tables 2 and 3. We did not have sufficient time to
incorporate all of the complexities of the alcohol taxes of some
states into our computations, so we left those states out of our
comparison. Consequently, although we can say that the District's
combined taxes on beer and wine are higher than those in most states
(for the price ranges we examined), we cannot say exactly where the
District ranks. We could not make adequate comparisons with enough
states to say whether or not the District's combined taxes on liquor
are higher than those in most states. However, for the states with
which we could make comparisons, the District's combined taxes on
liquor were higher than most of them, for at least part of the price
range we examined.\13
--------------------
\13 We were able to make comparisons with the tax rates on sales for
off-premises consumption in 30 states and the tax rates on sales for
on-premises consumption in 28 states.
THE DISTRICT WOULD HAVE SOME
DIFFICUTLY CONFORMING ITS
ALCOHOL TAXES TO THOSE IN
MARYLAND OR VIRGINIA
------------------------------------------------------------ Letter :6
There are significant differences among the alcohol tax structures of
the various jurisdictions surrounding the District. The District
cannot conform to all of those tax structures at the same time.
Moreover, the District would not be able to impose exactly the same
effective tax rates that exist in either Virginia or Montgomery
County, because those effective rates are difficult to estimate
precisely; might change frequently; and are likely to vary by
beverage type, brand, and container size. In order to conform its
combined statutory tax rates to those in Maryland, the District would
have to lower its taxes on all alcoholic beverages. The only way for
the District to make its statutory taxes similar to those in
Virginia, across the entire range of alcoholic beverages, would be to
adopt a single sales tax rate for alcoholic beverages that is close
to Virginia's and to adopt excise tax rates that are all very close
to Virginia's. This change would lower taxes on beer and high-priced
wine in the District while raising the tax on liquor and low-priced
wine.
The District would face a difficult enforcement task if it adopted
the ad valorem excise tax rate on liquor that Virginia currently
levies. Virginia's liquor excise tax is computed as 20 percent of
the liquor price after the state has taken its combined wholesale and
retail price mark-up. To levy an equivalent tax, the District would
have to impose the 20-percent rate on the final prices that retail
stores charge their customers before applying the sales tax. This
tax would be more difficult to enforce than the District's current
liquor excise tax, because (1) it would have to be collected from a
much greater number of taxpayers; (2) taxpayers would have more of an
incentive to understate their liquor sales, because the tax rate paid
by each taxpayer would be higher than it currently is; and (3) to
verify compliance, District auditors would have to examine a
taxpayer's detailed sales receipts. Virginia does not face these
enforcement difficulties, because it collects the tax from its own
stores. No license state levies an ad valorem alcohol excise tax.
It would be difficult for the District to devise an ad valorem liquor
tax that closely approximates the tax that Virginia imposes on liquor
sold for on-premises consumption. Restaurants and bars in Virginia
buy their liquor from state-run stores at retail prices that already
reflect the 20 percent excise tax. Those prices are not equivalent
to either the wholesale prices that restaurants and bars in the
District pay for their liquor or to the prices that District
restaurants and bars charge their final customers. Consequently, the
District could not replicate Virginia's tax simply by imposing a tax
rate on either the wholesale or final prices. The District would
have to define a new tax base that approximates the state-determined
prices on which Virginia's excise tax is imposed.
We are unable to say whether average alcohol prices and revenues from
alcohol taxes would increase or decrease if the District conformed
its statutory alcohol tax rates to those in Virginia. In order to
estimate the effects on average prices and revenues, we would need to
know the distribution of alcohol sales in the District, by beverage
type and by price range. However, this information does not exist.
The total value of alcohol sales in the District each year is
unknown.\14 Without knowing how average prices would be affected, it
is not possible to predict how total alcohol consumption in the
District would be affected by this specific change in the District's
tax structure. The decrease in the combined tax on beer likely would
cause beer consumption to increase, and the increase in the combined
tax on liquor likely would cause liquor consumption to decline. In
order to determine whether such a trade-off would be desirable, we
would have to be able to estimate the relative sizes of the changes
in beer and liquor consumption. The data needed to make such
estimates do not exist. National surveys indicate that beer is the
alcoholic beverage of choice among youths who drink alcoholic
beverages. There is also some evidence that beer is
disproportionately preferred by those who drink a lot during a
typical session and that drinkers who prefer beer are more likely to
drive while intoxicated than those who prefer wine or liquor.\15
--------------------
\14 Although it is possible to determine the total value of alcohol
sales in the District taxed at the 8-percent rate, which applies only
to sales of alcohol for off-premises consumption, according to
District officials, it is not possible to determine the total value
of alcohol sales taxed at the 10-percent rate. That latter rate
applies not only to sales of alcohol for on-premises consumption; it
also applies to sales of food in restaurants, bars, and other
establishments, as well as to car rental agreements and prepaid phone
cards.
\15 See Coate, Douglas and Michael Grossman, "Effects of Alcoholic
Beverage Prices and Legal Drinking Ages on Youth Alcohol Use,"
Cambridge, MA: National Bureau of Economic Research, Working Paper
No. 1852, 1986; and Berger, Dale and John Snortum, "Alcoholic
Beverage Preferences of Drinking-Driving Violators," Journal of
Studies on Alcohol, 46(3), 1985, pp. 232-239.
SALES TAX INCREASES HAVE MORE
THAN OFFSET THE EFFECT OF
INFLATION ON EXCISE TAX RATES
FOR MOST ALCOHOLIC BEVERAGES
------------------------------------------------------------ Letter :7
All of the District's per-unit excise tax rates have declined in
inflation-adjusted terms since they were last changed. During this
time, however, the District increased its ad valorem special sales
tax rates on all alcoholic beverages. For most of the beverage items
we examined, the increases in the sales tax rates have, to date, more
than compensated for the lack of indexation of the excise tax rates.
The District's excise tax rates for beer, wine, and liquor were last
changed in 1989, 1990, and 1978, respectively. Each has declined in
inflation-adjusted terms since those last changes. Table 4 shows
what the rates would have been at the time we did our review if they
had been increased to keep pace with inflation. The differences
between the current tax rates and the inflation-adjusted rates for
beer and wine are relatively small, because the last changes were
relatively recent and price inflation has been very moderate. In
contrast, the inflation-adjusted excise tax rate on liquor would be
more than twice as high as the current rate.
Since the time that all of these excise tax rates were last
increased, two changes have been made to the special sales tax rates
for alcoholic beverages. In 1992 the sales tax rate for alcoholic
beverages sold for off-premises consumption was increased from 6
percent to 8 percent. In 1994 the rate for on-premises consumption
was increased from 9 to 10 percent. There was one additional change
in the on-premises sales tax rate--from 8 to 9 percent in 1989--since
1978, when the liquor excise tax rate was last changed.\16
Table 4
Current and Inflation-Adjusted Excise
Tax Rates
Fiscal
year
current Tax rate
rate was Current tax adjusted
Beverage introduced rate for inflation
---------------------------- ---------- ------------ --------------
Beer 1989 $.09 per $.11 per
gallon gallon
Wine 1990 $.30 per $.34 per
gallon gallon
Liquor 1978 $1.50 per $3.14 per
gallon gallon
----------------------------------------------------------------------
Source: GAO computations based on tax rates provided by the District
of Columbia's Office of Tax and Revenue and the U.S. Department of
Labor's Consumer Price Index's alcoholic beverage component for the
Washington metropolitan area.
The increases in the sales tax rates have more than compensated for
the lack of indexation of the excise tax rates for most of the
beverage items we examined. For each alcoholic beverage item
included in tables 2 and 3, we computed the tax burdens at two points
in time--now and the date at which the excise tax on that item was
last changed. We also computed each tax burden as a percentage of
the final price of the item. (See table 5.) Only for the
lowest-priced liquor item sold for off-premises consumption has the
tax burden declined noticeably since the earlier date. Currently,
the District's combined taxes on liquor account for 13.0 percent of
the final price that consumers pay for an $8.00, 1-liter bottle of
liquor. In 1978, the District's combined taxes on that same bottle
accounted for 15.7 percent of its final price.\17
Table 5
Current Combined Tax Burdens Compared to
Those When Current Excise Tax Rates Were
First Introduced
Combined tax
on
alcohol as a
percentage of
Current final
combined price, at the
tax time
on alcohol that current
Retail as a excise
price percentage tax rates were
in of first
Beverage item 1998 final price introduced
------------------------------------------ ------ ------------ --------------
Off-site consumption
Six-pack of beer $4.00 9.3% 7.2%
Six-pack of beer 7.00 8.7 6.5
750-ml bottle of wine 4.00 9.5 7.4
750-ml bottle of wine 12.00 8.5 6.2
1-liter bottle of liquor 8.00 13.0 15.7
1-liter bottle of liquor 30.00 9.3 8.5
On-site consumption
12-ounce bottle of beer 2.00 10.4 7.9
12-ounce bottle of beer 4.00 10.2 7.7
6-ounce glass of wine 2.00 10.7 8.2
6-ounce glass of wine 4.00 10.4 7.8
2-ounce shot 2.00 11.2 9.9
2-ounce shot 7.50 10.3 8.1
--------------------------------------------------------------------------------
Sources: GAO computations based on tax rates provided by the
District of Columbia's Office of Tax and Revenue and the U.S.
Department of Labor's Consumer Price Index's alcoholic beverage
component for the Washington metropolitan area.
The District's alcohol excise taxes will continue to decline
gradually in inflation-adjusted terms. In order to keep the real
value of its combined tax rates on alcohol close to what they are
currently, the District would have to increase its excise tax rates
periodically, in step with inflation.
--------------------
\16 Dates of sales tax rate changes were provided by the District of
Columbia's Office of Tax and Revenue.
\17 The final price equals the retail price plus the sales tax paid.
Thus, the final price for the $8.00 bottle is currently $8.64. To
compute the final prices for the earlier dates, we deflated the
retail price for each item (after subtracting out the excise tax) so
that it would represent the pretax price that existed at the time
that the applicable excise tax rate was last changed. We then added
the appropriate excise and sales taxes to obtain the final price.
AN INCREASE IN THE DISTRICT'S
ALCOHOL TAXES WOULD LIKELY
REDUCE ALCOHOL CONSUMPTION AND
SOME RELATED PROBLEMS
------------------------------------------------------------ Letter :8
Taxes on alcohol are a means of reducing alcohol misuse while at the
same time raising revenue. Economic theory and empirical evidence
indicate that higher alcohol taxes increase the prices for alcoholic
beverages, and higher prices affect alcohol consumption. However,
there is some uncertainty regarding the extent to which the taxes are
passed through to consumers as higher prices, and empirical research
estimates vary on the degree to which changes in prices affect
alcohol consumption.
Researchers have found that changes in alcohol prices affect most
categories of drinkers, but to different degrees. Youths and young
adults appear to be more sensitive to price than older adults. Some
recent studies have found that higher alcohol taxes and prices are
associated with declines in drunken driving, motor vehicle
fatalities, rapes, and robberies. The special geographic
circumstances of the District--where all of its suburbs are in other
jurisdictions--may serve to weaken the effect that an increase in the
District's taxes would have on local alcohol consumption.
ALCOHOL TAXES INCREASE
ALCOHOL PRICES, BUT THE
RELATIONSHIP VARIES WITH
CIRCUMSTANCES
---------------------------------------------------------- Letter :8.1
Economists have suggested that the extent to which any excise tax
increase is passed along to consumers varies, depending on the
characteristics of the markets where consumers purchase their
beverages. Such characteristics would include how much competition
among sellers exists in the markets. There is a clear presumption in
the economic literature that in the long run, under perfectly
competitive market conditions, tax increases on consumer goods are
completely passed along by producers, wholesalers, and retailers to
the final consumers in the form of higher prices for the taxed goods.
