Canadian Centre for Policy Alternatives Sobering...

By Greg Flanagan

The Alberta Liquor

Retailing Industry

Ten Years after

Privatization

S o b e r i n gR e s u l t :

Parkland Institute is an Alberta research network that examines public policyissues. We are based in the Faculty of Arts at the University of Alberta andour research network includes members from most of Alberta’s Academicinstitutions as well as other organizations involved in public policy research.

Parkland Institute was founded in 1996 and its mandate is to:

• conduct research on economic, social, cultural and political issues facingAlbertans and Canadians.

• publish research and provide informed comment on current policy issuesto the media and the public.

• sponsor conferences and public forums on issues facing Albertans.• bring together academic and non-academic communities.

The Canadian Centre for Policy Alternatives is one of Canada's leadingprogressive policy research organizations. Founded in 1980, the CCPA issupported by its 8,000 individual and organizational members.

The Centre, through its work, seeks to promote economic and social literacyamong Canadians on important issues affecting their lives.

The CCPA publishes reports, books and other publications, including amonthly magazine.

Canadian Centre for Policy Alternatives

Parkland Institute

Sobering Result:The Alberta Liquor Retailing Industry

Ten Years after Privatization

Greg Flanagan

June 2003ISBN 0-88627-338-2

Published by:Canadian Centre for Policy Alternatives

and Parkland Institute

Table of contents

About the author ...................................................................................................................................... i

Acknowledgements .................................................................................................................................. i

Abstract ...................................................................................................................................................... ii

Executive Summary ................................................................................................................................ iiiConclusions: .......................................................................................................................................................................... v

1 Introduction ...........................................................................................................................................11. A Brief Description of Alcoholic Beverage Distribution in Alberta ...................................................................... 11.2 Purpose and Methodology ......................................................................................................................................... 21.3 Outline ............................................................................................................................................................................. 3

2 The Importance of Alcoholic Beverages ......................................................................................... 52.1 Alcohol Consumption Concerns ................................................................................................................................ 52.2 International Comparisons ......................................................................................................................................... 52.3 Canadian, Provincial, and Municipal Comparisons ........................................................................................ 62.4 The Rationale for Public Regulation ....................................................................................................................... 10

3 Social Issues ........................................................................................................................................133.1 Crime ............................................................................................................................................................................. 133.2 Impaired Driving ......................................................................................................................................................... 163.3 Socially Responsible Marketing ............................................................................................................................... 18

4 Alcohol Taxes .......................................................................................................................................214.1 Why Tax Alcoholic Beverages? ................................................................................................................................. 214.2 Unit Tax versus Ad Valorem Tax ...............................................................................................................................224.3 Federal Tax System .....................................................................................................................................................224.4 Provincial Tax Revenues ............................................................................................................................................23

5 Liquor Prices ....................................................................................................................................... 285.1 Alcohol Price Elasticity ................................................................................................................................................285.2 External Social Costs ..................................................................................................................................................285.3 Optimal Pricing with External Social Costs ............................................................................................................295.4 Price History .................................................................................................................................................................305.5 Current Prices ..............................................................................................................................................................32

6 The Evolving Market for Alcohol .................................................................................................... 366.1 Physical Availability .....................................................................................................................................................366.2 Market Structure or Industrial Organization .........................................................................................................376.3 The Alberta Liquor Retail Market Structure ...........................................................................................................386.4 Product Differentiation ..............................................................................................................................................396.5 Excess Capacity—Too Many Retailers ..................................................................................................................... 416.6 The Growth of Chain Stores ..................................................................................................................................... 416.7 Wholesale Costs ..........................................................................................................................................................42

7 Conclusions ........................................................................................................................................ 45

Endnotes ..................................................................................................................................................47

Bibliography ........................................................................................................................................... 50

List of tables

Table 2.1 Per capita consumption of pure alcohol equivalent (litres) in 1996 ...................................................... 6

Table 2.2 Trends in per capita consumption of pure alcohol equivalent (litres) in 1996 .................................. 6

Table 2.3 Percentages of alcohol beverage expenditures in 1996 ......................................................................... 11

Table 3.1 Edmonton gaming and liquor act incidents, Edmonton Police Service, rates per 100.000 persons 15yrs or older ........................................................................................................................ 16

Table 4.1 Alberta excise tax rates ....................................................................................................................................24

Table 4.2 Alberta excise tax rates converted to percentage change ......................................................................25

Table 4.3 Liquor revenue if 1992 revenue per capita were maintained ..............................................................27

Table 5.1 Summary statistics on price of ten alcohol beverage products surveyed January 2003 ................................................................................................................................................................34

Table 6.1 Total government and agency stores (Alberta private) ...........................................................................36

Table 6.2 The number and percent of chain stores ...................................................................................................42

Table 6.3 Sales, costs, and profit data Alberta and British Columbia (1000s) .....................................................43

Table 6.4 Landed costs of a few popular products in four provinces .................................................................. 44

List of figures

Figure 2.1 Volume of Sales of Alcoholic Beverages in Litres of Absolute Alcohol Per Capita 15 Years andOver ................................................................................................................................................................................. 7

Figure 2.2 Average Expenditure per Household on Alcohol in 2000 ..................................................................... 8

Figure 2.3 Average Expenditure per Household on Alcoholic Beverages in 2000 Selected MetropolitanAreas ................................................................................................................................................................................ 8

Figure 2.4 Average Expenditure on Alcoholic Beverages per Household by IncomeQuintiles Canada (10 Provinces) 2000 .......................................................................................................................... 9

Figure 2.5 Alcohol Expenditures as a Percentage of Total Expenditures (Income) ............................................. 9

Figure 3.1 Crime Rates: All Offences, Total ................................................................................................................. 14

Figure 3.2 Crime Rates: Robbery ................................................................................................................................... 15

Figure 3.3 Crime Rates: Break and Enter. .................................................................................................................... 15

Figure 3.4 Rate of Persons Charged with Impaired Driving by Province, 1998 .................................................. 17

Figure 3.5 Rate of Persons Charged with Impaired Driving by Nine largest Metropolitan Areas, 1998 .................. 17

Figure 3.6 Alberta Impaired Driving Charges .............................................................................................................. 18

Figure 4.1 Liquor Revenue as a Percentage of Total Provincial Government Revenue ................................... 26

Figure 4.2 Alberta and BC Provincial Revenue Comparison ................................................................................. 26

Figure 4.3 Tax Revenue Absolute Alcohol Sales ....................................................................................................... 27

Figure 5.1 The Effect of an External Cost on the Private Market for a Product .................................................. 30

Figure 5.2 Alcohol Beverage Inflation Rates ............................................................................................................... 31

Figure 5.3 Alberta Inflation Rates 'Alcoholic Beverages' and 'All Other Products ............................................. 31

Figure 5.4 The Difference in CPI between Alcohol Beverages and All Other Items ....................................... 32

Figure 5.5 CPI, Alcohol Beverages Sold in Stores Comparison of Alberta and British Columbia(1992=100) ................................................................................................................................................................ 33

Figure 6.1 Liquor Outlets (Ontario and British Columbia adjusted for private outlets) .................................... 37

Sobering Result i

About the author

Greg Flanagan is a public finance economist based in Calgary. He has taught for over 25 years inAlberta at various colleges, including 17 years at Mount Royal College. He served as dean at St. Mary’sCollege, 1998-2002. He holds degrees from University of Calgary (BA Economics), York University(MES Political Economy), and the University of British Columbia (MA Economics). His researchinterests focus on the economics of public policy. He is a founding director of Parkland Institute,Faculty of Arts, University of Alberta, and co-author of two textbooks: Economics in a CanadianSetting, HarperCollins Publishers, 1993; and Economics Issues, a Canadian Perspective, McGraw-Hill,1997; as well as numerous papers and articles. Most recently, he taught public finance (economics) atthe University of Calgary.

Acknowledgements

I would like to thank the reviewers for their detailed suggestions and comments: Ricardo Acuna,Parkland Institute; Soren Bech, BC Government Employees Union; Bruce Campbell, Canadian Cen-tre for Policy Alternatives; Mark Dickerson, University of Calgary; Roger Epp, Augustana University;Trevor Harrison, University of Lethbridge; Dean Neu, University of Calgary; and Steve Patten, Uni-versity of Alberta. I thank Barry Pashak for his help in obtaining price data in the short window oftime to effect comparable data, and also for commenting on the paper; Barry Bieller, Charlie Ruddick,and Gord Hall, Government of British Columbia, for their time discussing with me the BC system ofdistribution and regulation; and also Robin Rutherford and other members of the BCGEU whodiscussed the operation of the BC distribution and retail system. I also thank Dawn Macdonald,Edmonton Police Service for providing data for Table 3.2; John Szumlas, Alberta Association of Liq-uor Retailers; and all of the anonymous individuals working in the Alberta retail liquor trade fortaking the time to inform me of the retail industry workings, and their concerns and issues. Andthanks to Kerri-Anne Finn for the final proofing and layout of the study. Appreciation is also ex-tended to The National Union of Public and General Employees (NUPGE) for contributing to thefunding of this study; and the Canadian Centre for Policy Alternatives (CCPA) and Parkland Insti-tute for commissioning the study. I am, of course, entirely responsible for any errors in the analysisand for the final views expressed.

Greg Flanagan

ii Canadian Centre for Policy Alternatives / Parkland Institute

Abstract

In 1993/94 the Alberta Government implemented major policy changes involving the control, tax-ing, and distribution of liquor products. These changes included privatization of the retail and ware-housing functions, switching from an ad valorem (percentage of price) to a unit (flat) tax system ofalcohol excise taxes, and the ending of direct control of liquor regulation and moving to a legislativeand enforcement approach. Ten years later the retail industry has evolved into monopolistic competi-tion with its inherent excess capacity and high costs. The government has lost effective control of theliquor industry which will likely continue to evolve into an oligopolistic market structure as chainstores get greater control. Against the trends in other jurisdictions, liquor consumption has increased(with its potential risks of increasing social ills), wholesale costs have risen, and retail prices haveincreased. Although retail prices have increased, the tax revenues to government have fallen signifi-cantly.

Sobering Result iii

Executive Summary

was 1.72% of the average consumer’s budget. Thiswas 7.5% lower than in 1996 at 1.86%. Two-thirds of alcoholic beverages are sold in retail out-lets, and approximately one-half of the sales areon beer products (Table 2.3).

Comparing provinces, Alberta has the high-est per capita consumption of alcohol when meas-ured by dollar value of expenditures or absolutealcohol consumed (Figure 2.2). Calgary rates thehighest for major metropolitan areas, while Ed-monton is a modest eighth (Figure 2.3). The con-sumption of alcoholic beverages is positively cor-related with income (Figure 2.4). When calcu-lated as a percentage of income, Alberta’s expen-ditures on alcohol are closer to the national aver-age and that of its neighbours (Figure 2.5).

Social issues relating to alcohol consumptioninclude ill- health effects, fetal alcohol syndrome,family violence and divorce, lost work and pro-ductivity, and crime including impaired driving.Against the general backdrop of falling crime ratesin Canada (Figures 3.1, 3.2, & 3.3), there is localevidence that crime is associated with alcoholavailability and price (Table 3.1). Greater ethno-graphic study needs to be done in order to deter-mine if liquor privatization, and the resultingchanges in marketing, are associated with crimerates. Alberta has the second highest impaireddriving charge rates in Canada (Figure 3.4). Ed-monton and Calgary are first and third, respec-tively, in metropolitan area comparisons (Figure3.5). These rates appear to be relatively stable overthe last seven years (Figure 3.6).

Public policy issues are created by the specialcharacteristics of liquor as a consumer product.The potential misuse of alcohol and its associ-ated problems lends support to those who call for

In 1993/94 the Alberta Government imple-mented major policy changes involving the con-trol, taxing, and distribution of liquor products.These changes included 1) privatization of theretail and warehousing functions, 2) switchingfrom an ad valorem (percentage of price) to a unit(flat) tax system of alcohol excise taxes, and 3)the ending of direct control of liquor regulationand moving to a legislative and enforcement ap-proach.

The main changes in the industry ten yearsafter privatization include the following: Retailoutlets (excluding off sales) have more than tri-pled from 310 to 983. The individual stock itemshave increased over five- fold, from 3,325 to17,000. Jobs have increased from approximately1,300 to 4,000, while wages have fallen from over$14 per hour (in current dollars), plus a benefitpackage and civil service pension, to approxi-mately $7 per hour. Warehousing has changed.The one publicly-operated warehouse in St.Albert, in 1993, is now operated by a private firm,and three private firms are licensed to wholesalebeer out of warehouses in Calgary and Edmon-ton.

Approximately 75% of Canadians consumealcoholic beverages. Estimates suggest 10% ofconsumers purchase 50% of the products sold.Average consumption of alcoholic beverages hasbeen falling over the decades across Canada (Fig-ure 2.1). Using international comparisons, Ca-nadians are more moderate drinkers when com-pared to Americans and Europeans (Tables 2.1 &2.2). The declining importance of expenditure onalcohol by Canadians is also shown by the fall inexpenditures on alcohol as a percentage of con-sumer expenditure. Alcohol expenditure in 2000

iv Canadian Centre for Policy Alternatives / Parkland Institute

public regulation and control in the marketing ofalcoholic beverages. A major responsibility of gov-ernment is to protect the interest of the generalpublic against the abuse of alcohol and its costsimposed by a minority.

The socially responsible marketing of alco-hol appears to be less effective with the privateretailing of liquor when compared to a public re-tail system. The efforts to restrict or prevent salesto certain high-risk individuals are incompatiblewith the profit motive in private marketing. Regu-lation and enforcement become necessary, which,at the very least, add additional public costs. Ad-ditionally, there is greater potential for illegal liq-uor manufacturing or smuggling and sales undera system with a large number of private retailers.Again, policing, necessary in order to prevent thisactivity, incurs increased public costs.

Taxes on alcohol are imposed by both federaland provincial governments. There are a numberof arguments to justify this type of excise tax. Liq-uor consumption generates social costs, liquor isa luxury product, it has an inelastic demand, andits consumption is complementary to leisure. Atax on liquor does two main things. It increasesthe price of alcoholic beverages, reducing con-sumption, as well as any associated social ills (Fig-ure 5.1). It brings in revenue to governments, andthe tax revenue obtained can (potentially) be usedto help correct or compensate for these negativeeffects.

Alcohol excise taxes can be either charged perunit or as a percentage of price (ad valorem). Anad valorem tax is greater the higher the price ofthe product, and has the advantage that it auto-matically rises with inflation. This type of tax alsohas a progressive tax effect if, as in the case foralcohol beverages, people with higher incomes buymore expensive alcohol products. Alberta shiftedto a unit tax system after privatization. The ad-vantage of a unit tax is that the amount of tax isindependent of the price of the product. Also, aunit tax does not change with price fluctuations—

an advantage where prices are volatile. A unit taxwill favour expensive products over cheap onesbecause the tax will comprise a smaller percent-age of the purchase price of the expensive prod-uct. The initial (1993) unit tax rates were de-creased as the private marketing of alcohol led toprice increases (Tables 4.1 & 4.2). This has re-sulted in lost revenue to the province from alco-hol beverages. Liquor revenue is down, measuredeither: as a percentage of total government rev-enues (Figure 4.1), on a per capita constant dol-lar basis (Figure 4.2), or on the basis of currentdollar per unit absolute alcohol sold (Figure 4.3).The lost revenue may exceed $500 million overthe last ten years (Table 4.3).

Liquor prices in Alberta track the Canadianaverage closely, but spike with privatization in1993, and then level out as tax rates decreased(Figure 5.2). When liquor prices are comparedwith “all other items” in Alberta, it appears theunit tax has dampened the changes in liquorprices, until they take off in 2001 (Figure 5.3).And, over the period of privatization, liquor priceshave been more volatile in Alberta than in therest of Canada (Figure 5.4). Over the last decade,liquor prices have increased more in Alberta thanin British Columbia (Figure 5.5). Current retailprices are similar in the two provinces, on aver-age, but show considerable variability betweenstores in Alberta (Table 5.1). If the unit tax sys-tem had maintained relative tax revenue, whole-sale prices would be considerably higher in Al-berta today than they are. These lower taxes havedampened the upward pressures on wholesaleprice, limiting retail price increases. The interest-ing thing about liquor is that a low price is not(or should not be) an objective in the responsiblecontrol of alcohol consumption. Public welfare ishigher when price is high, consumption is low,and revenues obtained compensate for the socialcosts.

The number of liquor stores in Alberta hasincreased dramatically and is high compared to

Sobering Result v

other jurisdictions (Table 6.1 & Figure 6.1). Liq-uor retailing in Alberta is currently a“monopolistically competitive” industry, one inwhich there are a large number of firms, but eachfirm produces a product that is “differentiated”from that produced by its competitors. Productdifferentiation occurs in a number of ways. Theseinclude location, the specific selection of prod-ucts to stock, product expertise, store decor, open-ing times, advertising, customer loyalty programs,and discounting. Since each firm produces a simi-lar but somewhat different product compared toevery other firm in the industry, there is an in-centive for each firm to play up the difference inits product in order to boost its sales. Product dif-ferentiation allows for an inefficient number ofoutlets and significantly increases the costs of re-tailing. Prices have increased, but not to the de-gree they might have, because the share taken asgovernment revenue has fallen.

The wide variability in prices now found inAlberta demonstrates the degree of differentiation.Additionally, liquor producers appear to be uti-lizing their market power in order to set pricesand pass on increased servicing and marketingcosts due to the numerous individual stores nowin the Alberta liquor retailing industry (Table 6.3).These costs have increased the landed costs, push-ing up wholesale prices (Figure 6.4).

Chain stores are now increasing their pres-ence as they realize economies of scale from ad-ministrative advantages, advertising, and mini-warehousing (Table 6.2). However, these advan-tages are currently constrained by the wholesaleprice system, where each firm pays the samewholesale and transportation charges, regardlessof the actual cost.

Conclusions:

• The retail liquor industry in Alberta has beenevolving over the past decade following the

change from a government monopoly to acompetitive private market and the changefrom an ad valorem tax to the unit (or flat)tax markup.

• Evidence on the links between alcohol con-sumption and social ills is overwhelming. Ab-solute alcohol consumption is high in Albertarelative to the rest of Canada and has begunto climb since 1997. The potential for in-creased social costs is real.

• Socially responsible marketing would educatethe public about such dangers as drinking anddriving and fetal alcohol syndrome. The pub-lic’s objective is to minimize the abuse of al-cohol through the limit and control of thesale of liquor, in particular to prevent the saleto underage consumers and the intoxicated.In contrast, the objective of private firms isto sell product. A publicly-owned and con-trolled system of distribution does not havethis inherent incompatibly of incentives.

• Private retailing of liquor has required greaterregulation and enforcement costs. Some ofthese costs are incurred in the Ministry whileothers are shifted to local police departments.

• Alberta Government revenues from the saleof alcohol have stayed constant in absolutecurrent dollars. This means that, with infla-tion, population growth, and growth in sales,revenues have fallen between 1993 and 2001.Prices have increased, but not to the degreethey might have because the share taken asgovernment revenue has fallen.

• The change from a government monopoly toa private market has resulted in amonopolistically competitive market struc-ture. Some consumer advantages includegreater convenience with the increase in

vi Canadian Centre for Policy Alternatives / Parkland Institute

number of stores, including more stores inrural small towns, and stores are open longerand later hours. The disadvantages includeinefficiencies in the form of excess capacity,duplication, and redundancy, particularly inurban centres. This inefficiency generates con-siderable higher costs of retailing, even thoughwages are at half of those compared to otherjurisdictions.

• The liquor retailing market has been handi-capped in its ability to achieve marketefficiencies by the control on the wholesaledistribution and transportation costs and therestriction requiring stand-alone outlets.However, there are some economies of scaleyet to exploit, facilitating the expansion ofretail chains. Future directions in the indus-try appear to favour large grocery chains suchas Safeway, the Real Canadian Superstore, theCalgary Cooperative Association Ltd., IGA,and others.

• Prices for liquor products in Alberta are com-parable to those found in British Columbia(January 2003). However, tax revenues re-turned to government are much lower in Al-berta. This means the privatization effort hasbeen supported and subsidized by the gov-ernment through a reduction in the tax shareof the final retail price. This reduction in taxrevenue has limited the greater escalation ofprices due to the cost increases caused by ex-cess capacity.

In summary, the Alberta government has losteffective control of the liquor industry, which willlikely continue to evolve into an oligopolisticmarket structure as chain stores get greater con-trol. Against the trends in other jurisdictions, liq-uor consumption has increased (with its poten-tial risks of increasing social ills), wholesale costshave risen, and retail prices have increased. Al-though retail prices have increased, the tax rev-enues to government have fallen significantly.

Sobering Result 1

In the provincial referendum of November 1923,following an unsuccessful attempt at prohibition,Albertans chose a government-controlled liquordistribution and sales system that promised tocontrol crime and raise money for the province.1

The Alberta Liquor Control Board (ALCB) be-gan operation the following year. A grand com-promise had been achieved between prohibition-ists and the ‘moderates’ who wanted legal accessto alcoholic beverages. Seventy years later, on Sep-tember 2, 1993, Dr. Steven West2 announced inthe legislature a major policy shift in the regula-tion and control of liquor. This announcementoverturned a liquor distribution system that ap-peared to be working well, for all intents and pur-poses, for over seven decades.