The alcohol industry, however, does not operate under purely
competitive conditions. Researchers have concluded that the alcohol
industry is oligopolistic, meaning that it is dominated by a few
large suppliers.\18 There is no generally accepted theory of how
prices are determined in an oligopolistic industry; therefore, the
exact extent to which alcohol excise taxes will affect the prices of
alcoholic beverages is uncertain. Nor is there clear empirical
evidence to indicate what portion of a tax increase will be passed
along to alcoholic beverage consumers in the form of increased
prices. Some researchers have estimated that past alcohol excise tax
increases have caused prices to increase by more than the full amount
of the tax increase. Other economists have suggested that the extent
to which excise tax increases are passed along to consumers varies
depending on the characteristics of the markets where consumers
purchase their beverages. Such characteristics include how
responsive market demand is to price changes and how much competition
among sellers exists in the markets. In the absence of more
conclusive evidence, most researchers trying to model the effects of
taxes and prices on alcohol consumption assume that an excise tax
increase will cause sellers to raise their prices to consumers by at
least the full amount of the tax.
--------------------
\18 For example, in 1990 two producers supplied over 50 percent of
the domestic beer market.
EMPIRICAL ESTIMATES INDICATE
THAT PRICES AFFECT ALCOHOL
CONSUMPTION AND SOME OF ITS
HARMFUL CONSEQUENCES
---------------------------------------------------------- Letter :8.2
Numerous empirical studies confirm the conclusion of economic theory
that the higher prices that result from tax increases will have a
negative effect on alcohol consumption. However, precise estimates
of the degree to which consumption is affected are difficult to
obtain given the limitations of available data on alcohol prices and
consumption.\19
Researchers have found that light, moderate, and heavy drinkers in
the general population cut back on consumption when alcohol prices
are increased. However, study results vary concerning the relative
price sensitivity of light, moderate, and heavy drinkers. The
preponderance of the evidence on youth drinking indicates that youths
and young adults are more sensitive to price than older adults,
particularly those adults who have developed a long-term lifestyle
that includes heavy drinking. Researchers also found that beer is
the beverage of choice among youth who drink alcoholic beverages and
that youth seem to be more responsive to changes in alcohol prices
than the population in general.
Researchers also have found negative relationships between alcohol
prices or tax rates and the adverse consequences associated with
alcohol misuse, especially between alcohol use and auto crashes and
fatalities. For example, a recent study demonstrated that higher
state beer excise tax rates had a significant impact on lowering
total driver fatalities, night driver fatalities, and alcohol-related
fatalities for drivers of all ages and for drivers 18 to 20 years
old. Another study found that alcohol prices had a negative effect
on binge drinking--a 9-percent reduction in the number of binge
episodes per month resulted from a 10-percent increase in price.
Other research indicates that higher alcohol taxes (or prices) have a
negative and statistically significant effect on suicide rates;
possibly on the liver cirrhosis death rate; on mortality rates from
other cancers to which alcohol contributes; and on violent crimes,
such as rape and robbery. Some authors observe that the bulk of
evidence supports the conclusion that increasing alcohol taxes would
extend life expectancy.
--------------------
\19 See appendix III for a review of the specific empirical findings
of researchers who studied the effects of price increases on alcohol
consumption and a discussion of some of the data limitations that the
researchers have faced.
COMPETITION FROM MARYLAND
AND VIRGINIA MAY WEAKEN THE
EFFECTS OF A DISTRICT TAX
INCREASE
---------------------------------------------------------- Letter :8.3
If the District raises its alcohol taxes while Maryland and Virginia
do not, then some consumers who currently purchase their alcohol in
the District may shift the location of some of their purchases to
neighboring jurisdictions. The actual public health benefits of an
increase in alcohol taxes would be reduced to the extent that the tax
increase merely shifted the location of purchases rather than
reducing consumption.
Prior research indicates that the rates of local sales and/or excise
taxes across jurisdictions within a region can influence where the
consumers of that region shop. The importance of these so-called
"border effects" of tax rate differences depends on the specific
border situation in question, but several characteristics of the
District's metropolitan area imply that policymakers should not
ignore these effects when considering changes in the District's
alcohol taxes. First, most of the District's residents live within a
relatively short distance of alcoholic beverage retail outlets in
Maryland or Virginia. Second, every work day the District has a
large influx of commuters, who might shop in the District with
relative ease if they had a sufficient incentive. Finally, residents
in the immediate metropolitan area have a wide range of choices of
bars and restaurants in all three jurisdictions. If the District
were the only jurisdiction in the region that raised its taxes on
alcohol, then the after-tax prices of alcoholic beverages sold in the
District would increase relative to prices in the surrounding
jurisdictions, and retailers operating in the District could lose
some business to competitors in the surrounding jurisdictions.\20
In comparison to, say, a 10-cents-per-gallon increase in the federal
beer tax, a 10-cents-per-gallon increase in the District's beer tax
would have less of an effect on alcohol consumption (and associated
problems) of individuals who currently buy alcohol in the District.
Those individuals could avoid a District tax increase that was passed
on to consumers by making their purchases in Maryland or Virginia.
If the cost of shifting the location of their purchases is less than
the cost imposed by the District tax increase, then their cost of
consumption would not have risen by the full 10 cents per gallon.\21
In contrast, these individuals could not avoid a full increase in the
federal excise tax that was passed on to consumers by shifting their
purchases. Because the District tax would increase the cost of
consumption for some consumers by less than a similar federal tax
increase would, it would produce smaller aggregate behavioral
changes.\22 In addition to reducing the beneficial behavioral effects
of an alcohol tax increase, the shifting of sales would reduce the
potential revenue gains for the District. It would be difficult to
accurately estimate the size of the shift in sales that would occur
from any given increase in the District's alcohol taxes because of
the many factors involved.
--------------------
\20 Local laws that prohibit or limit the transportation of alcoholic
beverages from one jurisdiction to another may reduce cross-border
shopping, but to an unknown degree. For example, Virginia limits the
importation of alcoholic beverages, not for resale, to no more than 1
gallon or 4 liters in any one trip. Maryland limits personal
importation to 1 quart per person per trip and no more than 2 quarts
in any one calendar month.
\21 The cost of shifting locations of purchase will vary across
individuals. Individuals who commute between the District and the
surrounding jurisdictions every day are likely to have lower costs of
shifting their purchase locations than those who would have to make a
special trip into or out of the District just to purchase alcohol.
Individuals living close to the border of Maryland and the District
may experience no difference in transportation costs, whether they
shop in one jurisdiction or the other. The number of individuals who
would find it less costly to shift their purchases than to pay the
increased tax in the District would depend on the size of the tax
increase.
\22 A District tax increase could conceivably increase drunk driving
if it leads some District residents to drive farther away from home
to consume alcohol and then drive back. However, the increase could
have an opposite effect if it leads Maryland and Virginia residents
to drink locally, rather than drive into the District.
MANY STATES EARMARK PORTIONS OF
THEIR ALCOHOL TAX REVENUES FOR
SPECIFIC PURPOSES
------------------------------------------------------------ Letter :9
As of 1993, 24 states earmarked at least a portion of their alcohol
excise tax revenues for specific purposes. The percentage of the
alcohol tax revenue that was earmarked in each of these states ranged
from 4.5 percent in Colorado to 100 percent in West Virginia. The
purposes for which the alcohol revenue were earmarked also varied
substantially across these states--from public schools, to local
governments, to convention promotion. Appendix IV shows how much
revenue was earmarked for each purpose in each state in fiscal year
1993. In 15 of the states some of the revenue was specifically
earmarked for alcohol treatment, substance abuse, and/or mental
health programs. Since 1994, the District has earmarked 10 percent
of its sales tax on alcoholic beverages sold for on-premises
consumption to the Washington Convention Center Authority Fund.
STUDIES HAVE SHOWN THAT SOME
ALCOHOL PREVENTION APPROACHES
ARE EFFECTIVE, BUT MANY NEED
MORE RIGOROUS EVALUATION
----------------------------------------------------------- Letter :10
Communities have opted to use alcohol prevention approaches that fall
into three general categories. The first approach emphasizes
education and skill-building programs directed toward individuals in
schools, families, colleges, and specific population groups (e.g.,
women and minorities). The second approach is more population-based,
using legal and regulatory strategies to influence the physical and
social environments in which drinking occurs or is promoted. For
example, state and local governments seek to control the availability
of alcohol by regulating the location, hours of operation, and number
of establishments that sell alcoholic beverages. A third approach
combines these two by creating multiple communitywide strategies,
such as using education and skill-building programs to support new
laws and regulations. When programs directed toward individuals have
shown success, their effects have been small. Many of these programs
need to be evaluated more rigorously and over time to determine their
effectiveness. Research has shown that several laws and regulations
that control the physical availability of alcohol or the social
environment in which drinking occurs have resulted in lowered
consumption and fewer alcohol-related problems. However, research on
other preventive approaches influencing children's social
environment, such as controlling advertising, has been less
definitive. Studies have shown that combining the use of individual
and population-based legal and regulatory strategies can be
successful; additional studies of this approach are still under way.
APPROACHES THAT FOCUS ON
EDUCATION AND SKILL BUILDING
HAVE SHOWN MODEST SUCCESS
--------------------------------------------------------- Letter :10.1
Alcohol prevention approaches that are directed toward the individual
generally use education, information, and skill-building activities
to change attitudes and beliefs that influence drinking behavior and
enhance people's ability to resist underage and abusive drinking.
These approaches, although directed toward the individual, are most
often presented in group settings, such as schools, families, and
colleges and universities; or may focus on specific population
groups, such as minorities and women. Table 6 describes different
types of education and skill-building programs that are commonly used
to combat alcohol abuse and prevent drinking among youth.
Table 6
Education and Skill-Building Programs
Type of program Prevention strategy used
---------------------- ----------------------------------------------
School-based Life skills training focuses on enhancing an
individual's social interaction, interpersonal
conflict resolution, and assertiveness.
Resistance education strategies teach young
people to identify and resist specific
situations that could create pressure for them
to drink alcohol.
Normative education strategies try to correct
erroneous beliefs that young people have about
the prevalence and acceptability of alcohol
use among their peers and to promote
conservative attitudes about its use.
Parent and family Programs range from providing education about
alcohol use and abuse to strengthening the
family's ability to socialize the children in
a positive manner.
College or university Strategies focus on the early detection of
alcohol abuse, particularly binge or heavy
drinking, and intervention.
Specific population Strategies address the specific needs of
groups groups, such as minorities and women, and the
communities in which they live.
----------------------------------------------------------------------
Source: NIAAA.
Since the 1960s, school-based programs have played a key role in the
prevention efforts of many states and communities, primarily because
they give easy access to a young audience. Research shows that
although these programs are one of the most popular alcohol
prevention approaches and continue to target thousands of today's
young drinkers and potential drinkers, experts still debate their
effectiveness. The main goals of school-based programs are to
decrease the overall prevalence and level of drinking among youth;
reduce the progression of alcohol consumption to problem levels; and,
ideally, prevent young persons from starting to drink. One of their
major strategies is to influence knowledge, beliefs, or attitudes
about alcohol and its effects. Participants in one life skills
program reported lower alcohol use than nonparticipants after 5
years.\23 The prevention literature suggests, however, that the
success of most school-based programs in preventing the onset of
drinking and reducing the use of alcohol has been small.