The Alberta government implemented thenew policy through the fall and winter of 1993and into the spring of 1994. The policy includedthree major changes in the way distribution ofalcoholic beverages was to be treated. First, thegovernment backed away from the direct regula-tion and control of liquor, moving to a muchlooser legislative and enforcement model. Second,the ALCB retail stores were closed and the build-ings or leases were sold. Retailing liquor in theprovince changed from a virtual monopoly, gov-ernment-owned and operated system, into a pri-vately-owned competitive market-driven indus-try. And third, the method of taxing liquor waschanged from an ad valorem system of percent-age markup over costs to a unit or ‘flat’ markupsystem.

Was this change purely ideological—a beliefin the transcendence of market competition? Cer-tainly, neither the demise of the ALCB nor thederegulation of alcohol was demanded by the

public; it was not an election issue in the 1993campaign; there was little if any debate, analysis,or other consideration of the implications for so-cial well-being that would arise from this dramaticchange in government policy; and there was noreferendum this time out.

1. A Brief Description of AlcoholicBeverage Distribution in Alberta

Prior to privatization in 1993, the distributionand sale of alcoholic beverages were almost com-pletely under the direct control of the government.There were 208 Alberta Liquor Control Board(ALCB) retail stores, 49 ALCB- regulated agencystores (first allowed in 1990), 30 private retail beerstores (introduced in the 1970s), and 23 privatewine boutique stores (first allowed in 1985), for atotal of 310 outlets. Additionally, there were 548licensed off-sales outlets in hotels and five manu-facturers’ off-sales. The number of products orstock-keeping units (SKUs) listed in the publicly-owned and operated central warehouse in St.Albert was 3,325. There were approximately 1,300full and part-time positions in ALCB retail stores,along with additional warehouse employees.Wages for a sales clerk were over $14 per hour (incurrent 1993 dollars) and full-time employees hadpublic service pensions and a full benefit pack-age. Prices were the same across the province inALCB-controlled liquor stores. The governmentset the retail prices by adding an ad valorem tax(markup) to the landed cost of the product. Thedifference between the landed cost and the retailsales constituted the gross revenue from sales. Thecosts of retailing, warehousing, and administra-

1 Introduction

2 Canadian Centre for Policy Alternatives / Parkland Institute

tion were deducted to obtain the net excise taxrevenue (profit) from alcohol, which was trans-ferred to government general revenues.3

Today, private firms retail, warehouse, and dis-tribute liquor—alcoholic beverages—in Alberta.The Alberta government continues to regulate theindustry. Specifically, the Alberta Gaming andLiquor Commission (AGLC) is responsible forissuing liquor licenses and for collecting alcoholicbeverage tax revenues. The Commission, throughthe Investigations Branch, also acts as an enforce-ment agent of the Gaming and Liquor Act and theGaming and Liquor Regulation. The current Min-ister of Gaming is Ron Stevens.

All spirits, wine, and beer are shipped bymanufacturers to privately-operated warehousesapproved by the AGLC. The current four licensedwarehouse companies include Connect LogisticsServices Inc. (CLS), a member of the Tibbett andBritten Group (UK), who own and operate themain warehouse in St. Albert. Brewers Distribu-tor Ltd. warehouses and distributes beer productsfor Molson and Labatt breweries from Edmon-ton and Calgary. Big Rock Brewery manufacturesand distributes beer from its plant/warehouse inCalgary. And Sleeman Breweries Ltd. warehousesand distributes its products from a warehouse inCalgary. The number of separate identified prod-ucts (SKUs) is over 17,000, reportedly five timesthe number prior to 1994. However, a SKU isissued for any variation in a product: for exam-ple, the same beer marketed in 355ml cans, 341mlbottles, and 625 ml bottles would count as threeSKUs.

Manufacturers or agents set basic cost priceson liquor products called the C.I.F. invoice price,which includes freight, agent fees, and insurance.Federal excise taxes and import duties are addedto make the landed cost. The wholesale price tolicensees includes the landed cost plus the recy-cling fee, the container deposit, the GST, and theAlberta markup (flat rate). Wholesale prices andshipping costs (subject to minimum order and

quantity) are the same regardless of the distanceof a retail store from the warehouse(s). The retailstores must pay cash at the time of ordering fortheir inventory and order minimum case lots.Retailers are completely free to set their own re-tail prices, including selling at below the whole-sale cost. Beer products dominate accounting for50% of provincial sales.

As of January 2003, there were 897 privateretail liquor stores and 86 general merchandiseliquor stores (rural locations), for a total of 983.Additionally, there are approximately 600 generaloff-sales liquor licensees. The provincial govern-ment does not restrict the number of outlets ortheir location, leaving this up to municipal andlocal authorities. It does limit hours of openingfrom 10:00 a.m, to 2:00 a.m. (with an additionalhalf hour for delivery), and restricts opening onlyone day a year: on Christmas Day, stores mustremain closed. The legal drinking age is 18 yearsof age (reduced from 21 in 1971), and 82% ofAlbertans (age 18 years and older) have consumedalcohol.4 Access to the market was initially bannedoutright to grocery chain stores, and then allowedonly as stand-alone store fronts. Chain stores arean increasing aspect of the current market struc-ture. Many of the current stores are ‘mom andpop’ operations where employment and wages(net income) are difficult to ascertain. However,the Ministry estimates there are over 4,000 fulland part-time jobs in the retail sector5 and oneestimate puts the 1996 average wage at about$7.00 per hour.6

1.2 Purpose and Methodology

The purpose of this study is to critically assess theliquor retail industry in Alberta after almost tenyears of experience since the liquor control poli-cies in Alberta were so substantially altered. Theassessment will apply an economics (public fi-nance) approach to review and analyze the cur-rent retailing of alcoholic beverages and the im-

Sobering Result 3

plications of this public policy change on thewelfare of Albertans. The future directions ofchange in the industry will also be considered.This analysis will build on the early work of Laxeret al and the various versions of Douglas West’sstudy of the industry. The Laxer study,7 comingso soon after privatization, was more theoreticaland quite prescient in its predictions. The Weststudy8 was more descriptive and empirical, com-ing out after greater experience with privatization.These studies, taken together, provide a goodbackground to the process of change in retailingliquor in Alberta in the early and mid- 1990s.

This study takes a more analytical approachto the current situation in liquor retailing in Al-berta. In order to do this thoroughly, it must touchon the theoretical issues surrounding the uniqueaspects of liquor as a consumer product. Theseinclude the concepts of externalities, optimal pric-ing, price and income elasticity, and industrial (ormarket) structure: monopoly, oligopoly, perfectcompetition, and monopolistic competition.Understanding the market structure of the indus-try as it affects the current evolving private retail-ing of liquor is vitally important. The change frompublic ownership of the retailing part of the in-dustry to the private ownership and proliferationof retail outlets in a competitive environment isevaluated. The costs associated with each marketstructure and the scale (economies) issues will bediscussed to give some insight into the currentand future organization of the industry. The con-sumer and social issues affected by policies on tax,price, and availability are considered. However,liquor retailing is not like other product retailingwhere the economics of the situation may be pre-dominant. There are numerous social issues re-lated to alcohol consumption, and these must beaddressed by public policy relating to liquor re-tailing.

The information used in this report is calcu-lated primarily from Statistics Canada data, andfrom reports issued by Alberta and other govern-

ments, Health Canada, police services, and theAlberta Alcohol and Drug Abuse Commission(AADAC). A small primary price survey con-ducted in January 2003 provided price data for24 Alberta retail stores over ten commonly avail-able alcoholic beverage products. Other price datawere obtained from Connect Logistics Liquorwholesale price list and the British Columbia Liq-uor Stores Product guide. Economic analysis, in-cluding optimal price theory and industrial struc-ture theory, supports much of the discussion.Qualitative information was obtained throughinterviews with various stakeholders and numer-ous short discussions with liquor store owners andstaff. The objective of these interviews was to sup-plement, strengthen, and deepen the understand-ing of the industry obtained from the publisheddata and reports.9

1.3 Outline

After this introductory section, the report willconsider the relative and changing importance ofalcoholic beverages in Canadian spending pat-terns. Data on international, national, provincial,and municipal consumption trends are presented.Section 2 also outlines the public policy issuescreated by the special characteristics of liquor as aconsumer product. The third section reviews someof the social issues relating to alcohol consump-tion. It presents national, provincial, and localcrime-rate statistics in order to consider the ef-fects on crime, including impaired driving, giventhe greater availability of liquor with the increasein privatization retailing. The conflict betweenprofit incentives and socially responsible market-ing is touched on. Section 4 evaluates specific taxa-tion of alcoholic beverages, including the ration-ale for liquor taxes. The changes in tax assessment,collection, and revenues obtained in Alberta arecompared to those in British Columbia wherethe tax system used is similar to the former ALCB


regime. The fifth section looks at alcohol avail-ability, prices, and other characteristics of deliv-ery affecting consumers and society. Section 6contains the analysis of the current market struc-ture in the private retailing of alcoholic beverages.Limited cost information, combined with per-

sonal observations and discussions with retailers,is used to assess the consequences of the indus-trial structure on the liquor retaining industry.Future directions are also considered. The con-clusions of the study are enumerated in Section7.

Sobering Result 5

Approximately three-quarters of Canadians con-sume alcoholic beverages. Average consumptionof alcoholic beverages has been falling in Canada,and Canadians are more moderate drinkers whencompared to Americans and Europeans. Albertahas the highest per capita consumption of alco-hol in Canada when measured by dollar value ofexpenditures or absolute alcohol consumed.Calgary rates the highest for major metropolitanareas, while Edmonton is a modest eighth. Theconsumption of alcoholic beverages is positivelycorrelated with income. When calculated as a per-centage of income, Alberta’s expenditures on al-cohol are closer to the national average and thoseof its provincial neighbours.

The importance of expenditure on alcohol isalso declining when measured as percentage ofconsumer income. Two-thirds of alcoholic bever-ages are sold in retail outlets, and approximatelyone-half of the sales are on beer products. Al-though consumption expenditures have fallen, itis still important to regulate liquor sales given thenature of alcohol and that estimates suggest tenpercent of consumers purchase fifty percent of theproducts sold.

2.1 Alcohol Consumption Concerns

Health Canada reported in 1995 that about 16.7million Canadians or 72.3% of women and menaged 15 and over have consumed alcohol in thepast 12 months, a drop of 5.4% since 1989. Thisreport also shows the extent to which people areconcerned about others’ liquor consumption.10

The proportion of Canadians who drinkcontinues to decline nationally; however,

2 The Importance of Alcoholic Beverages

provincial variations exist. The vast major-ity of both current and former drinkers(79.2 per cent) feel their own consump-tion has not harmed them. On the otherhand, 73.4 per cent of all Canadians --drinkers and non-drinkers alike -- say theyhave been harmed in some way at somepoint in their lives by others' drinking. Inthe past 12 months, 10.5 per cent of cur-rent drinkers reported at least one harmfuleffect resulting from their drinking. Physi-cal health problems are reported by 6.2 percent of current drinkers and of those whoare parents, 1.3 per cent perceive their al-cohol use as harmful to their children. Ap-proximately one in five current drinkers(20.3 per cent) state that they drove afterconsuming two or more drinks in the pre-vious hour. In 1989 the percentage was 22.8per cent.