Communities also use education and skill-building programs directed
to individuals in family units, colleges, and specific groups, such
as women and minorities. Research results have suggested that parent
participation in alcohol prevention programs could be effective in
reducing alcohol use, but such programs generally have a difficult
time getting large numbers of parents to participate on a regular
basis. Research has also shown that a prevention program for
freshman students at one university succeeded in reducing alcohol
consumption and problems associated with excessive drinking.\24
Researchers are exploring whether minorities, women, and other
special populations could benefit from prevention programs tailored
to their needs.
--------------------
\23 Gilbert J. Botvin et al., "Long-term Follow-up Results of a
Randomized Drug Abuse Prevention Trial in a White Middle-Class
Population," Journal of the American Medical Association, 273:14
(1995), pp. 1106-12.
\24 The High Risk Drinkers Project targeted freshman students at the
University of Washington and resulted in participants having levels
of both drinking and alcohol-related problems significantly lower
than those of a randomly assigned control group after 2 years.
NIAAA, "Prevention of Alcohol Problems," Chapter 9 of Alcohol and
Health: Ninth Special Report to the U.S. Congress, U.S. Department
of Health and Human Services, pp. 305-306, 1997.
APPROACHES THAT INFLUENCE
PHYSICAL AND SOCIAL
ENVIRONMENTS IN WHICH
DRINKING OCCURS
--------------------------------------------------------- Letter :10.2
The research literature suggests that certain legal and regulatory
strategies that influence the physical and social environments in
which alcohol is consumed are effective in reducing consumption and
alcohol-related problems. Prevention approaches that use these
strategies generally fall into two categories: (1) those intended to
influence individual drinking practices, such as enforcement of
impaired driving laws; and (2) those aimed at regulating the
availability of alcoholic beverages, such as restricting the number
and location of establishments selling alcohol. Prevention research
has produced evidence demonstrating the effectiveness of several
legal and regulatory strategies, such as laws prohibiting the sale of
alcohol to minors, server training programs, and various measures to
deter drinking and driving. Research has not been conclusive,
however, regarding the effectiveness of laws and regulations that,
for example, control the hours and days of alcohol sales, restrict or
ban alcohol advertisements, or require warning labels on alcoholic
beverages. Table 7 shows various types of legal and regulatory
strategies and their success in reducing alcohol consumption and
problems associated with excessive drinking.
Table 7
Selected Legal and Regulatory Prevention
Strategies
Research results on
Type of approach Prevention strategy used effectiveness of strategy
------------------------- ------------------------- --------------------------
Availability of alcohol Limiting the number of Research suggests that
alcohol outlets in a restricting the density of
given geographical area alcohol outlets may be
or community. effective in reducing
alcohol consumption and
motor vehicle crashes.
Limiting the hours and Extended hours of sale do
days alcohol can be sold. not necessarily increase
alcohol consumption or
related problems.
Restrictions on days when
sales are permitted appear
Establishing 21 as the to reduce traffic
minimum legal age for the crashes.
purchase and consumption
of alcohol. Raising the legal drinking
age to 21 effectively
reduced youth drinking and
related problems.
Server intervention, Training programs to Server training can modify
control laws, and legal enhance the skills of servers' and managers'
liability servers and managers to knowledge and beliefs
identify intoxicated and about alcohol service and
underage individuals and change serving practices.
discourage them from
drinking. Liability laws can affect
the behavior of persons
State and local laws and who serve alcohol,
regulations define who resulting in patrons
may serve or be served consuming less alcohol.
alcohol, the type of
alcohol sold, and the
time during which alcohol
can be sold.
Alcohol advertisements Restrictions or bans on The effects of advertising
advertising, restrictions or bans are
counteradvertising unclear.
efforts or warnings, and Limited evidence suggests
health promotion that counteradvertising
campaigns. and health promotion may
be effective.
Warning labels do not
Containers for alcoholic appear to influence
beverages must have a drinking behavior.
clear, nonconfusing label
warning of alcohol-
related hazards.\a
Drinking and driving Drinking and driving laws Evidence suggests that
that set blood alcohol drinking and driving laws
concentration levels for have significantly
young and adult drivers decreased alcohol-related
and use deterrence traffic crashes and
measures, such as problems.
interlock devices.\b
--------------------------------------------------------------------------------
\a In 1989, Public Law 100-690 required that containers for alcoholic
beverages be labeled to warn women about the risk of birth defects
from drinking during pregnancy and to warn the public about the
dangers associated with drinking and driving a car or operating
machinery.
\b Judges can order that an interlock device be installed in the cars
of people convicted of alcohol-impaired driving. This device
requires that the driver pass a dexterity or an alcohol-free test
before the vehicle will start.
Source: NIAAA.
VISIBLE ENFORCEMENT OF LEGAL
AND REGULATORY STRATEGIES
CAN ENHANCE THEIR
EFFECTIVENESS
--------------------------------------------------------- Letter :10.3
The literature suggests that visible enforcement programs and
education can enhance the beneficial effects of certain legal and
regulatory strategies. The deterrent effect of a law depends, at
least in part, on the public's belief that violations are likely to
be detected and violators punished. The District of Columbia and the
states of Maryland and Virginia use a number of enforcement
techniques to ensure compliance with state and local laws covering,
among other things, who may serve or be served alcohol, the type of
alcohol that can be sold, and the time during which alcohol can be
sold. Appendix V describes selected provisions of alcohol control
laws, along with related penalties, for the District, Maryland, and
Virginia.\25
Alcohol beverage control (ABC) boards establish conditions for
issuing licenses to sell alcohol and rely on a cadre of enforcement
officials\26 to monitor compliance with these regulations and impose
penalties for violations. Owners of licensed drinking establishments
and alcohol servers can be punished by fines and short jail terms if
they violate alcohol-related laws and regulations. The District and
neighboring jurisdictions in Maryland and Virginia devote different
levels of resources to enforcement. For example, the ratio of
enforcement officials to licensees in the District is about 1 to 400;
the ratio in the City of Alexandria is about 1 to 150.\27 In
Virginia, unlike Maryland and the District, ABC enforcement officials
have enhanced authority, which, among other things, allows them not
only to fine the establishment that sells alcohol to underage
purchasers, but also to fine the employee and the youth attempting to
purchase alcohol. ABC officials in the District and Maryland would
like similar enforcement authority granted to their staff, because
they believe that this authority is a highly effective deterrent to
underage drinking. Following is a description of several legal and
regulatory strategies whose effectiveness has been demonstrated by
research.
-- Studies of server training programs reveal that they can modify
servers' and managers' knowledge and beliefs about alcohol
service and bring about changes in serving practices that help
reduce the rate and amount of alcohol consumed by patrons. Such
training increased staff intervention with intoxicated patrons
and increased servers' willingness to suggest alternative
beverages and forms of transportation. Research also shows that
server intervention can be greatly enhanced through increased
enforcement of alcohol control laws and server liability laws.
-- Several researchers have explored the effectiveness of
increasing the visibility of enforcement and the rigorousness of
prosecution of alcohol control laws. They found declines in
both the number of arrests for driving under the influence of
alcohol obtained at bars and restaurants and incidents of
alcohol service to researchers posing as intoxicated patrons.
Research also shows that liability laws have affected the
behavior of persons who serve alcohol, which in turn affects the
drinking practices of patrons. Server liability laws place the
server at risk of committing a violation for serving alcohol to
underage drinkers or highly intoxicated patrons. Further, under
civil liability, or dram shop laws, an alcohol server has
potential legal responsibility for damage that intoxicated
patrons and underage drinkers inflict on themselves or others.
The financial loss that bar and restaurant servers and managers
may incur is expected to deter serving practices that could
increase a patron's risk of a motor vehicle accident and other
liabilities.
-- The minimum legal drinking age (MLDA) policy has been heavily
studied, with numerous research findings demonstrating the
effectiveness of a higher MLDA in preventing injuries and deaths
among youth. For example, the MLDA of 21 is estimated to save
more than 1,000 young lives each year. A recent review of 50
studies provided evidence that raising the legal drinking age to
21 reduced youth drinking and related problems, such as traffic
crashes. In response to many concerns about people under the
age of 21 easily obtaining alcohol, research has also suggested
that the MLDA could become even more effective with increased
enforcement, including deterrents for adults who might sell or
provide alcohol to minors.
-- Lower legal blood alcohol concentration (BAC) limits for youth
and adults have been found to decrease alcohol-related traffic
fatalities. Many states have lowered legally allowable BAC
limits for young drivers in an effort to reduce their
involvement in alcohol-related crashes; for these states, BACs
range from .00 to .05. An analysis of the first four states to
lower BAC levels found that these states experienced a decline
in teenage nighttime fatal crashes 30 percent greater than
declines in nearby comparison states that did not lower the BAC
limit.\28 A number of states have also lowered the legal BAC for
adults from .10 to .08 percent. Studies of a subset of these
states project that if all states adopted a .08 percent BAC law
for adults, at least 500 to 600 fewer deaths would occur
annually.\29
--------------------
\25 Some counties have enacted restrictions on various forms of
advertising of alcoholic beverages, such as bans on advertising
prices or restrictions on how and where alcoholic beverages may be
publicly displayed. In 44 Liquormart v. Rhode Island, 517 U.S. 484
(1996), however, the Supreme Court found that a Rhode Island law that
banned advertisement of retail liquor prices (except at the place of
sale) violated the First Amendment guarantee of freedom of speech.
Therefore, some current restrictions on advertising of alcoholic
beverages may present constitutional problems.
\26 In the District, these officials are called investigators; in
Virginia, enforcement agents; and in Montgomery and Prince George's
Counties, inspectors.
\27 The District's Fiscal Year 1998 Appropriations Act directs the
District to hire 12 new investigators, with half focusing on
prohibiting sales to minors. The District of Columbia Financial
Responsibility and Management Assistance Authority Board has not yet
authorized the D.C. ABC Board to hire the investigators and
continues to study how it can find the money for the new hires. In
addition, the D.C. budget has not been augmented to cover the 12
additional hires.
\28 Ralph Hingson, "Prevention of Alcohol-Impaired Driving," Alcohol
Health and Research World, Vol. 17, No. 1(1993), pp. 28-34.
\29 The District of Columbia lowered its BAC to .00 percent for
minors in 1994 and currently has a BAC of .10 percent for adults. A
provision that would reduce states' highway construction dollars if
the states fail to lower the legal blood alcohol limit for adults to
.08 percent by 2001 is currently under consideration by Congress as
part of the Intermodal Surface Transportation Efficiency Act of 1998.
EFFORTS COMBINING MULTIPLE
APPROACHES ARE GAINING
POPULARITY, BUT EVALUATIONS
ARE STILL UNDER WAY
--------------------------------------------------------- Letter :10.4
Lately, communitywide prevention efforts that combine multiple
strategies have become a more popular response to problems related to
alcohol misuse. These communitywide programs incorporate strategies
to both regulate the physical and social conditions in which drinking
occurs and educate individuals about alcohol use and enhance their
ability to reduce or resist drinking. Early research indicated that
using multiple approaches produced only temporary changes in drinking
behavior and the prevalence of alcohol-related problems. Programs
were redesigned in response to these findings, and preliminary data
suggest that these newer multifaceted strategies may be more
successful in reducing alcohol consumption than either individual or
environmental approaches alone. For example, in several states
implementation of a strong educational program, along with lowering
the legal BAC limit for teen drivers, was reported to significantly
reduce nighttime fatal automobile crashes. A 5 year community
prevention trial that combined several strategies, including
mobilizing the community, increasing enforcement of drinking and
driving laws, and enforcing underage sales laws, resulted in a
reduction in alcohol-involved traffic crashes of about 10 percent a
year and significant reductions in alcohol sales to minors.\30
Evaluations of several major studies of communitywide approaches are
still under way and no final outcome data are available. One of the
studies, Project Northland-Phase II, focuses on reducing drinking and
alcohol-related problems among 15- to 17-year-olds and includes a
combination of school and media curricula; youth social action
programs; parent involvement and education; and community task forces
for numerous policy and social interventions (e.g., enforcing
existing laws prohibiting alcohol sales to minors and restricting
alcohol sales at sporting, music, and other public events).\31
--------------------
\30 Addiction: A Community Prevention Trial to Reduce
Alcohol-Involved Trauma, Vol. 92, Supplement 2, June 1997.
\31 Kelli A. Komro et al., "Project Northland - Phase II: Research
and Evaluation Design of a Community-Wide Program to Reduce
Adolescent Alcohol Use." Proceedings published from a Kettil Bruun
Society Thematic Meeting, Fourth Symposium on Community Action
Research and the Prevention of Alcohol and Other Drug Problems,
February 8-13, 1998, New Zealand, forthcoming.
LIMITED EVIDENCE TO
DETERMINE WHAT WORKS
--------------------------------------------------------- Letter :10.5
Based on the current research literature, only a few alcohol
prevention approaches have been adequately evaluated and proven
effective. With the exception of several well-designed studies, such
as Project Northland and research on the MLDA, most published
evaluations of the effectiveness of alcohol prevention programs and
strategies were shown to be methodologically weak. Detailed reviews
of the alcohol prevention literature by NIAAA, the Institute of
Medicine, and other experts found limitations in the study designs
that affect the evaluation of outcomes and may compromise
conclusions. Common problems include questions about the validity of
self-reported data, the selection of inappropriate research designs
and statistical analyses, lack of comparable experimental and control
groups, and the potential impact of high attrition rates.
Evaluations of early school-based programs, for example, relied
heavily on self-reported data to measure alcohol use, which raises
concerns about possible underreporting or overreporting by program
participants. Although recent studies have attempted to address many
methodological challenges that commonly face researchers of
prevention programs, concerns continue to surface. For example,
although research has shown that some legal and regulatory approaches
are effective, the inability to control factors beyond the study
interventions makes it difficult to determine the exact nature of the
relationship between the prevention strategy and changes in drinking
behavior.
ENFORCEMENT STRATEGIES IN
DISTRICT OF COLUMBIA
METROPOLITAN AREA
--------------------------------------------------------- Letter :10.6
The District of Columbia, Maryland, and Virginia support a number of
enforcement strategies to enforce laws intended to prevent underage
drinking and the misuse of alcohol by adults. Some of these
strategies are directed at sales outlets, while others are aimed at
preventing underage access to alcohol and promoting responsible
drinking. ABC Boards, ABC enforcement officials, and local police
departments are responsible for implementing these strategies.
Although the following strategies have not been formally evaluated,
officials we interviewed cite them as successful.
-- A reverse sting operation, commonly referred to as cops in
shops, was an effort in which police officers, posing as store
clerks, apprehended minors using false identification to
purchase alcoholic beverages while ABC officials and/or police
officers waited outside in cars. Some of the officials we
interviewed ranked this program as the most successful
enforcement program in their jurisdictions, claiming numerous
citations issued to youth and subsequent decreases in underage
attempts to purchase alcohol. Officials said that establishment
owners welcomed this program, because only underage violators
were fined, not the establishment. Additionally, the
establishments did not have to pay the police officer, who
worked as a clerk for several hours in the liquor outlet. This
program, funded through a federal grant with the assistance of
the Washington Regional Alcohol Program of Northern Virginia,
was implemented for 1 year in the District and several Maryland
and Virginia counties.
-- Maryland requires an alcohol awareness training course for every
licensed establishment. Alcohol beverage control officials in
the District and Virginia support efforts to offer such
training, and the District requires training for establishment
owners who have violated liquor laws. Officials we interviewed
believe that statewide laws should require that a trained person
be on the premises of an establishment at all times, or else the
training has little effect on consumption. Although Maryland
law does not require this, Montgomery County supplemented
Maryland law to require a trained person on the premises at all
times.
-- Beer keg registration, a strategy in use in the District,
Maryland, and Virginia, requires that every keg sold must be
registered, with information on a label as to who bought it,
what kind of identification was shown, and where the keg came
from. This is to discourage adults from purchasing kegs for
youth; if the label is removed, legal responsibility rests with
the person hosting the party. Officials told us this has
resulted in a decline in adult beer keg purchases for underage
drinkers and has cut down on the number of youth drinking
parties.
-- Montgomery County, Maryland, and Virginia use sting operations
involving underage decoys who accompany ABC enforcement
officials to establishments to attempt to purchase alcohol. If
an establishment sells alcohol to a decoy, it can be penalized.
Montgomery County also uses these underage volunteers in a
program to monitor hotel and motel room service operations. In
this sting operation, the ABC official rents a room and the
underage volunteer calls for room service, ordering an alcoholic
beverage. The official waits unseen to observe if the youth is
illegally allowed to purchase alcohol. According to Montgomery
County officials, this program has been highly effective; hotels
had 100 percent compliance rates with underage drinking laws
following 2 consecutive years in which they had a 66 percent
violation rate. Prince George's County officials disagree with
the concept of a sting operation; they believe it
inappropriately entraps establishments.
THE DISTRICT OF COLUMBIA
FUNDS VARIOUS ALCOHOL ABUSE
PREVENTION AND TREATMENT
PROGRAMS
--------------------------------------------------------- Letter :10.7
In addition to its enforcement efforts, the District funds a variety
of health and education programs to prevent and treat alcoholism,
most of which are components of overall substance abuse programs. In
fiscal year 1997, District agencies spent about $66 million providing
substance abuse services to its residents.\32 Most of these dollars
were used for treatment services. The District Department of
Health's Addiction Prevention and Recovery Administration (APRA) is
one of the major providers of substance abuse treatment and
prevention services, with total spending of about $24 million in
fiscal year 1997. The District of Columbia Public Schools are a
source of funding for prevention activities for school-aged youth.
APRA funds alcohol prevention programs and offers counseling and
treatment services to residents, either directly or through
contractors. Alcohol prevention activities directed to youth who
have not begun to use alcohol range from disseminating information
and educating targeted populations, such as school-aged and college
youth, to helping community groups develop programs. During fiscal
year 1997, APRA spent about $1.2 million of its federal block grant
on alcohol prevention activities and treatment services, most of it
spent on treatment.\33 Major prevention activities included a
telephone hotline and neighborhood outreach centers. APRA worked
with other government agencies and community groups to provide
prevention activities to youth and adults.
The District also offers school children a systemwide drug prevention
education program, using funding under the Safe and Drug Free Schools
and Communities Act of 1994. Administered by the District of
Columbia Public Schools, the Substance Abuse Prevention Education
(SAPE) Program spent about $1.7 million during fiscal year 1997,
providing a variety of prevention activities to public, parochial,
and private school students; teachers and other school staff;
parents; and community groups. The SAPE Program provides education,
training, program development, and information dissemination to teach
its participants about the use and abuse of alcohol and other drugs.
--------------------
\32 Overall substance abuse programs include alcohol and other drug
prevention and treatment services. For fiscal year 1997, the
District's total expenditures for substance abuse services fell into
four categories: Medicaid (about $27 million), APRA services (about
$24 million), criminal justice programs (about $7 million), and other
agency programs, such as in the District's public schools (about $8
million).
\33 Under the Substance Abuse Prevention and Treatment Block Grant,
which is awarded to the District by the U.S. Department of Health
and Human Services, a minimum of 37 percent of the total amount of
the grant must be spent on alcohol treatment and prevention
initiatives.
CONCLUSIONS
----------------------------------------------------------- Letter :11
Although the District's alcohol tax structure differs from the tax
structures of most states, its combined taxes on alcohol are
generally higher. The District cannot conform its alcohol tax
structure to those in all surrounding jurisdictions at the same time,
because the tax structures among those neighboring jurisdictions
differ significantly. Moreover, the District would not be able to
impose exactly the same effective tax rates as those in either
Virginia or Montgomery County, because those effective rates are
difficult to estimate precisely. The District's taxes on beer are
currently the highest in the region. Increasing taxes on alcoholic
beverages has been associated with reductions in alcohol consumption
and related health and social problems. The special geographic
circumstances of the District--where all of its suburbs are in other
jurisdictions--could weaken the effect that an increase in the
District's taxes would have on local alcohol consumption.
Strategies to prevent youth from using alcohol and adults from
drinking excessively generally either try to educate individuals and
build their resistance skills or use legal and regulatory controls to
affect the availability and consumption of alcohol. Although
communities across the nation have invested significant resources in
these efforts, there is mixed evidence about which prevention
approaches are most effective. The best current evidence suggests,
however, that some legal and regulatory strategies, when enforced,
can help reduce illegal drinking and alcohol-related problems. If
District officials are interested in investing in new alcohol
prevention initiatives, it appears that greater efforts to enforce
existing laws and regulations might produce the best short-term
results. At the same time, however, more rigorous evaluations of
prevention strategies and programs would be needed in order to
provide better information about the effectiveness of the full range
of prevention approaches.
AGENCY COMMENTS AND OUR
EVALUATION
----------------------------------------------------------- Letter :12
We obtained written comments on a draft of this report from NIAAA,
PRC, and SAMHSA; and oral comments from the District of Columbia's
Office of Tax and Revenue, Department of Consumer and Regulatory
Affairs, and Department of Health; the District of Columbia Financial
Responsibility and Management Assistance Authority; and the Virginia
Department of Alcohol Beverage Control. NIAAA commented that in
general, the report makes a good start in identifying the complex
issues involved in designing programs and policies to reduce alcohol
abuse in a specific jurisdiction, particularly one with the unique
characteristics of the District. It also provided detailed comments
and suggestions for improving our presentation, which we incorporated
where appropriate. SAMHSA and the Department of Consumer and
Regulatory Affairs said they generally agreed with the findings of
the report. The other oral comments involved minor wording
clarifications, which we made where appropriate.
Officials from PRC suggested several technical changes to the report
that we incorporated where appropriate. In response to their comment
that our review of the literature did not sufficiently acknowledge
the success of prevention strategies that combine educational and
environmental interventions, we added the results of a major study
that combined several strategies. They also said that our review
gave inadequate recognition to some limitations of the economic
literature on the effects of alcohol taxes. Their conclusion is that
because of these limitations, no evidence exists regarding the
effects of local alcohol taxes. Our report makes clear that there is
much uncertainty regarding the size of the effect that an increase in
the District's alcohol taxes would have on consumption. However, we
believe that economic theory and the weight of the available
empirical evidence suggest that a District tax increase likely would
have some effect on alcohol consumption. We made changes to address
PRC's other concerns about our presentation of the alcohol prevention
literature when the concerns were supported by the evidence we
reviewed, including additional studies PRC provided.
--------------------------------------------------------- Letter :12.1
We are sending copies of this report to other appropriate
congressional committees and other interested parties. Copies will
also be made available to others upon request.