2.2 International Comparisons

The World Health Organization (WHO) hasconducted an impressive international study onper capita alcohol consumption and its socialimpacts and costs. The report listed 137 coun-tries in order, from heaviest to lightest consum-ing countries, for 1996, the most recent year forwhich data were available. The estimates rely onpopulation data from the United Nations (UN).Adult population figures (age 15+) were used toadjust for the differing age structures of nationalpopulations. Table 2.1 shows a selection of thecountries from the WHO document.11

Ranking countries from the highest to lowestper capita consumption, Canada was ranked


Table 2.1 Per capita consumption of pure alcohol equivalent (litres) in 1996RANK COUNTRY TOTAL BEER SPIRITS WINE1 Slovenia 15.15 5.76 0.89 8.5029 United Kingdom 9.41 6.34 1.72 1.9431 United States of America 8.90 5.36 2.43 1.1248 Canada 7.52 4.23 2.16 1.19137 Cambodia 0.34 0.14 0.20 -

48th—well below the United States and the UnitedKingdom. The United Kingdom was lower thanall European countries. Table 3 in the WHOdocument shows the changing consumption pat-terns over time. It compares total adult per capitaconsumption of pure alcohol for two periods,1970-1972 and 1994-1996. Three-year averageswere used to minimize the impact of short-termtemporal fluctuations in adult alcohol consump-tion. Data are available for both time periods forthe 137 countries, drawing again on World DrinkTrends and on the Food and Agricultural Organi-zation (FAO) and UN statistical databases.

The WHO table shows that developing coun-tries and countries in transition were more likelyto increase their recorded adult per capita con-sumption of alcohol than the developed coun-tries. Forty-seven percent of the developing coun-tries or countries in transition showed increasedalcohol consumption since 1970, whereas 35%of the developed countries recorded higher con-sumption of alcohol per adult.12

Using data from the WHO document, Table2.2 compares Canada, the United Kingdom, andthe United States. Both Canada and the UnitedStates have experienced a considerable decline inper capita consumption, with Canada decreasingalmost double the United States percentage, start-ing from a slightly lower average consumption

level. The United Kingdom increased betweenthese two periods.

This international comparison indicates thatconsumption of alcoholic beverages in Canada isrelatively low for a developed country, has beenfalling over the 25-year period covered, and islower than that in United States and the UnitedKingdom (and Europe).

2.3 Canadian, Provincial, andMunicipal Comparisons

Alcohol consumption levels have changed inCanada and also vary considerably across the na-tion. Comparing provinces, Alberta has the high-est per capita consumption of alcohol when meas-ured by dollar value of expenditures or absolutealcohol consumed. Calgary (tied with Toronto)rates the highest for major metropolitan areas,while Edmonton is a modest eighth. The con-sumption of alcoholic beverages is positively cor-related with income. When calculated as a per-centage of income, Alberta’s expenditures on al-cohol are closer to the national average and tothose of its provincial neighbours.

Figure 2.1 illustrates the changing levels ofsales of alcohol in Canada between 1983 and2001. Sales fell significantly between 1983 and1997 for all jurisdictions. The Canadian average

Table 2.2 Trends in per capita consumption of pure alcohol equivalent (litres) in 1996COUNTRY 1970-1972 1994-1996 PERCENT CHANGECanada 9.16 7.62 -16.81United States of America 9.92 8.98 -9.48United Kingdom 7.35 9.25 +25.85

Sobering Result 7

fell from 10.44 to 7.2. Alberta fell from 12.46 to8.1. The West (and the North) of the country hastraditionally consumed more alcohol. However,Alberta and British Columbia have had the great-est decrease in sales, with British Columbia reach-ing the national average in 2000. After 1997, thenational average has started to climb somewhat,to 7.7 in 2001. Alberta never reached the lowernational average and also started to increase after1997, to 8.6 litres per person in 2001.

Another way to look at the importance of al-cohol is to consider its place in the individual’s orfamily’s budget. The production of the consumerprice index (CPI) uses a basket of goods derivedfrom what average Canadians actually spend theirincome on. The ‘basket’ includes hundreds ofproducts collected into a number of consumercategories. Each individual product and each con-sumer category is given a weight determined bythe percentage of household income spent on theproduct(s). The construction of the CPI weightscan give us some indication of the financial im-portance of alcoholic beverages to Canadians.13

The main sources of expenditure data on con-sumer goods and services are the family expendi-

ture surveys conducted periodically by StatisticsCanada. Until 1996, these were called the FamilyExpenditure Survey, thereafter they are called theSurvey of Household Spending and the Food Ex-penditure. These surveys provide estimations usedto derive weights in the CPI basket. Average yearlyexpenditures per household are calculated for eachcommodity class by province or sub-provincialarea. These are then applied to the estimatednumber of households in each geographical area,giving aggregate expenditures for each commod-ity class. Aggregate expenditures for Canada areobtained from estimated aggregate expendituresfor each basic commodity group for each geo-graphical unit.

Using the data obtained from the 2000 Sur-vey of Household Spending and the Food Expendi-ture, Figure 2.2 shows the average expenditure onalcohol products by province and the Canadianaverage. Alberta is tied with Ontario at $721 and$722, respectively, compared to a national aver-age of $677.14

When cities are compared, Calgary and To-ronto are the highest at $837 spent on alcoholbeverages per household; the next highest city is

Figure 2.1 Volume of Sales of Alcoholic Beverages in Litres of Absolute Alcohol Per Capita 15 Years and Over

6

7

8

9

10

11

12

13

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

Fiscal Year

Lit

res

/ca

pit

a

Canada

Alberta

British Columbia

Saskatchewan

Ontario


Montreal at $747. Edmonton households spenton average a relatively modest $678.

Alberta and Ontario have the highest con-sumption among the provinces, as measured byconsumer expenditures (Figure 2.2). However, Al-berta and Ontario both are relatively rich prov-inces with high average incomes. Expenditures onalcohol are related to income; the higher the in-

come group the higher the expenditure on alco-hol.

Johnson et al have estimated price and incomeelasticities [see box] for beer, wine, and spirits foreach of the provinces of Canada using data overthe period 1956-83. The estimates vary markedlyacross provinces. For Canada (calculated asweighted averages of provincial estimates), all in-

Figure 2.2 Average Expenditure per Household on Alcohol in 2000

$0 $100 $200 $300 $400 $500 $600 $700 $800

Canada

Newfoundland/Labrador

Prince Edward Island

Nova Scotia

New Brunswick

Quebec

Ontario

Manitoba

Saskatchewan

Alberta

British Columbia

Figure 2.3 Average Expenditure per Household on Alcoholic beverages in 2000 Selected Metropolitan Areas

$0 $100 $200 $300 $400 $500 $600 $700 $800 $900

Saint John

Charlottetown

Victoria

Quebec City

Winnipeg

St.John’s

Edmonton

Saskatoon

Regina

Ottawa

Vancouver

Halifax

Montreal

Toronto

Calgary

Sobering Result 9

come elasticities were found to be positive: .27for beer, between 1.02—1.33 for spirits, and be-tween .19—2.62 for wine. The estimated incomeelasticity of beer is small (inelastic), while the es-timated income elasticities for spirits and wineare substantially larger (elastic), especially in thelong run.15 The positive relationship between in-come and liquor consumption is also demon-strated in Figure 2.4.

As the consumption of alcoholic beverages ispositively correlated with income, the differencein incomes in each province should be taken intoconsideration when comparing provincial alco-hol consumption levels. Figure 2.5 shows alcoholexpenditures as a percentage of total consumerexpenditures. There is far less variance in provin-cial expenditures when income levels are consid-ered. Although Newfoundland and Quebec stand

Figure 2.4 Average Expenditure on Alcoholic Beverages per Household by Income Quintiles Canada (10 Provinces) 2000

Highest Quintile $82,402 and over

Fourth Quintile $55760 to $82,402

Third Quintile$37000 to $55,760

Second Quintile$21216 to $37,000

Lowest QuintileLess than $21,216

All Classes

$0

$200

$400

$600

$800

$1,000

$1,200

$1,400

Exp

en

dit

ure

s

Figure 2.5 Alcohol Expenditures as a Percentage of Total Expenditures (Income)

0.00% 0.25% 0.50% 0.75% 1.00% 1.25% 1.50%

Canada

Newfoundland/Labrador

Prince Edward Island

Nova Scotia

New Brunswick

Quebec

Ontario

Manitoba

Saskatchewan

Alberta

British Columbia

Percentage


ElasticityElasticityElasticityElasticityElasticity (e) is an extremely useful concept. It is a detailed numerical calculation using data on a product’sdemand. It measures the response of the quantity bought to either a price change, an income change, or anyother variable of interest.

Demand is the general inverse relationship between price and quantity bought/sold. The ratio of the per-centage change in the quantity bought when a percentage change in price occurs is called the price elasticityof demand. If the percentage change in quantity exceeds the percentage change in price, the ratio is greaterthan one, and the good is considered price elastic. When this ratio is less than one, the good is considered priceinelastic. High price elasticity (elastic demand) means a small price increase will cause a large reduction insales of the good. For example if a product’s price went up ten percent and the consumers bought one-half thequantity, then e = 5. Consumers respond strongly to price. Low price elasticity (inelastic demand) means alarge price increase will cause a small reduction in sales of the good. For example, if a product’s price doubledand consumers bought only ten percent less then e = .1. Consumers are not very responsive to price.

Income elasticity measures the change in quantity bought with changes in consumers’ incomes. It is calcu-lated the same as price elasticity, only it is the ratio of percentage change in quantity divided by percentagechange in income. For normal goods the ratio it has a positive sign. That is, the quantity increases with incomeincreases. Again, a ratio number greater than one indicates elastic demand—consumers buy more propor-tionally than the income increase. If the elasticity number is less than one, consumers buy proportionally lessthan the increase in income. For example, with liquor products all were found to be normal—higher incomesmeant greater expenditures on them (e = +).

out, each has lower absolute expenditures on al-coholic beverages than either Alberta or Ontario,but each spends a higher proportion of incomeon alcoholic beverages. When expenditures onalcoholic beverages are taken as a percentage oftotal consumer expenditures (or income), Albertaappears similar to the national average and to itsneighbour provinces.

Figure 2.5 also brings home the point thatconsumer expenditures on alcohol are a smallpercentage of the average household’s budget.From a consumer expenditure basis, alcoholicbeverages as a category are just 1.21% for Canadaand 1.16% for Alberta. The declining importanceof expenditure on alcohol by Canadians is alsoemphasized by the Survey(s) of Household Spend-ing and the Food Expenditure. The percentage ofconsumer expenditure on alcohol in 2000 at1.72% (Figure 2.4) is 7.5% lower than the 1996CPI basket weight for all alcohol beverages, whichwas 1.86% (as listed in Table 2.3).