Major contributors to this report are listed in appendix VI. If you
have any questions, please contact Mr. White on (202) 512-9110 or
Ms. Lillie-Blanton on (202) 512-7119.
James R. White
Associate Director, Tax Policy
and Administration Issues
Marsha Lillie-Blanton
Associate Director, Health Services
Quality and Public Health Issues
ALCOHOLIC BEVERAGE PRICING BY THE
GOVERNMENTS OF VIRGINIA AND
MONTGOMERY COUNTY
=========================================================== Appendix I
STATE OF VIRGINIA
--------------------------------------------------------- Appendix I:1
The Virginia Department of Alcoholic Beverage Control operates state
stores that have the exclusive authority to sell liquor both to final
consumers for their own off-premises use; and to resellers, such as
hotels, restaurants, and taverns, that serve liquor on-premises. The
retail "shelf" prices in the state stores are uniform across the
state. Private sector resellers pay the same state shelf prices that
final consumers do. Final consumers pay the state sales tax on top
of the shelf price. Resellers add their own mark-ups on top of the
state's shelf price and then add the sales tax to the prices they
charge their customers.
The shelf prices are computed by formula. The department takes the
delivered case cost (what it pays to its supplier); adds a mark-up
(which is a percentage of the case cost, plus a $1 handling fee); and
then adds its 20 percent ad valorem excise tax to arrive at the shelf
price for the case. The department's mark-up percentages vary by
proof content and container size as follows:
Table I.1
Virginia's Mark-up Percentages for
Liquor
Mark-up percentages
-------------------
Equal to
Less or
than greater
125 than
Container size (in liters) proof 125 proof
------------------------------------------------- ------ -----------
1.750 45% 60%
1.000 50 60
.750 50 60
.375 55 67
.200 60 74
.050 30 74
----------------------------------------------------------------------
Source: State of Virginia, Department of Alcohol Beverage Control.
In the mid-1980s the department experimented by lowering prices on
popular liquor items in selected northern Virginia stores to be more
competitive with the District. The revenues for these stores
declined, because the lower prices did not attract enough business
away from the District to make up for the revenue that the stores
lost from customers who had already been patronizing the stores. The
last price survey that the department conducted in the early 1990s
showed that prices in the District were generally lower than in
Virginia for popular liquor brands in 1.75 liter size bottles. The
survey also showed that prices on liquor sold in bottle sizes of 750
ml or less were generally lower in Virginia. Representatives from
the District's Alcohol Retailers Association indicated that this
varied relationship between liquor prices in the District and
Virginia likely still existed.
MONTGOMERY COUNTY, MARYLAND
--------------------------------------------------------- Appendix I:2
Montgomery County's Department of Liquor Control operates 22 stores
that offer the full array of alcoholic beverage types. These are the
only stores in the county permitted to sell liquor for off-premises
use. Licensed private sector stores may sell beer and wine. Hotels,
restaurants, and clubs can serve all types of alcoholic beverages,
but only for on-premises consumption. Taverns are licensed to serve
beer and wine but not liquor. All of the licensed private sector
sellers of alcohol are required to purchase all of their alcoholic
beverages from dispensaries operated by the Department of Liquor
Control.
Montgomery County's Department of Liquor Control sets the wholesale
prices that its dispensaries charge to both private sector licensees
and to its own stores. The board also sets the retail shelf prices
that all of its own stores charge. The county's price mark-ups vary
by type of beverage. In the case of liquor, the county begins with
the delivered case cost that it pays to its suppliers, adds the state
excise tax, and then adds a percentage mark-up to arrive at its
wholesale price for the case. These wholesale mark-ups vary by
container size as follows:
Table I.2
Montgomery County's Mark-up Percentages
for Liquor
Mark-
up
percen
Container size (in liters) tages
-------------------------------------------------------------- ------
1.750 27.0%
1.000 27.0
.500 51.2
.375 51.2
.200 60.0
.110 60.0
.050 60.0
----------------------------------------------------------------------
Source: Distilled Spirits Council of the United States.
For its own stores, the county adds a further mark-up of 18 percent
to the wholesale case price to arrive at the retail shelf price.
Consumers pay the 5 percent state sale tax on this price. The same
procedure is used for wine, except that a different state excise tax
rate applies, and the county uses different wholesale and retail
mark-ups: 35 percent and 28 percent.
Beer wholesale and retail prices are not established by formal
mark-up rules. The county relies on price surveys of wholesale and
retail prices in nearby jurisdictions for guidance in setting beer
prices.
In the limited time available to us, we could not complete a retail
price survey of sufficient quality that would enable us to make a
useful comparison of retail liquor prices in Montgomery County with
retail prices in the private sector. However, because Maryland
requires all wholesalers operating in the state to (1) publish the
prices that they charge retailers for liquor and wine and (2) charge
the same price to all retailers in the state (excluding Montgomery
County), we were able to compare the wholesale prices that the county
charges for those beverages with the wholesale prices that would have
been charged in the county if the control system did not exist.\1
Our comparison of prices in effect during January 1998 indicates that
Montgomery County's wholesale prices for some alcoholic beverage
items are higher than those of private sector wholesalers operating
in the rest of Maryland, but for other items the county's prices are
lower.\2 The county's prices were higher than those of the private
sector wholesalers for 9 of the 14 liquor items that either the
Virginia ABC Commission or DISCUS identified as top sellers. The
county's wholesale prices also were higher for 15 out of the 28
liquor items and 15 out of the 29 wine items that we randomly
selected from the county's price lists.\3 In the absence of
information on sales volumes for each beverage item, we were not able
to determine whether Montgomery County's average prices for liquor
and wine are higher or lower than they would be if the county did not
control prices. Consequently, we were not able to determine whether
the average effective tax rate on alcohol in the county is above or
below the average statutory tax rate applicable in the county. The
fact that the relationship between the county's prices and the
private sector prices varies across beverage items means that the
relationship between effective and statutory tax rates on alcohol
sold in Montgomery County is likely to vary across beverage items
also.
--------------------
\1 Each alcoholic beverage item sold in Maryland is carried by only
one private sector wholesaler. Therefore, there was only one private
sector price that we needed to compare against the Montgomery County
price.
\2 Both Montgomery County and the private sector wholesalers offer
various discounted and special prices. We compared nondiscounted,
nonspecial prices only.
\3 For 17 of the 42 liquor items in our comparison, Montgomery
County's prices were more than 5 percent greater than those of the
private sector wholesalers, and for 8 liquor items the county's
prices were more than 5 percent lower than the private sector prices.
For 11 of the 29 wine items, Montgomery County's prices were more
than 5 percent greater than those of the private sector wholesalers,
and for 10 wine items the county's prices were more than 5 percent
lower than the private sector prices.
STATE ALCOHOLIC BEVERAGE EXCISE
TAX RATES AND STATE SALES TAX
RATES APPLIED TO ALCOHOL
========================================================== Appendix II
Excise tax rates ($ per
gallon)
-------------------------
State sales
tax on
alcohol
State Beer Liquor Wine (percent) Other tax rates
------- ------- ------- ------- -------------- ----------------------------
Alabama $0.53 \a $1.70 4.000 Wine: over 14% alcohol
content sold through state
stores
Alaska 0.35 $5.60 0.85 None Liquor: under 21% alcohol
content, $0.85 per gallon
Arizona 0.16 3.00 0.84 5.000
Arkansa 0.23 2.50 0.75 4.625 Beer: under 3.2% alcohol
s content, $0.16 per gallon;
over 3.2% and under 5%
alcohol content, $.23 per
gallon; $0.008 per gallon
enforcement tax; malt
liquor, $0.20 per gallon;
10% on-premises gross
receipts tax (for clubs).
Wine: under 5% alcohol
content, $0.25 per gallon;
over 5% alcohol content,
$0.75 per gallon; $0.05 per
case tax; 3% off-premises
and 10% on-premises gross
receipts tax (for clubs).
Liquor: under 5% alcohol
content, $0.50 per gallon;
over 5% but under 21%
alcohol content, $1.00 per
gallon; 3% off-premises and
14% on-premises gross
receipts tax.