How important is the retail side of liquor?Table 2.3 shows the breakdown of liquor sales

through liquor stores and those served in licensedpremises. The 1996 survey of consumer expendi-tures shows that the retail store component isabout two-thirds of all sales and beer is about one-half of all alcoholic beverages sold in stores.16 Thedata on sales shown in the AGL reports indicatethe same ratios for Alberta and also that 50% ofsales are on beer.17

2.4 The Rationale for Public Regulation

Public policy issues are created by the special char-acteristics of liquor as a consumer product. Alco-holic beverages are potentially dangerous goods,particularly in the hands of inexperienced youth,and the misuse of alcohol and its associated prob-lems lend support to those who call for publicregulation and control in the marketing of alco-holic beverages. A major responsibility of govern-ment is to protect the interest of the general pub-lic against the abuse of alcohol and its costs.

There is little doubt that attitudes about theuse of alcoholic beverages have changed for the

Sobering Result 11

vast majority of Canadians who are consumingless alcohol. An important question arises in theevaluation of liquor distribution. What is thepublic purpose? What are the public’s objectivesin being involved in the sale and distribution ofalcoholic beverages at all? When adjusted for in-come differences, expenditures on alcohol aresimilar across the nation and on average a smallfraction of income is spent on liquor. However,the information on expenditure and consump-tion patterns in section 2.2 is based on averageconsumer behaviour. This can be misleading whenconsidering the use or abuse of alcohol. Althoughliquor sales are relatively insignificant to averageCanadians, they are of considerable social impor-tance when ‘hard core’ consumers are considered.For example, the World Health Organizationstudy on alcohol states:18

Estimates of per capita consumption of al-cohol across the entire population aged 15years or older can provide policy makerswith some sense of the magnitude andtrends likely to be found in alcohol relatedproblems. Among those who drink at all,the heaviest drinking 10 per cent typicallyaccount for 50 per cent or more of all alco-hol consumed. Per capita alcohol consump-tion trends thus may be a good proxy forproblems of chronic heavy drinking suchas cirrhosis of the liver, but less so for prob-lems typically more widely distributed suchas alcohol-related traffic casualties (Edwardset al., 1994). Adult per capita alcohol con-sumption estimates can be indicative of theextent of alcohol-related problems.

Many who manage retail liquor outlets sup-port the view that a small proportion of custom-ers account for a large proportion of sales. Thepotential misuse of alcohol and its associated prob-lems lends support to those who call for publicregulation in the marketing of alcoholic bever-ages.

Controlling the distribution and sale of alco-holic beverages has a long history in Canada com-ing out of prohibition. The public distaste at theturn of the 20th century with the abuses of alco-hol led provincial governments, including Alberta,to implement various systems of prohibition ofalcoholic beverage sales and consumption. Gov-ernments responded to the public’s wishes. Aftera period of experience, however, prohibition wasdeemed an unsuccessful experiment, causing moreproblems than it solved: increased crime, the in-volvement of organized crime, and all of the dis-tortions that are caused when those willing tobreak the law can profit greatly by selling a re-stricted substance. Prohibition ended in 1924 inAlberta. The political compromise between pro-hibitionists and others was the establishment ofprovincial liquor boards to take over the controland sale of alcoholic beverages. The sale of alco-holic beverages was gradually expanded in the1970s and 1980s when the province allowed pri-vate firms to sell cold beer and wine and with theestablishment of agency stores to give better serv-ice to rural areas.

Alcoholic beverages are still considered bymany to be potentially dangerous goods, particu-larly in the hands of inexperienced youth. Nu-merous social ills are correlated with alcohol over-use or abuse:19 More recently, the problem andsocial costs of fetal alcohol syndrome (FAS) have

Table 2.3 Percentages of alcohol beverage expenditures in 1996All Alcohol sales: Served Purchased from stores

Beer Wine Liquor1.86 .58 1.28 .60 .30 .31


been appreciated. Costs can run in the millionsof dollars in the social services and the criminaljustice systems for just one FAS victim, as a papercommissioned by Health Canada reports:20

FAS is expensive and pervasive. It is im-possible to measure the exact costs to soci-ety of each FAS child. Beyond the healthcare costs, there are the costs of special edu-cation, foster-care, incarceration, financialassistance and social services, and the indi-rect costs related to maladaptive behaviours,goods and services not produced, and, mostimportant, the loss of human potential. Theestimated lifetime costs related to healthand education for one person with FAS mayexceed $1.4 million over a lifetime(Streissguth et al, 1996). These costs arestaggering considering that FAS is entirelypreventable. The costs must be paid by theaffected children and families and by therest of society. For this reason the conse-quences of alcohol exposure in utero are ofconcern to us all.

A major responsibility of government is toprotect the interest of the general public againstthe abuse of alcohol and its costs imposed by aminority. This is as true today as it was duringthe call for prohibition. This responsibility for the

sale and distribution of alcoholic beverages hasbeen recognized by governments across Canada.It is still recognized in Alberta today. However,since 1994 the Alberta government has imple-mented the regulatory function through the lawand its enforcement rather than direct controlthrough liquor distribution and sales. Is this ap-proach sufficiently effective, both in complianceand costs, to meet the demands for socially re-sponsible liquor distribution and use? The poten-tial conflict in incentives of this approach is dis-cussed further in Section 3.

The control of alcohol should also fit withother important government objectives. Effi-ciency, both economic and productive, should alsobe of concern to governments. The general socialconditions, for example, working conditions, citi-zens’ incomes, education, and health are of primeimportance. Alcohol distribution and consump-tion is intimately intertwined with all of these con-cerns. From the public’s point of view, the ulti-mate question regarding the regulation of liquoris whether the alcohol distribution system mini-mizes the dysfunctional social consequences thatoften result from alcohol consumption whilebringing in the most tax revenue possible for thegovernment to spend on education, health, andsocial services in order to offset the effects of abuseof alcohol.

Sobering Result 13

3 Social Issues

on due to societal, cultural, and demographicchanges with respect to the consumption of alco-hol. We want to isolate the small changes due tothe liquor distribution system from the largersocietal changes happening in the background.The statistics we have are unsatisfactory. What isneeded is a theoretical model that would betterexplain the differing factors causing alcohol abuseand its effects. Mostly we have only correlations.Greater analytic work beyond the scope of thisreport needs to be pursued on this issue. None-theless, some data will be presented below.

3.1 Crime

It is well accepted that much crime is associatedwith alcohol abuse. For example, as reported inJuristat:24

Alcohol, drugs, and other intoxicants areknown to play a role in the commission ofmany crimes including homicide. In 2000,police reported that 33% of homicide vic-tims and 44% of accused persons had con-sumed alcohol and/or drugs at the time ofthe offence, consistent with the pattern seensince 1991 when this information was firstcollected in the Homicide Survey. As vic-tims of homicide, men were 50% morelikely than women to have consumed alco-hol and/or drugs, and as accused, were 25%more likely than women to have consumedalcohol or drugs.

It might be expected that crime would in-crease with an increase in private retailing in al-cohol beverages, where the awareness and preven-

There are numerous social issues relating to alco-hol. Against the general backdrop of falling crimerates in Canada there is local evidence that crimeis associated with alcohol availability and price.Greater ethnographic study needs to be done inorder to determine if liquor privatization, and theresulting changes in marketing, are associated withcrime rates. Alberta has the second highest im-paired driving charge rates in Canada, and Ed-monton and Calgary are first and third, respec-tively, in metropolitan area comparisons.

It is well known and documented that socialills are associated with alcohol use and abuse. Thesocial, health, and economic burdens of excessalcoholic consumption include alcoholism, illness,injury, and loss of life, loss of worker production,property damage, crimes and violence, social dis-cord and family tension, impaired driving and itsconsequent loss of life, and a host of other prob-lems.21 Additionally alcohol harms disproportion-ately youths, Aboriginal people, inner city resi-dents, and the poor. Babor presents a good sum-mary of the economic and epidemiological lit-erature.22 In 1995-96 (the most recent available),the total federal and provincial tax revenues fromliquor were $3.78 billion, while the social costswere estimated at $5.25 billion.23

To the degree that average consumption ofalcohol is falling, these social ills will also decrease,all other things constant. Does the increased avail-ability of liquor in Alberta since privatization af-fect liquor consumption and therefore its aber-rant effects? This is a very difficult question toanswer analytically. One would need to separateout the effect of the change in liquor distribution(from a public monopoly to a private competi-tive market) from the background changes going


tion of criminal activity may be less prudent thana publicly-run distribution system. Along with thefall in alcohol consumption (Figure 2.1), the crimerates in Canada have been falling in most catego-ries from the highs of the early 1990s. The fol-lowing figures illustrate the crime rates in threecategories: all crimes, robbery, and break-and-en-try. These categories are selected based on the as-sumption that crime in general may be affectedby the general retailing circumstances as well asthe robbery rate, if retail liquor stores are morevulnerable, and that private retail liquor outletsmay be an additional attractive break-and-entertarget.

Police forces do not necessarily keep statisticson the type of retail outlet in their accounting foreither robbery or break-and-entry. Some data fromthe Calgary Police Service (CPS) indicate in-creased criminal activity in conjunction with theincrease in liquor stores under privatization. How-ever, these stores may have initially provided amore convenient and vulnerable target comparedto the alternatives. The CPS has worked with liq-uor merchants in order to improve security andpreparedness for criminal activities, to good ef-fect.25

The data used in the following figures are Sta-tistics Canada provincially aggregated data. Fig-ure 3.1 illustrates the crime rates for all offencesfor Alberta, and includes British Columbia andCanada for comparison.

The evidence illustrated in these figures indi-cates that British Columbia’s total crime rate ishigher, by as much as 50%, than the national av-erage but has fallen in the 1990s to 2001 at thesame rate of change as the national rate. The Al-berta rate rose over the 1980s, peaking in 1991.Then it fell dramatically, becoming equal to thenational average in the mid-1990s. It levelled outin the late 1990s again, becoming higher than thenational average after 1995.

The robbery rates in British Columbia, Al-berta, and the nation in the mid-1970s were muchthe same. Thereafter, British Columbia rates grewto be considerably greater than the Alberta rateand the national average, while Alberta’s rate fellbelow the nation’s rate. Although Alberta’s ratewas below that of the nation in the mid- 1990s, itstarted to climb after 1995. British Columbia’srate during this period, although high, has fallenat a very fast rate.

Figure 3.1 Crime Rates: All Offences Total

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

Year

All C

rim

es/1

00,0

00 p

op

ula

tio

n

Alberta

British Columbia

Canada

Sobering Result 15

Break-and-enter crime rates are somewhat lessvolatile. Alberta’s rate and the national rate fallsteadily from 1991, and the Alberta rate dropsbelow the nation’s in 1993. British Columbia’sbreak-and-enter crime rate did not start fallinguntil 1996 and then dropped quickly from 51%to 34% higher than the national average.