Califor 0.20 3.30 0.20 6.000 Liquor: over 50% alcohol
nia content, $6.60 per gallon
Wine: sparkling wine, $0.30
per gallon
Colorad 0.08 2.28 0.32 3.000
o
Connect 0.19 4.50 0.60 6.000 Liquor: under 7% alcohol
icut content, $2.05 per gallon
Wine: over 21% alcohol
content and sparkling wine,
$1.50 per gallon
Delawar 0.16 3.75 0.97 None Liquor: under 25% alcohol
e content, $2.50 per gallon
Florida 0.48 6.50 2.25 6.000 Liquor: alcohol content
under 17.259%, $2.25 per
gallon; alcohol content
over 55.78%, $9.53 per
gallon
Liquor: retail tax $0.10
per ounce for on-premises
consumption
Wine: alcohol content over
17.259%, $3.00 per gallon;
sparkling wine, $3.50 per
gallon
Wine: retail tax $0.10 per
4 ounce on-premises
consumption
Beer: retail tax $0.04 per
12 ounce on-premises
consumption
Georgia 0.48 3.79 1.51 4.000 Liquor: $0.83 per gallon
local tax
Wine: alcohol content over
14%, $2.54 per gallon,
$0.83 per gallon local tax
Beer: $0.53 per gallon
local tax
Hawaii 0.92 5.92 1.36 4.000 Wine: sparkling wine, $2.09
per gallon and wine
coolers, $0.84 per gallon
Beer: $0.53 per gallon for
draft beer
Idaho 0.15 \a 0.45 5.000 Beer: alcohol content over
4%-$0.45 per gallon
Illinoi 0.07 2.00 0.23 6.250 Liquor: alcohol content
s under 14%, $0.23 per
gallon
Wine: alcohol content over
14%, $0.60 per gallon
Indiana 0.12 2.68 0.47 5.000 Liquor: alcohol content
under 15%, $0.47 per
gallon
Wine: alcohol content over
21%, $2.68 per gallon
Iowa 0.19 \a 1.75 5.000 Wine: alcohol content under
5%, $0.19 per gallon
Kansas 0.18 2.50 0.30 8.0/10.0\b Wine: alcohol content over
14%, $0.75 per gallon
Beer: alcohol content under
3.2%, a sales tax of 4.25%
Kentuck 0.08 1.92 0.50 6.000 Liquor: alcohol content
y under 6%, $0.25 per gallon;
$0.05 per case
Liquor, beer, and wine: 9%
wholesale tax
Louisia 0.32 2.50 0.11 4.000 Liquor: alcohol content
na under 6%, $0.32 per gallon
Wine: alcohol content from
14% to 24%, $0.23 per
gallon, over 24% and
sparkling wine, $1.59 per
gallon
Beer: $0.048 per gallon
local sales tax
Maine 0.35 \a 0.60 6.0/7.0\b Wine: alcohol content over
15.5% , sold through state
stores; sparkling wine,
$1.25 per gallon
Liquor: Sold only through
state stores if alcohol
content is over 4%
Beer: Sold through private
outlets
Marylan 0.09 1.50 0.40 5.000
d
Massach 0.11 4.05 0.55 5.000 Liquor: alcohol content
usetts under 15%, $1.10 per
gallon, over 50%, $4.05 per
proof gallon; 0.57% gross
receipts tax on private
club sales
Wine: sparkling wine, $0.70
per gallon
Beer: 0.57% gross receipts
tax on private club sales
Michiga 0.20 \a 0.51 6.000 Wine: alcohol content over
n 16%, $0.76 per gallon
Minneso 0.15 5.03 0.30 8.500 Liquor: $0.01 per bottle
ta (except miniatures)
Wine: alcohol content 14%
to 21%, $0.95 per gallon;
under 24% and sparkling
wine, $1.82 per gallon;
$0.01 per bottle (except
miniatures)
Beer: alcohol content under
3.2%, $0.077 per gallon
Mississ 0.43 \a 0.35 7.000
ippi
Missour 0.06 2.00 0.36 4.225
i
Montana 0.14 \a 1.06 None Wine: alcohol content over
16%, sold through state
stores; 7% surtax
Beer: 7% surtax
Nebrask 0.23 3.00 0.75 5.000 Wine: alcohol content over
a 14%, $1.35 per gallon
Nevada 0.09 2.05 0.40 6.500 Liquor: alcohol content
under 14%, $0.40 per gallon
and under 21%, $0.75 per
gallon
Wine: alcohol content 14%
to 22%, $0.75 per gallon;
over 22%-$2.05 per gallon
New 0.30 \a \a None Liquor and wine: all sales
Hampsh are through state stores
ire
New 0.12 4.40 0.70 6.000
Jersey
New 0.41 6.06 1.70 5.000 Wine: alcohol content over
Mexico 14%, $6.06 per gallon
New 0.16 6.44 0.19 4.000 Liquor: alcohol content
York under 24%, $2.54 per gallon
North 0.48 \a 0.79 4.000 Liquor: sales tax applies to
Caroli on-premises consumption
na only
Wine: alcohol content over
17%, $0.91 per gallon
North 0.16 2.50 0.50 7.000 Wine: alcohol content over
Dakota 17%, $0.60 per gallon;
sparkling wine, $1 per
gallon
Ohio 0.18 \a 0.32 5.000 Wine: alcohol content over
14%, $1 per gallon;
vermouth, $1.10 per gallon;
sparkling wine, $1.50 per
gallon
Oklahom 0.40 5.56 0.72 4.5/12.0\b Liquor: $1 per bottle on-
a premises tax
Wine: alcohol content over
14%, $1.44 per gallon;
sparkling wine $2.08 per
gallon; $1 per bottle on-
premises
Beer: alcohol content under
3.2%, $0.36 per gallon; $1
per case on-premises
Oregon 0.08 \a 0.67 None Wine: alcohol content over
14%, $0.77 per gallon
Pennsyl 0.08 \a \a 6.000 Liquor and wine: all sold
vania through state stores
Rhode 0.10 3.75 0.60 7.000 Wine: sparkling wine, $0.75
Island per gallon
South 0.77 2.72 0.90 5.000 Liquor: $5.36 per case
Caroli charge and 9% surtax on
na retail "shelf" price
Wine: $0.18 per gallon
additional tax
South 0.27 3.93 0.93 4.000 Liquor: alcohol content
Dakota under 14%, $0.93 per gallon
and 2% wholesale tax
Wine: alcohol content 14%
to 20%, $1.45 per gallon,
over 21% and sparkling
wine, $2.07 per gallon and
2% wholesale tax
Tenness 0.13 4.00 1.10 6.0/15.0\b,c Liquor and wine: $0.15 per
ee case charge; alcohol
content under 7%, $1.10 per
gallon
Beer: 17% wholesale tax
Texas 0.19 2.40 0.20 6.25/14.0\b Wine: alcohol content over
14%, $0.408 per gallon
Beer: alcohol content under
4%, $0.198 per gallon
Utah 0.35 \a \a 4.750 Liquor and wine: sold
through state stores only
Beer: alcohol content over
3.2%, sold only in state
stores
Vermont 0.27 \a 0.55 5.000 Wine: alcohol content over
16%, sold through state
stores
Beer: alcohol content 6% to
8% -$0.55; 10% on-premise
sales tax
Virgini 0.26 \a 1.51 4.500 Wine: alcohol content under
a 4%, $0.2565 per gallon and
over 14%, sold through
state stores
Washing 0.15 \a 0.87 6.500 Wine: alcohol content over
ton 14%, $1.72 per gallon
Beer: state excise tax plus
$4.78 per barrel additional
tax
West 0.18 \a 1.00 6.000 Wine: alcohol content over
Virgin 14%, sold through state
ia stores
Wiscons 0.06 3.25 0.25 5.000 Wine: alcohol content over
in 14%, $0.45 per gallon
Wyoming 0.02 \a \a 4.0 Liquor and wine: all sales
through state stores
Distric 0.09 1.50 0.30 8.0/10.0\b Wine: alcohol content over
t of 14%, $0.40 per gallon and
Columb sparkling wine, $0.45 per
ia gallon
--------------------------------------------------------------------------------
\a In 18 states, the government directly controls the sales of liquor
and, in some cases, beer and wine. Revenue in these states is
generated from various taxes, fees, and alcohol beverage receipts.
\b The first rate is on sales for off-premises consumption; the
second rate is on sales for on-premises consumption.
\c The on-premises sales tax rate does not apply to beer.
Source: Federation of Tax Administrators, "State Alcoholic Beverage
Excise Tax Rates as of January 1, 1998"; and "Tax Briefs" prepared by
the Distilled Spirits Council of the United States, December 1996.
EMPIRICAL EVIDENCE RELATING TO THE
EFFECTS OF ALCOHOL TAX INCREASES
ON ALCOHOL CONSUMPTION AND
ASSOCIATED BEHAVIOR
========================================================= Appendix III
Empirical research conducted since the early 1980s generally
concludes that increases in the prices of alcoholic beverages reduce
drinking; heavy drinking; and related outcomes, such as motor vehicle
and other accidents; liver cirrhosis mortality; "crime"; and reduced
education, employment, and labor productivity.\4 According to a
review of recent research, price-induced reductions in alcohol
consumption "are not limited to infrequent, light, or moderate
drinkers, but also occur among frequent and heavy drinkers." The
review also finds "youth and young adults, the age groups where
alcohol-related problems are disproportionately high, are generally
more responsive to increases in price than are adults."\5
--------------------
\4 Chaloupka et al., "The Effects of Price on the Consequences of
Alcohol Use and Abuse," in M. Galanter, ed. Recent Developments in
Alcoholism, vol. 14: The Consequences of Alcoholism: Medical,
Neuropsychiatric, Economic, Cross-Cultural, New York: Plenum
Publishing Corp. forthcoming; Chaloupka et al., "The Effects of
Price on Alcohol Consumption," Alcohol Health and Research World,
forthcoming.
\5 Chaloupka et al., Alcohol Health and Research World, forthcoming.
RESEARCHERS AGREE THAT ALCOHOL
TAXES AFFECT CONSUMPTION, BUT
EMPIRICAL ESTIMATES VARY
------------------------------------------------------- Appendix III:1
Higher prices for alcoholic beverages could be achieved by higher
taxation. There is a clear presumption that higher taxes on
alcoholic beverages are correlated with higher prices for those
beverages. In general, however, the link between alcohol taxes and
alcohol prices requires further study. Economists believe that the
extent to which any excise tax increase is passed along to consumers
varies depending on the characteristics of the markets in which
consumers purchase their beverages.\6 Such characteristics would
include how responsive market demand is to price changes and how much
competition among sellers exists in the market. Most researchers
studying the economics of alcohol consumption assume that the full
amount of the excise tax increase is passed along to consumers in the
form of higher prices. In the absence of more complete evidence,
researchers believe these are the best assumptions that can be made.
Many researchers have used the variation in state-level excise tax
rates across states as a proxy for the variation of alcohol prices
across states.
Researchers have estimated a range of values for the degree to which
consumption of beer, wine, or liquor responds to changes in the
prices of these beverages. A comprehensive survey of empirical
research conducted between 1983 and 1992 on the effect of price
increases on alcohol consumption found that in response to a 10
percent beer price increase, beer consumption would decline by
between 1.2 percent and 10.7 percent, with most studies estimating
that the change in consumption would be less than 5 percent.\7 \8
Generally, studies have tended to show that liquor and wine
consumption is somewhat more responsive to price changes than is beer
consumption. Experts estimated that liquor consumption would decline
by between 5 percent and 10 percent in response to a 10 percent
liquor price increase, but most of the estimates for wine were in the
range of 5 to 20 percent.\9
Other, generally more recent, studies have used data from surveys of
individual alcohol consumption. These studies have found higher
estimates for the consumption response to an increase in the price of
alcohol.\10 NIAAA's 1997 report to Congress reviewed and summarized
the post-1992 studies of the effect of alcohol price increases on
consumption.\11 According to this report, there continues to be
substantial variation in estimates of the responsiveness of alcohol
consumption to changes in alcohol prices.
One reason why the effects of price increases on alcohol consumption
remain uncertain is the quality of the data that researchers have to
work with. To make more precise estimates of the effects of price
increases on alcohol consumption, one would need to use accurate
measures of the prices that individual consumers pay for various
types of alcohol rather than consumption data aggregated to a state
or national level. However, collecting price data for a large sample
of consumers is difficult and costly.\12 There may also be problems
with the data on alcohol consumption that have been used in the
empirical literature--self-reported consumption data--which tend to
understate actual consumption.\13 The use of alternatives to
self-reported consumption data, such as expenditures on alcoholic
beverages, may introduce a different set of errors and biases.\14
--------------------
\6 Kenkel, Donald and Willard Manning, "Perspectives on Alcohol
Taxation," Alcohol Health and Research World, 20(4), 1996, pp.
233-234.
\7 Leung, Siu Fai and Charles Phelps, "My Kingdom for a Drink. . .
? A Review of Estimates of the Price Sensitivity of Demand for
Alcoholic Beverages," in Hilton, Michael and Gregory Bloss, eds.
Economics and the Prevention of Alcohol-Related Problems, National
Institute on Alcohol Abuse and Alcoholism Research Monograph No. 25,
NIH Pub. No. 93-3513, U.S. Department of Health and Human
Services, 1993.
\8 Nine of these studies estimated the sensitivity of beer
consumption to beer prices and got a statistically significant
result. All but one used consumption data aggregated to the state or
national level, rather than individual consumption data.
\9 National Institute on Alcohol Abuse and Alcoholism, "Economic
Aspects of Alcohol Use and Alcohol-Related Problems," Chapter 8 of
Alcohol and Health: Ninth Special Report to the U.S. Congress, U.S.
Dept. of Health and Human Services, June 1997, pp. 282, 284-285.
\10 Grossman, Michael, "The Economic Analysis of Addictive Behavior,"
in Hilton, Michael and Gregory Bloss, eds. Economics and the
Prevention of Alcohol-Related Problems, NIAAA Research Monograph No.
25, National Institutes of Health Pub. No. 93-3513, U.S.
Department of Health and Human Services, 1993, pp. 91, 114;
Chaloupka, Frank, "The Effects of Price on Alcohol Consumption,"
forthcoming, 1997, p. 5.
\11 NIAAA, "Economic Aspects of Alcohol Use and Alcohol-Related
Problems," Chapter 8 of Alcohol and Health: Ninth Special Report to
the U.S. Congress, U.S. Dept. of Health and Human Services, June
1997.
\12 In the absence of suitable data on alcohol prices, many
researchers have used the variation in state-level excise tax rates
as a proxy for the variation of alcohol prices across states. When
using tax variables as proxies for price variables in studies of
alcohol consumption, one needs to control for all state-specific
factors that might affect both alcohol tax rates and alcohol
consumption. For example, religious or cultural factors might cause
certain states to have both lower than average alcohol consumption
and higher than average tax rates on alcohol. The alternative,
computing alcohol price indices from expenditure and quantity data,
is also subject to problems. For example, see Johnson et al.,
"Alternative Approaches to the Measurement of Consumption and Price
of Alcoholic Beverages, Canada, 1957-1983," Journal of Studies on
Alcohol, 51(1), 1990, p. 82.
\13 Comparisons of representative survey data with aggregate alcohol
sales data indicate that self-reported alcohol consumption levels
tend to underestimate actual consumption. If the reporting errors in
drinking levels are correlated with drinking levels and other
relevant variables, statistical estimates based on these data may be
biased. Because the relationships between reporting errors and other
variables are not well understood, it is not clear how large the
resulting biases may be or how they might be mitigated.
\14 A consumer's high overall expenditure on alcohol may result from
larger levels of consumption of lower priced beverages or from
smaller levels of consumption of relatively higher priced beverages.
EVIDENCE INDICATES THAT BOTH
LIGHT AND HEAVY DRINKERS ARE
RESPONSIVE TO ALCOHOL PRICES,
ALTHOUGH SOME OF THE HEAVIEST
DRINKERS MAY BE LESS RESPONSIVE
------------------------------------------------------- Appendix III:2
Light, moderate, and fairly heavy drinkers respond to alcohol price
increases by cutting back on consumption. However, among a
relatively small number of the very heaviest drinkers--those often
considered to be addicted to alcohol--some researchers have found
very little, if any, response to changes in price, while others have
found some price responsiveness. One study found that consumers in
the middle of the distribution of drinkers were the most sensitive to
price changes, and very light and very heavy drinkers were less
sensitive. This study also found that the higher the price of
alcohol, the less likely consumers were to have any days of heavy
drinking.\15 Another study that examined the effects of alcohol
prices on the frequency of heavy drinking and drunk driving found
that a higher price of alcohol was associated with significant
reductions in the frequency of heavy drinking for males of all ages,
for females of all ages, and for females aged 21 and younger, but not
for males aged 21 and younger.\16
In another study, the same researcher has found that reported
familiarity with the health consequences of drinking was important in
determining the extent to which the heaviest drinkers responded to
price changes. The least-informed heavy drinkers did not appear to
be sensitive to price changes, but the best-informed heavy drinkers
appeared to be very sensitive. The author notes that the
heaviest-drinking, least-informed consumers might be alcoholics who
are in denial over the adverse consequences of drinking.\17 He and a
colleague also note that his finding is consistent with results of
the Manning et al. study that found that very light and very heavy
drinkers were less sensitive to price than others. The least
well-informed consumers in his study were, on average, also very
heavy drinkers.
--------------------
\15 Manning, Willard et al., "The Demand For Alcohol: The
Differential Response to Price," Journal of Health Economics, 14(2),
1995, pp. 137-139.
\16 Kenkel, Donald, "Drinking, Driving, and Deterrence: The
Effectiveness and Social Costs of Alternative Policies," Journal of
Law and Economics, XXXVI, October 1993, pp. 889-890.
\17 Kenkel, Donald, "New Estimates of the Optimal Tax on Alcohol,
Economic Inquiry, XXXIV, April 1996, p. 307.
YOUTH AND YOUNG ADULT ALCOHOL
CONSUMPTION MAY BE MORE
RESPONSIVE TO PRICE THAN THAT
OF OLDER DRINKERS
------------------------------------------------------- Appendix III:3
Most researchers have found that youth and young adults exhibit more
responsiveness to changes in alcohol prices than do older drinkers.
One explanation of the greater price sensitivity of younger drinkers
is that younger drinkers may have less income to spend than their
older counterparts. Whatever the reason for the greater price
sensitivity of younger drinkers, there may be public policy
implications. If older drinkers--those with a long-term lifestyle
that includes heavy drinking--are less sensitive to price while
younger drinkers are more sensitive to price, higher alcohol taxes
may have a two-fold effect. Higher alcohol prices may be an
effective policy for reducing youth alcohol consumption and its
related problems, as well as in reducing the likelihood of developing
a long-term lifestyle that includes heavy drinking.\18
Most researchers have found that beer is the beverage of choice among
youths who drink alcoholic beverages.\19 Some researchers have
concluded that beer is disproportionately preferred by higher risk
groups--for example, by those who drink a lot during a typical
session far more than by those who drink moderately. It also has
been noted that beer drinkers are more likely to drive while
intoxicated than drinkers of other alcoholic beverages.\20
Additionally (as noted above) researchers believe that the
responsiveness of alcohol consumption to changes in its price is
greater for youth than for adults.\21 One study estimated that a
10-percent increase in the price of beer could cause youths'
consumption to decline by 23 percent.\22 For those 17 to 29, another
study found that on average, a 10-percent increase in the price of
beer would lead to about a 7-percent decrease in consumption in the
long run.\23 Yet another study found that after the states increased
their legal drinking ages to 21 in the late 1980s, the price
sensitivity of youth alcohol use fell.\24
Recent studies show that the drinking behavior of youths who are
frequent, heavy, or binge drinkers is especially sensitive to alcohol
price changes.\25 One study found that a 10-percent decline in the
price of beer would increase the number of youths (aged 16 to 21) who
drink beer 4 to 7 times per week by about 10 percent. The same
10-percent decline in price would cause the number of youths (aged 16
to 21) who consumed no beer per week to fall by about 7 percent.\26
Another study by some of the same experts found that the number of
youths who drink six or more cans of beer on a typical drinking day
would decline by about 31 percent in response to a price increase of
10 percent; the number of youths who drink only 1 to 2 cans of beer
on a typical drinking day would decline by about 12 percent in
response to a price increase of 10 percent.\27 In contrast, another
recent study suggests that prices would have little impact on
drinking and binge drinking among male college students.\28 The
effects of alcohol taxation on heavy and binge drinking are of
special interest because of the high fatality rates from drunken
driving that are associated with it. Alcohol involvement in motor
vehicle accidents is estimated to be three times higher in the 18- to
20-year-old group than it is in the general population.\29 Results
from other studies indicate binge drinking and heavy drinking are
inversely related to price among adults as well.\30
--------------------
\18 Chaloupka, Frank and Henry Wechsler, "Binge Drinking in College:
The Impact of Price, Availability, and Alcohol Control Policies,"
Contemporary Economic Policy, 14(4), October 1996, pp. 112-124;
Kenkel, Donald, personal communication, 1998, p. 4.
\19 Coate, Douglas and Michael Grossman, "Effects of Alcoholic
Beverage Prices and Legal Drinking Ages on Youth Alcohol Use,"
Journal of Law and Economics, XXXI, April 1988, p. 152; Berger, Dale
and John Snortum, "Alcoholic Beverage Preferences of Drinking-Driving
Violators," Journal of Studies on Alcohol, 46(3), 1985, p. 232;
Grossman et al., 1998, p. 40; Saffer, Henry and Michael Grossman,
"Beer Taxes, the Legal Drinking Age, and Youth Motor Vehicle
Fatalities," The Journal of Legal Studies, XVI, June 1987, pp.
353-354; Chaloupka and Wechsler, 1996, p. 116.
\20 Berger and Snortum, pp. 232-239.
\21 Kenkel 1993, p. 895; Leung and Phelps, p. 23; Chaloupka, Frank,
personal communication, 1998, p. 1.
\22 Phelps, Charles, "Death and Taxes: An Opportunity for
Substitution," Journal of Health Economics, 7, 1988, p. 10.
\23 Grossman, Michael et al., "An Empirical Analysis of Alcohol
Addiction: Results from the Monitoring the Future Panels, Economic
Inquiry, XXXVI, January 1998, pp. 40, 45.
\24 Laixuthai, Adit and Frank Chaloupka, Youth Alcohol Use and Public
Policy, Cambridge, MA: National Bureau of Economic Research, Working
Paper No. 4278, February 1993, pp. 1, 14-16.
\25 Coate, Douglas and Michael Grossman, "Effects of Alcoholic
Beverage Prices and Legal Drinking Ages on Youth Alcohol Use,"
Journal of Law and Economics, XXXI, April 1988, pp. 145, 151, 164.;
Grossman, Michael, "The Economic Analysis of Addictive Behavior," in
Hilton, Michael and Gregory Bloss, eds. Economics and the Prevention
of Alcohol-Related Problems, NIAAA Research Monograph No. 25, NIH
Pub. No. 93-3513, U.S. Department of Health and Human Services,
1993, pp. 109, 111, 114-115; Leung and Phelps, pp. 19-20; NIAAA, p.
282; Phelps, pp. 9-10.
\26 Coate and Grossman, 1988, pp. 145, 151, 164.
\27 Leung and Phelps, pp. 18-20.
\28 Chaloupka and Wechsler, 1996, p. 120.
\29 Chaloupka et al., "Alcohol Control Policies and Motor-Vehicle
Fatalities," Journal of Legal Studies, XXII, January 1993, pp. 165,
181.
\30 Manning, Willard et al., p. 139; Kenkel, pp. 889-890.
EVIDENCE SUGGESTS THAT
INCREASING TAXES MAY REDUCE
SOME HARMFUL CONSEQUENCES OF
DRINKING
------------------------------------------------------- Appendix III:4
A number of studies have examined the relationship between alcohol
prices or tax rates and adverse consequences associated with alcohol
misuse. According to a summary of the most recent research, it has
been clearly demonstrated that increases in alcohol prices "can
significantly reduce many of the problems associated with alcohol
abuse, as well as improve educational attainment."\31 Problems
associated with alcohol use and abuse include drinking and driving
and motor vehicle accidents, liver cirrhosis and other health
effects, decreased educational attainment and employment, and
violence and other crime.\32
One of the most studied relationships is the relation between alcohol
use and auto accidents and fatalities. There is a consensus in the
empirical literature that an increase in the price of alcoholic
beverages would reduce the number of lives lost in vehicle
fatalities.\33 According to one study, the occurrence of drunk
driving declines as its full price increases.\34 The study also found
the risk of death or injury from an auto accident rises precipitously
with the intensity of drinking; i.e., binge drinking.\35 Another
study found that higher state beer excise tax rates were associated
with reductions in motor vehicle fatalities for youths aged 15
through 24.\36 Likewise, in another study the state beer excise tax
rate exhibited large negative and statistically significant
associations with total driver fatalities, night driver fatalities,
and alcohol-involved fatalities for both drivers of all ages and
drivers 18 to 20 years old.\37 Other researchers found higher state
beer tax rates to be weakly associated with a reduced propensity to
drive drunk.\38
A recent study of the relation between beer prices and drunken
driving included a relatively comprehensive set of explanatory
variables and examined a variety of different model specifications.
This study found that a 10-percent increase in the price of beer
would result in an almost 10-percent decrease in the fatality rate
from drunken driving, a 14-percent decrease in the fatality rate from
nighttime drunken driving, and a 14-percent decrease in the fatality
rate from drunken driving for those aged 18 to 20.\39 Another recent
study estimated that alcohol prices have a negative and significant
effect on binge drinking--the behavior that leads to drunken
driving--with a 10-percent increase in the price of alcohol leading
to a 9-percent decrease in the expected number of binge episodes per
month.\40
Two studies have found that the excise tax rate on liquor has a
negative and significant effect on the liver cirrhosis death rate.\41
In contrast, other researchers found higher alcohol prices were not
significantly related to lower death rates from liver cirrhosis.