There appears on the surface to be little cor-relation between crime and the policy changes inAlberta to liquor distribution. The all offences rate

fell after a peak in 1991 to a low in 1994, where ithas remained pretty much constant. A steady dropin robbery rates was stopped at the time of priva-tization when it rose and then remained flat.Break-and-enter has been on a continual decline.This may be explained by the initial lack of aware-ness of new retail entrants, who, after advice bypolice, better prepared themselves against thepossibility of robbery.

Figure 3.2 Crime Rates: Robbery

0

20

40

60

80

100

120

140

160

180

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

Year

Ro

bb

ery

/100,0

00 p

op

ula

tio

n

Alberta

British Columbia

Canada

Figure 3.3 Crime Rates: Break-and-Enter

0

500

1000

1500

2000

2500

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

Year

Bre

ak

& E

nte

r/1

00

,00

0 p

op

ula

tio

n

Alberta

British Columbia

Canada


However, as one concerned academic notes:“Nor can one hide behind a fog of empirical un-certainties about the connections between liquor,disorder, and crime. In the end, academic statis-tical exercises are no substitute for live ethno-graphic realities.”26 More localized data may re-veal some interesting results. For example,evidence indicates that liquor act violations orincidents have risen considerably since the priva-tization of liquor retailing. The following figuresillustrate liquor-related incidents in violation ofthe Alberta Gaming and Liquor Act in the city ofEdmonton.

The data in Table 3.1 would suggest that, al-though crime rates have been falling throughoutthe nation over the 1990s, specific violations ofthe Alberta Gaming and Liquor Act, a provincialstatute, have risen dramatically in conjunctionwith the greater availability of liquor.

3.2 Impaired Driving

Other extremely serious criminal activity occursin impaired driving. The following quote bringshome the immense tragedy of this:27

Drinking-driving collisions are one of thelargest sources of alcohol-involved deaths

and injuries. In the 20-year period between1977 and 1996, the estimated numbers ofdeaths involving a drinking driver is in ex-cess of 35,000. The problem is particularlytragic among the young, but it is by nomeans restricted to them. Alcohol is theleading contributor to deaths on our high-ways, and in recent years it has been de-tected in about 40% of all drivers killed.The societal impact of injuries is muchlarger than that of deaths. The number ofpeople seriously and often permanentlyinjured is conservatively estimated to be atleast 10 times the number of people killed.Research indicates that the more serious thecollision, the more likely it is that alcoholis involved. Impaired driving is one of thelargest contributors to the social and eco-nomic costs of alcohol abuse in Ontario andCanada. It is clear that there are unaccept-able numbers of alcohol-related deaths andinjuries on our roads.

Figure 3.4 compares impaired driving chargesacross the provinces and for Canada. In 1998,Alberta had the second highest rate of impaireddriving at 469 per 100,000 population. When thisis compared to a national average of 295, Albertawas 59% higher than the national average.

Table 3.1 Edmonton Gaming and Liquor Act incidents, Edmonton Police Service,Rates per 100.000 persons 15yrs or older

YearLiquorActGenerally

Consumein PublicPlace

Conveyingin MotorVehicle

IllegalPoss-ession

Intoxi-cations

Minor-LicensedPremises

Minor-ObtainLiquor

SupplytoMinor

Sale andKeep forSale

Total

1994 1.97 5.42 8.89 0.08 8.27 1.27 1.11 0.08 0.17 27.281995 2.15 5.56 11.52 0.03 8.87 1.53 1.27 0.18 1.01 32.111996 2.82 6.78 7.27 0.16 11.81 2.63 0.63 0.10 0.67 32.871997 2.53 8.99 7.42 0.42 17.49 1.84 1.25 0.00 0.16 40.101998 2.41 11.62 6.19 0.03 18.61 2.06 0.99 0.00 0.19 42.101999 3.92 12.56 6.20 0.17 19.94 1.65 1.14 0.02 0.26 45.862000 5.51 10.63 5.73 0.12 26.05 1.76 0.91 0.02 1.18 51.912001 5.61 14.13 6.89 0.17 36.36 2.13 0.83 0.00 0.14 66.252002 25.01 0.24 8.95 0.22 38.07 0.96 0.89 0.01 0.21 74.55

Sobering Result 17

Figure 3.5 compares impaired driving chargesin 1998 for the nine largest metropolitan areas inCanada. Edmonton and Calgary came in at firstand third, respectively, in comparison with thenine largest metropolitan areas.28

Figure 3.6 illustrates impaired driving statis-tics over time. Alberta’s high rate of impaired driv-

ing has continued over the period these statisticshave been compiled since 1995. To the degree thatcharges indicate the problem, impaired driving isfar too prevalent; however, it does not appear tobe getting any worse. Conviction and related pen-alties are clearly not a sufficient deterrent to im-

Figure 3.4 Rate of Persons Charged withImpaired Driving by Province, 1998

683

469

389

381

375

307

295

251

224

218

0 100 200 300 400 500 600 700 800

Saskatchewan

Alberta

Manitoba

New Brunswick

Prince Edward Is.

Nova Scotia

Canada

Quebec

British Columbia

Newfoundland

Rate per 100,000 populations 16 years and older

Figure 3.5 Rate of Persons Charged with Impaired Driving by Nine largest Metropolitan Areas, 1998

374

297

264

245

201

178

166

148

126

0 50 100 150 200 250 300 350 400

Edmonton

Quebec

Calgary

Winnipeg

Ottawa-Hull

Montreal

Hamilton

Vancouver

Toronto

Rate per 100,000 populations 16 years and older


paired driving, and more needs to be done to re-duce impaired driving in Alberta.

Impaired driving and the collisions, injuries,and deaths resulting from it can be reduced bypublic action. Prevention of these injuries anddeaths is still an important priority for the gen-eral public. As well as other actions, research hasdemonstrated that factors that influence the con-sumption of alcohol by individuals andpopulations will also influence rates of drivingafter drinking and resulting problems:29

Measures that determine the economic, le-gal and social availability of alcohol will alsoinfluence drinking-driving rates. Thus, forexample, alcohol taxes play an importanthealth role in reducing drunk-driving fa-talities. In general, any measure that willtend to increase alcohol consumption ormake alcohol more accessible will also tendto increase rates of drinking-driving. Con-versely, measures that tend to reduce alco-hol consumption will tend to decrease ratesof drinking-driving.

3.3 Socially Responsible Marketing

The socially responsible marketing of alcohol ap-pears to be less effective with the private retailingof liquor when compared to a public retail sys-tem. The efforts to restrict or prevent sales to cer-tain high-risk individuals are incompatible withthe profit motive in private marketing. Regula-tion and enforcement becomes necessary which,at the very least, adds additional public costs.Additionally, there is greater potential for illegalliquor manufacturing or smuggling and sales un-der a system with a large number of private retail-ers. Again policing, necessary in order to preventthis activity, incurs increased public costs.

Unlike most market commodities, the adher-ence to socially responsible marketing of consumeralcohol is an important public concern. Sociallyresponsible marketing means promotion of mod-erate drinking behaviour, educating the publicabout the potential risks of alcohol, particularlyfetal alcohol syndrome, drinking and driving, andnot selling to those underage or to intoxicated per-sons. Socially responsible behaviour is a cost tothe retailer in time, store space for promotional

Figure 3.6 Alberta Impaired Driving Charges

0

2,500

5,000

7,500

10,000

12,500

15,000

17,500

20,000

22,500

25,000

27,500

30,000

1995 1996 1997 1998 1999 2000 2001

Year

Nu

mb

er

Total Charges

Guilty

Superior Court

Other Disposition

Stay/Withdrawn

Acquited

Sobering Result 19

material, and in lost revenue (profit) from notselling. The private retailing of liquor creates anincentive-incompatible situation regarding so-cially responsible marketing. Although there aremany conscientious business people who can ap-preciate more than the bottom line, the fact isthat private stores exist to make a profit. Whenthese retailers are running meagre margins, work-ing long hours, and are open late at night (earlymorning), the potential damage of alcohol abuseis increased.

A particularly important concern to the pub-lic is liquor sales to minors. The minimum agefor purchasing liquor in Alberta is 18. The changeto private retail stores has changed the incentivesto control this age-limitation. Understanding thisconflict of incentives, AGLC requires private re-tailers to ask for identification if a patron appearsto be under 25. Failure to do so is a punishableoffence if this identification is not performedwhen appropriate. The government has to usecrime and punishment techniques to overcomeand change the private incentives. This problemwas brought to the forefront when a televisionstation had under-age persons successfully pro-cure alcoholic beverages at a number of liquorstores.30 Subsequently, the AGLC raised the finesfor selling to a minor and for not requesting IDfrom those appearing to be under 25.

The effectiveness of this approach is in doubt.Compliance depends on the probability of loss tothe offender. This loss, in turn, not only dependson the levels of fines but it also depends on theenforcement effort. Making this approach to regu-lation work, at the very least, adds additional costson the administrative and regulatory regime, overand above that of a publicly-run system wherethe incentives for enforcement are compatiblewith the public control on sales. Under publicly-owned liquor retailing sales, staff can be trainedin how to handle and prevent under-age pur-chases, including procurement for a minor by anadult. The incentive structure is appropriate in

this situation as the government employee has noself-interest in selling to a minor or an intoxicatedperson.

Are private liquor retailers helping educate thepublic about the potential risks of alcohol use?Liquor retailers in Alberta in 1999 were asked toplay a vital role in a campaign aimed at combat-ing FAS. The Alberta Liquor Store Association, aprivate voluntary organization, sent individualstores pamphlets, decals, and posters warningpregnant women of the severe danger posed totheir unborn children by drinking alcohol. Theassociation president recognized the incentiveincompatibilities of this program:31

Greg Krischke, president of the association,says the government wanted the help of liq-uor retailers in raising awareness of FAS.Krischke says the association agreed it couldplay an important role in that. As respon-sible retailers and as an industry that is con-cerned about responsible consumption, wefeel that it’s important to assist in gettingthis message out to the public.

Krischke indicated he had not yet receivedany feedback from individual liquor retail-ers as to how supportive of the FAS cam-paign they will be, but that generally theyaccept initiatives supported by the associa-tion. He admits that some retailers may bereluctant to participate in a program thatcould result in reduced sales, but says mostliquor store owners see themselves as partof the community and want to see respon-sible consumption.

The program does not appear to be working.No literature relating to FAS, or for that matteranything that could be construed as fulfilling so-cially responsible objectives, was observed by theauthor and others in the many private stores vis-ited during this study.


Contrast this approach with the public liq-uor retailing organization in British Columbia,the BCLDB:32

The LDB continued to actively educate itsstaff and customers about the risks of drink-ing alcohol when pregnant. In addition toa yearly in-store campaign targeted at FetalAlcohol Syndrome (FAS) awareness, theLDB distributed brochures and posterscontaining valuable information about al-cohol and pregnancy that were previouslydeveloped in consultation with the B.C.FAS Society and endorsed by British Co-lumbia doctors, nurses and midwives.These materials have been made readilyavailable to customers, as well as to healthcare and community workers across theprovince and across the country. In 2001,the LDB distributed more than 5,000 FASbrochures and 400 posters.