These studies did find a significant negative relationship between
alcohol prices and suicide rates and mortality rates from other
cancers to which alcohol contributes. They also found weak or
insignificant effects of alcohol price on death rates from homicide
and from falls, fires, and other accidents.\42
Two recent studies investigated the relationship between alcohol use
and crime.\43 These researchers found significant relations between
the real tax rate on beer and the incidence of rape and robbery.
Other recent studies examined the impact of alcohol use and heavy use
on the level of education attained.\44 These researchers note that
there is evidence that heavy drinking is associated with reductions
in the average number of years of schooling completed and reduction
in employment as well as a tendency toward alcohol abuse in later
life. They observe that the bulk of evidence supports the conclusion
that increasing alcohol taxes would extend life expectancy.\45
Several other experts have suggested that people who misuse alcohol
are less likely to be employed and tend to have lower incomes than
people who do not.\46
--------------------
\31 Chaloupka, Frank et al., in Recent Developments in Alcoholism,
forthcoming, pp. 21-22.
\32 Chaloupka, Frank et al., in Recent Developments in Alcoholism,
forthcoming, pp. 9-10, 15-21.
\33 Phelps, 1988; Kenkel, 1993.
\34 Mullahy, John and Jody Sindelar, "Do Drinkers Know When to Say
When? An Empirical Analysis of Drunk Driving," Economic Inquiry,
XXXII, July 1994, pp. 383-394. The full price of drunk driving
includes not only the cost of purchasing alcoholic beverages, but
also the costs of driving while intoxicated. The latter costs can
include license revocation, fines, imprisonment, and social stigma.
\35 The study found that youths who have consumed 6 or more drinks
and then drive face a 100-fold increase in risk of death, compared
with nondrinkers.
\36 Saffer, Henry and Michael Grossman, "Beer Taxes, the Legal
Drinking Age, and Youth Motor Vehicle Fatalities," Journal of Legal
Studies, June 1987, pp. 351-353, 373-374.
\37 Chaloupka et al., 1993. Drinkers who prefer beer are more likely
to drive while intoxicated than those who prefer wine or liquor.
\38 Mullahy and Sindelar, pp. 383-394.
\39 Ruhm, C.J., "Alcohol Policies and Highway Vehicle Fatalities,"
Journal of Health Economics, 15(4), 1996, pp. 435-454.
\40 Sloan, F. et al., "Effects of Tort Liability and Insurance on
Heavy Drinking and Drinking and Driving," Journal of Law and
Economics, 38(1), 1995, pp. 49-78.
\41 Chaloupka et al., "Alcohol Addiction: An Econometric Analysis,"
presented at the annual meeting of the Allied Social Science
Associations, Anaheim, CA, December 1992; Cook, Philip and George
Tauchen, "The Effect of Liquor Taxes on Heavy Drinking," Bell Journal
of Economics, 12, 1982, p. 379.
\42 Sloan, F. et al., "Effects of Prices, Civil and Criminal
Sanctions, and Law Enforcement on Alcohol-Related Mortality," Journal
of Studies on Alcohol, 55, 1994, pp. 454-465.
\43 Chaloupka, Frank and Henry Saffer, "Alcohol, Illegal Drugs,
Public Policy and Crime," presented at the annual meeting of the
Western Economic Association, San Francisco, CA, July 1992; Cook,
Philip and Michael Moore, "Economic Perspectives on Reducing
Alcohol-Related Violence," in Martin, S. E. ed., Alcohol and
Interpersonal Violence: Fostering Multidisciplinary Perspectives,
Washington, DC:
U. S. GPO, 1993, p. 193.
\44 Cook, Philip and Michael Moore, "Drinking and Schooling," Journal
of Health Economics, 12, 1993, p. 411; Yamada, T. et al., "The
Impact of Alcohol Consumption and Marijuana Use on High School
Graduation," Health Economics, 5, 1996, p. 77.
\45 Cook, Philip and Michael Moore, "Taxation of Alcoholic
Beverages," in Hilton, Michael and Gregory Bloss, eds., Economics and
the Prevention of Alcohol-Related Problems, NIAAA Research Monograph
No. 25, NIH Pub. No. 93-3513, U. S. Department of Health and
Human Services, 1993, pp. 33, 50.
\46 Kenkel, Don and D. Ribar, "Alcohol Consumption and Young Adults'
Socioeconomic Status," Brookings Papers on Economic Activity, 1,
1994, p. 119; Mullahy, J. and Jody Sindelar, "Alcoholism, Work, and
Income," Journal of Labor Economics, 11(3), 1993, p. 494; Harwood,
H. J. et al., "Economic Costs to Society of Alcohol, Drug Abuse and
Mental Illness," Research Triangle Institute, 1984.
HOW STATES EARMARKED ALCOHOLIC
BEVERAGE EXCISE TAX REVENUES, AS
OF FISCAL YEAR 1993
========================================================== Appendix IV
(Dollars in millions)
Earmarked
----------------------------------------------------
Total
collectio
State ns Amount Percent Purpose
--------------- --------- --------------- --------------- ------------------
Alabama $122.4 $9.4 7.7 Mental health
30.8 25.2 Human resources
17.8 14.5 Education
6.1 4.9 Counties and
cities
Arizona 40.9 15.2 37.2 Correction fund
Arkansas 4.2 1.1 26.2 University of
Arkansas Medical
Center
Colorado 23.2 1.0 4.5 Old-age pension
fund
Florida 539.2 8.8 1.6 Child and
adolescent
substance abuse
services
19.9 3.7 Alcoholic beverage
and tobacco trust
fund
Idaho 12.8 4.5 35.2 Counties and
cities
1.2 9.4 Alcohol programs
1.2 9.4 K-12 education
0.3 2.3 Community colleges
0.7 5.5 Welfare
Indiana 33.2 15.6 47.0 Prison
construction,
enforcement, and
administration
2.3 6.9 Local police
pension
2.8 8.4 Addiction services
8.3 25.0 Cities and towns
Kansas 57.7 12.2 21.2 County or city
where sold
1.5 2.6 Alcoholism
treatment and
prevention
Michigan 74.5 21.9 29.4 School aid fund
21.9 29.4 Convention
promotion
8.0 10.7 Liquor purchasing
revolving fund
Mississippi 38.1 3.1 8.1 Department of
Mental Health
1.0 2.6 Municipalities
Montana 15.4 3.7 24.0 Alcohol treatment
and
rehabilitation
2.6 16.9 Local government
Nevada 14.6 2.0 13.7 Counties and
cities
0.6 4.1 Alcohol and drug
abuse programs
New Jersey 88.0 11.0 12.5 Alcohol education,
rehabilitation,
and enforcement
New Mexico 17.6 9.1 52.0 Community
alcoholism and
detoxification
fund
North Carolina 156.1 2.1 1.3 County
rehabilitation
contribution
3.4 2.2 Law enforcement
2.9 1.9 Alcohol education
27.1 17.4 County and city
districts
21.0 13.5 Local-state tax
sharing
Ohio 83.5 0.3 0.4 State grape
industries
2.0 2.4 Alcohol treatment
and prevention
18.1 21.7 Debt service on
state economic
development bonds
Oklahoma 23.6 6.8 29.0 Cities and towns
Oregon 71.4 29.0 40.6 Counties and
cities
5.2 7.3 Alcoholism
programs
0.1 0.2 Wine industry
South Dakota 21.0 5.3 25.2 Municipalities and
local governments
Tennessee 64.2 1.7 2.6 Highway fund
18.8 29.3 Counties and
cities
11.2 17.4 Education
Texas 392.5 24.9 6.3 Law enforcement
29.8 7.6 Public education
(K-12)
51.2 13.0 Counties and
cities
Virginia 26.8 17.9 66.8 Local government
Washington 154.4 10.0 6.5 Drug enforcement
and education
20.7 13.4 Counties and
cities
0.1 0.1 Washington Wine
Commission
0.1 0.1 Wine grape
research
West Virginia 4.4 4.4 100.0 Counties and
municipalities'
state police
drunk driving
prevention funds
--------------------------------------------------------------------------------
Source: National Conference of State Legislatures, Earmarking State
Taxes (April 1995).
SELECTED PROVISIONS OF ALCOHOL
CONTROL LAWS IN THE DISTRICT OF
COLUMBIA, MARYLAND, AND VIRGINIA
=========================================================== Appendix V
(See figure in printed
edition.)
(See figure in printed
edition.)
(See figure in printed
edition.)
Note: Maryland and Virginia counties may enact local laws that
supplement state laws affecting penalties, hours of operations, or
other requirements.
Source: The statutes of the District of Columbia, Maryland, and
Virginia.
MAJOR CONTRIBUTORS TO THIS REPORT
========================================================== Appendix VI
GENERAL GOVERNMENT DIVISION,
WASHINGTON, D.C.
James A. Wozny, Assistant Director, Tax Policy and Administration
Issues
Charles C. Tuck, Economist-in-Charge
Anne Stevens, Senior Economist
HEALTH, EDUCATION AND HUMAN
SERVICES DIVISION, WASHINGTON,
D.C.
Helene Toiv, Assistant Director, Health Services Quality and Public
Health Issues
Ann M. Calvaresi Barr, Project Manager
Brenda R. James, Evaluator-in-Charge
OFFICE OF GENERAL COUNSEL
Roger J. Thomas, Senior Attorney
Shirley A. Jones, Senior Attorney
Richard T. Cambosos, Senior Attorney
RELATED GAO PRODUCTS
============================================================ Chapter 0
Substance Abuse and Mental Health: Reauthorization Issues Facing the
Substance Abuse and Mental Health Services Administration
(GAO/T-HEHS-97-135, May 22, 1997).
Drug Control: Reauthorization of the Office of National Drug Control
Policy (GAO/T-GGD-97-97, May 1, 1997).
Drug Control: Observations on Elements of the Federal Drug Control
Strategy (GAO/GGD-97-42, Mar. 14, 1997).
Substance Abuse Treatment: VA Programs Serve Psychologically and
Economically Disadvantaged Veterans (GAO/HEHS-97-6, Nov. 5, 1996).
Drug and Alcohol Abuse: Billions Spent Annually for Treatment and
Prevention Activities (GAO/HEHS-97-12, Oct. 8, 1996).
Substance Abuse Surveys (GAO/HEHS-96-179R, July 19, 1996).
Cocaine Treatment: Early Results From Various Approaches
(GAO/HEHS-96-80, June 7, 1996).
At-Risk and Delinquent Youth: Multiple Federal Programs Raise
Efficiency Questions (GAO/HEHS-96-34, Mar. 6, 1996).
Treatment of Hardcore Cocaine Users (GAO/HEHS-95-179R, July 31,
1995).
Residential Care: Some High-Risk Youth Benefit, But More Study
Needed (GAO/HEHS-94-56, Jan. 28, 1994).
Drug Use Among Youth: No Simple Answers to Guide Prevention
(GAO/HRD-94-24, Dec. 29, 1993).
Confronting the Drug Problem: Debate Persists on Enforcement and
Alternative Approaches (GAO/GGD-93-82, July 1, 1993).
Indian Health Service: Basic Services Mostly Available; Substance
Abuse Problems Need Attention (GAO/HRD-93-48, Apr. 9, 1993).
Drug Education: Limited Progress in Program Evaluation
(GAO/T-PEMD-93-2, Mar. 31, 1993).
Community-Based Drug Prevention: Comprehensive Evaluations of
Efforts are Needed (GAO/GGD-93-75, Mar. 24, 1993).
*** End of document. ***