There is another concern with the currentregulation and private retailing of liquor systemin Alberta: smuggling and illegal manufacturing.For example, beginning in 1994, Highwood Dis-tilleries, with the help of two other companies

and a former Canada Customs officer, obtained alicense to export liquor to the United States when,in fact, it was destined for the black market inCanada. “Convictions followed a 1995 RCMPsting involving two truckloads of whisky, rum,and vodka. Prosecutors showed that liquor des-tined for export was sold in Canada on the blackmarket. The product either never left the countryor was smuggled back. By avoiding excise dutyand sales taxes, the alcohol would sell for abouthalf the normal retail price.” 33

The extent of this problem is not as well docu-mented in Alberta as it is in other jurisdictions.34

Can the AGLC reasonably police this potentialproblem when Alberta has a fragmented distri-bution system with over 900 private retailers andanother 500-600 off-sales licensees? The vastmajority of retailers are honest business people.However, it is easier for alcohol smugglers andillegal manufacturers to sell to unscrupulous own-ers of private stores than it would ever be to apublicly-run retail network. The larger thenumber of independent stores there are, the moredifficult and costly is the effort to inspect, audit,and police this system.

Sobering Result 21

4 Alcohol Taxes

their taxation might also be considered inequita-ble. The equity case against such taxation has beencountered in the past with the argument that,because they are addictive, discouraging their useby taxing them is justifiable on moral grounds.”36

Today, moral arguments are generally less empha-sized, while health risks are more prominent; how-ever, there is still acceptance for these taxes.

4.1 Why Tax Alcoholic Beverages?

There are a number of arguments that can bemade to justify taxes on liquor. Higher prices re-strict consumption of alcohol, depending on theprice elasticity of demand [see box]. As there aresocial costs associated with the consumption ofalcoholic beverages, the lower quantity consumedunder taxation is appropriate and desired. Fromthis perspective, taxes on alcoholic beverages canbe considered a corrective Pigouvian tax.37 Whenthe social costs are added to the private costs, thedesired output and price (i.e., a high price andlow quantity) can be achieved with the tax. Thisconcept will be expanded upon in Section 5 whenoptimal prices are considered. However, it maybe difficult to estimate the social costs, or for thatmatter the social benefits, making the applicationof the appropriate tax complicated.38 Estimatesthat have been conducted indicate that the taxrevenue is far short of the social costs incurred.Tax revenue can (although not necessarily) be usedto correct or compensate for these social costs.For example, this tax revenue can help financehealth care in order to compensate for the ill-health effects of some alcoholic beverage use.

Taxes on alcohol are imposed by both federal andprovincial governments. There are several argu-ments to justify this type of excise tax. Liquor con-sumption generates social costs, liquor is a luxuryproduct, it has an inelastic demand, its consump-tion is complementary to leisure, and it brings inrevenue. A tax on liquor does two main things. Itincreases the price of alcoholic beverages, reduc-ing consumption, as well as any associated socialills. It brings in revenue to governments, and thetax revenue obtained can (potentially) be used tohelp correct or compensate for negative social ef-fects. Alberta government revenue from the saleof alcohol has fallen since the introduction of theunit tax (1993). The rates were decreased as theprivate marketing of alcohol led to price in-creases—increases which have been moderatedbecause the share taken as government revenuehas fallen.

The taxation of alcohol and tobacco prod-ucts is one of the oldest forms of taxation in theworld. In the review of the Excise Tax Act, it wasfound that: “All the countries surveyed continueto impose specific taxation on alcohol and tobaccoproducts at the level of the manufacturer, eitherat some point during the production process oron the first sale of the goods for wholesale distri-bution. This structure continues even in coun-tries with a greater reliance on a comprehensivevalue-added taxation structure.”35

Excise taxes (or duties) are taxes levied onspecific products, such as alcoholic beverages, to-bacco products, motive fuels (i.e., gasoline), andgaming. For one thing, consumer goods like to-bacco and alcohol are addictive and therefore theycan be made to yield large revenues. But becausethey are consumed widely by low-income groups,


A second argument for a liquor tax is thatalcoholic beverages are ‘luxuries,’ products notconsidered essential to living a full life, thereforealcohol taxes can be thought of as a luxury tax onthe more affluent in society. Of course, peopleacross the income spectrum purchase alcoholicbeverages, therefore this argument may be spuri-ous. However, it is an argument for higher ratesof tax (ad valorum) on pricier wines, spirits, andliquors.

A more esoteric efficiency case can be madefor taxing liquor at higher rates than other con-sumer goods due to the price inelasticity of de-mand for these goods. The Ramsey rule indicatesthat goods with a low price elasticity of demand(inelastic) be taxed at a higher rate for overall com-modity tax efficiency.39 A further refinement ofthis rule, the Corlett-Hague rule, indicates effi-cient commodity taxation also requires taxing ata higher rate commodities that are complemen-tary to leisure.40 Certainly the inelastic price elas-ticity of demand for alcoholic beverages guaran-tees the actual generation of revenues.

Governments tax alcohol because it brings inrevenue and they are historically accustomed totaxing alcohol products. Moreover consumers areaccustomed to paying these taxes, and there seemsto be general public acceptance of this form ofrevenue generation for government functions. Taxefficiency arguments, however, do support taxingliquor. There may be a policy tradeoff, though.Because the demand for alcoholic beverages isprice inelastic, it takes a proportionally larger priceincrease in order to curtail consumption. This situ-ation, however, brings in proportionally greaterrevenues per tax value than would occur withother, more price-responsive, products.

4.2 Unit Tax versus Ad Valorem Tax

Excise taxes can be either unit taxes or ad valoremtaxes. A unit tax is a certain charge on the prod-

uct per unit sold; for alcohol this charge is usu-ally per litre. An ad valorem tax is a percentagecharge on the value of the product, for example,a 10% tax on the product’s price. With an ad val-orem tax, the tax per product gets larger the higherits price. Both of these types of charges are usedas excise taxes in Canada.

An ad valorem tax causes a shift in the supplyprice by a constant percentage and has the advan-tage that it automatically rises with inflation. Thistype of tax also has a progressive tax effect if, as inthe case for alcohol beverages, the income elastic-ity of demand is positive and the price elasticity islow. What this means is that higher income earn-ers buy higher quality/priced alcoholic productsand, with an ad valorem tax, they will pay a highertax rate.

A unit tax causes the supply to shift by a con-stant amount. For this reason a unit tax has alsobeen referred to as a flat tax. The advantage of aunit tax is that the amount of tax is not affectedby the price of the product. A unit tax does notchange with price fluctuations, an advantagewhere prices are volatile. A unit tax on alcoholicbeverages will favour expensive products overcheap ones because the tax will comprise a smallerpercentage of the purchase price of the expensiveproduct. Tax incidence—who pays the tax—is alsoaffected. If the expensive product has lower priceelasticity of demand, the retailer can charge ahigher markup while paying a smaller percentagein tax.

4.3 Federal Tax System

Historically, commodity taxes on specific goodshave been an important element of the federaltax system, accounting for as much as 25% offederal revenues in the first half of this century.While their relative importance has declined overrecent years, these levies continue to represent avaluable source of revenue to the federal govern-

Sobering Result 23

ment. Among the most significant federal exciselevies are the charges imposed on alcohol and to-bacco products, which contributed approximately$3.1 billion in fiscal year 1995-96. Of this total,$1.94 billion was attributable to tobacco, $560million to beer, $480 million to spirits, and $110million to wine.41

Federal excise taxes are administered througha number of Acts. The Excise Act imposes exciseduties at various unit rates on domestically pro-duced spirits, beer, and tobacco products. Theoverall philosophy behind the Excise Act is one ofrigorous control and strict adherence to a set ofprescribed rules to ensure that revenue to the fed-eral government is maximized. The Excise Tax Actimposes excise taxes on both imported and do-mestic goods at a unit rate for wine and tobaccoproducts and at an ad valorem rate for cigars. TheCustoms Tariff imposes customs duties, equivalentto the excise duties that are applicable to domes-tically produced goods, on imported spirits, beer,and tobacco products. In addition, the Importa-tion of Intoxicating Liquors Act applies to bulk beerbrought into a province by an excise licensee.

The Excise Act imposes a tax on spirits at thefollowing rates: $11.066 for every litre of abso-lute ethyl alcohol distilled in Canada; $0.2459per litre on mixed beverages produced in a distill-ery that contain not more than 7% absolute ethylalcohol by volume; and 12 cents per litre on im-ported spirits, in addition to any of the dutiesotherwise imposed on every litre of absolute ethylalcohol.42 The Excise Act also imposes a tax onbeer: $27.985 per hecto-litre on all beer or maltliquor containing more than 2.5% absolute ethylalcohol by volume; $13.99 per hecto-litre on allbeer or malt liquor containing more than 1.2%absolute ethyl alcohol by volume but not morethan 2.5% absolute ethyl alcohol by volume; and$2.591 per hecto-litre on all beer or malt liquorcontaining not more than 1.2% absolute ethylalcohol by volume.

The Excise Tax Act imposes the following ex-cise taxes on wine: $0.0205 per litre on wines ofall kinds containing not more than 1.2% of abso-lute ethyl alcohol by volume; $0.2459 per litreon wines of all kinds containing more than 1.2%of absolute ethyl alcohol by volume but not morethan 7% of absolute ethyl alcohol by volume; and$0.5122 per litre on wines of all kinds containingmore than 7% of absolute ethyl alcohol by vol-ume.43

Although federal excise taxes on alcohol gen-erate significant income ($1,150 million in 1996),the predominant taxing authority on alcohol re-sides at the provincial level. Federal excise taxes,as well as the GST paid on alcohol products, be-come part of the cost base to the provinces. Theprovinces (except Alberta) mark up alcoholic bev-erage retail prices significantly in order to gener-ate ‘profits’—more appropriately called excise taxrevenues. When Alberta privatized, it set the mark-up on the wholesale price.

4.4 Provincial Tax Revenues

Prior to prohibition, the revenue from excise taxeson alcohol sales was a main source of governmentfinance. This was before the introduction of thepersonal and business income tax. Today, the rev-enue is of much less importance to provincial fi-nance, but is still significant. For example, in 1948alcohol tax revenue accounted on average for 17%of provincial revenues. But by 1970 it had fallento 5% and by 1981 to just 2.5%, where it hasmore or less remained.44

In Canada, alcohol administration and saleshave been provincial monopolies, and govern-ments therefore determined taxes implicitly bysetting retail prices. In most cases, an ad valoremtax is administered by setting a retail price basedon a percentage mark-up over cost. Under gov-ernment control of retail sales, the separation ofwholesale and retail prices is not an issue. Note in


Table 4.1 that, with Alberta’s pre-1993 ad val-orem taxes, the percentage tax increased with thealcohol content of the product. With the intro-duction of private alcoholic beverage retailing, Al-berta changed to a unit (flat) tax regime on alco-holic beverages. With the separation of wholesaleand retail prices in a privatized distribution sys-tem, the continued application of an ad valoremtax would have caused greater administrative prob-lems with constant fluctuations in wholesaleprices. The tax rates for Alberta under this unittax regime are shown in Table 4.1

The initial unit tax system was introduced inNovember 1993 as ALCB sold off stores to theprivate sector. As the private retail market devel-oped, prices started to rise and the Alberta gov-ernment reduced the flat tax rates in August 1994,phasing the reductions in through a surchargestarting at 10% and falling by 1% every four weeksover the year.

Table 4.2 repeats the data in Table 4.1, con-verting it to percentage change. This table showsthe percentage decrease in excise tax rates up un-til 2002. It is important to remember that thefigures in Table 4.1 are in current dollar figures.There was also some inflation. The tax rates werereduced in absolute terms and simultaneouslyinflation eroded the relative tax rates. The firstincrease in rates since privatization was made ef-fective on April 5, 2002. However, the tax onsmall-scale breweries products was again reduced.

It is worthwhile to compare Alberta’s tax rev-enue under the new unit tax regime to that ofBritish Columbia where liquor control is still simi-lar to what Alberta’s was prior to privatization.Alberta and British Columbia both had an advalorem alcoholic beverage tax system prior to1993 before the Alberta government shifted to aflat tax method simultaneously with the introduc-tion of privatization of the retailing of alcoholic

Table 4.1 Alberta Excise Tax RatesEffective Date: Pre

1993Nov1993

Aug1994*

May1995

Sept1997

Dec2000

Apr2002

Mar2003

Product Alcohol Content Percent $ Per LitreSpiritsa > 60% $17.87Spirits > or equal to 22.1% 149% $14.95 $12.95 $12.50 $13.30Spirits < or equal to 22% 159% $14.95 $12.95 $9.50 $9.90Ready to Drink and Cocktailsb $3.05 $4.05 $4.05Wine > or equal to 16.1% 177% $6.20 $5.50 $5.50 $6.10Wine < or equal to 16% 120% $4.35 $3.30 $3.05 $3.45Coolers/Cidersc 91% $2.10 $1.50 $1.25 $1.35 $1.35Beer 73% $1.06 $0.92 $0.89 $0.88Manufacturer produces worldwide annually (hectolitres)First 50,000 $0.50Next 20,000 $0.60Next 30,000 $0.75Over 100,000 $0.88> 200,000d $0.98< 200,000 $0.40< 10,000e $.20*in addition to these rates there was a plus 10% surcharge diminished by 1% per every 4 weeks Sources: A new Era in LiquorAdministration, ALCB, Dec. 1994. ALCB and AGLC Annual Reports. Review of Liquor Mark-up Structure and Related Findings andRecommendations, AGLC, Feb. 20, 2003.New as of March 21, 2003:aSpirits with alcohol content greater than 60%. The second tier becomes > 22% and < 60%.bThis category becomes refreshment beverage > 8% and < 16%.cThis category becomes refreshment beverage > 1% and < 8%.dApplies to more than 92% of the volume of beer sold.eNew category to lower the rate on small scale craft beer production.

Sobering Result 25

Table 4.2 Alberta Excise Tax Rates converted to percentage changeEffective Date: Nov

1993Aug1994

May1995

Sept1997

Dec2000

Apr2002

Mar2003

Product Alcohol ContentSpirits > 60% +34%Spirits > or equal to 22.1% -13.38% -3.47% +6.40%Spirits < or equal to 22% -13.38% -26.64% +4.21%Ready to Drink and Cocktails +32.79%Wine > or equal to 16.1% -11.29% 0 +10.91%Wine < or equal to 16% -24.14% -7.58% +13.11%Coolers/Ciders -28.57% -16.67% +8.00%Beer -13.21% -3.26% -1.12%Manufacturer produces worldwide annually (hectolitres)First 50,000 -43.18% -20.00%Next 20,000 -31.82% -33.33%Next 30,000 -14.77% -46.67%Over 100,000 0> 200,000 -20.0%—

46.67%< 200,000 +11.36%< 10,000 -50%

beverages in Alberta. Figure 4.1 illustrates alco-hol excise tax revenues as the percentage of totalprovincial revenues. At 3.02%, Alberta tax rev-enue in 1990 was a somewhat higher percentagethan the national average (2.5%), while BritishColumbia at 2.66% was closer to the national fig-ure.

In Alberta, liquor tax revenues declined to2.18% by 2002, while British Columbia revenues,although also falling somewhat, have experiencedgreater stability, remaining close to the Canadianaverage of 2.5%.

Figure 4.2 shows the same alcohol tax rev-enues in a different way. In this figure, revenuesare measured in constant dollars (1992) per capita.In the early 1990s, prior to privatization, Albertaobtained considerably higher per capita revenuefrom liquor sales compared to British Columbia.After privatization, the alcohol tax revenue in Al-berta began to decline and has become less thanBritish Columbia’s after 1997.

A third perspective to consider excise tax rev-enue on alcohol is to look at the revenue obtained(in current dollars) per unit of absolute alcohol

equivalent sold in each province. Absolute alco-hol converts the actual products sold into purealcohol equivalent. For example, a 750-millilitrebottle of spirits with 40% alcohol by volumewould constitute .40 X .750 = .3 litre of absolutealcohol equivalent. Figure 4.3 shows the revenuein current dollars realized on each litre of abso-lute alcohol sold. This figure shows how Alberta’srevenue has changed from being greater prior to1997 to less than that realized in British Colum-bia after 1997. Consumption in Alberta has risento 8.6 litres in 2001, while consumption in Brit-ish Columbia has fallen to 7.7 (Figure 2.1), yetthe dollar revenue per litre absolute alcohol hasbeen maintained and increased in British Colum-bia while it has fallen in Alberta.

Laxer et al asked in 1994: “Will it [the flattax regime] be as or more efficient than the previ-ous public system was in producing revenue forthe provincial government during its current pro-gram to balance the budget? In short, will priva-tization contribute to a more fiscally responsibleAlberta?” The answer to this question is clearlyNO. The revenue-neutral policy of the Alberta


government was ‘successful,’ liquor revenue wasconstant in absolute current dollars over the pe-riod. However, with inflation, growth in popula-tion, and growth in sales, a constant level of rev-enue is really a loss. Measured appropriately, thecurrent unit (flat) tax system in Alberta has re-sulted in lost revenue to the province from alco-hol beverages.

How much revenue might have been lost?This is a ‘what if ’ question that depends on theassumptions used. Table 4.3 shows the actual gov-

ernment revenue obtained from the sale of alco-holic beverages in Alberta (column B), and theper capita amount (column C), both in currentdollars. These amounts have not varied much eventhough there has been inflation and populationincreases. The table then shows the per capita rev-enue, if the tax revenue increased at the same rateas inflation (column D). Column E shows therevenue that would have been obtained with thepopulation increases. Finally, column F shows thedifference between the actual revenue obtained

Figure 4.1 Liquor Revenue as a Percentage ofTotal Provincial Government Revenue

1.5%

1.7%

1.9%

2.1%

2.3%

2.5%

2.7%

2.9%

3.1%

3.3%

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Year

Perc

en

t

Alberta

British Columbia

Figure 4.2 Alberta and BC Provincial Revenue Comparison

$100

$110

$120

$130

$140

$150

$160

$170

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Year

Co

nsta

nt

1992$ p

er

cap

ita

Alberta

British Columbia

Sobering Result 27

and what would have been obtained if the tax rateshad been maintained (in constant dollars).Thesefigures assume the same tax regime as in 1992 inplace throughout, that the inflation rate for liq-uor was the same as ‘all items’, and that the liquorconsumption per capita remained the same. Infact, the price increases were higher for alcoholproducts and per capita consumption has in-creased; however, these are likely the result of thechanges in the industry. The answer, given theseassumptions: The aggregated lost revenue between1993 and 2002 exceeds $500 million dollars.

Section 5 will review the experience of liquorprices in Alberta. If the new (1993/94) unit taxsystem had maintained relative tax revenue, whole-sale prices would be considerably higher todaythan they are. These lower taxes have dampenedthe upward pressures on wholesale price, offset-ting to a degree the rising retail prices. As shall beshown in the next section, however, consumershave been made no better off in terms of prices,while at the same time the government has lostrevenue.

Figure 4.3 Tax Revenue, Absolute Alcohol Sales

$18

$20

$22

$24

$26

$28

$30

1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Cu

rren

t$/p

er

litr

e

British Columbia

Alberta

Table 4.3 Liquor Revenue if 1992 revenue per capita were maintainedA B C D E F

Actual Potential DifferenceYear $millions current/capita per capita $millions $millions

1992 431 $163.61 $163.61 4311993 412 $154.27 $156.12 417 51994 434 $160.45 $164.62 445 111995 430 $156.94 $164.79 452 221996 484 $174.06 $186.77 519 351997 425 $149.80 $164.03 465 401998 425 $146.21 $161.85 470 451999 425 $143.60 $162.84 482 572000 469 $155.82 $182.93 551 822001 481 $157.24 $188.84 578 972002 482 $154.81 $192.27 599 117

Sum of lost revenue: 511


5 Liquor Prices

ing of alcoholic beverages is more complex. Lowprices for alcoholic beverages may reduce the pub-lic welfare through higher, if implicit, social costs.

5.1 Alcohol Price Elasticity

People buy more of any good the lower the price,all other things constant. This is also true for al-coholic beverages. However, the quantities of al-coholic beverages bought are relatively unrespon-sive to price changes, i.e., they are price-inelastic.

Johnson et al estimated short-run priceelasticities for beer, wine, and spirits to be .3, .9,and .5, respectively.45 This means that increasesin price will reduce the consumption of all bever-ages in the short run. Another way to put this is:a 1% increase in the price will reduce beer con-sumption by .3%, wine by .9%, and spirits by.5%. For example, if an additional 10% tax wasimposed on a $10 bottle of wine, the quantitysold (at $11) would decrease by 9%. They alsofound similar long-run elasticities for beer andwine, but found that spirits have very little long-run price-sensitivity.

These elasticities indicate that imposing taxeson alcoholic beverages is effective in generatingrevenue, and this is particularly true for spirits. Italso means that raising prices on beer and wine isan effective method in curtailing consumption,although for spirits, price increases would have tobe considerable in order to reduce consumption.

5.2 External Social Costs

As discussed above in Section 3, alcohol consump-tion generates social costs, referred to in econom-

Liquor prices in Alberta track the Canadian aver-age closely, but spike with privatization in 1993,and then level out as tax rates decreased. Whenliquor prices are compared with ‘all other goods’in Alberta, it appears that the unit tax system hasdampened the changes in liquor prices, until theytake off in 2001. Over the period of privatiza-tion, liquor prices have been more volatile in Al-berta than in the rest of Canada. And, over thelast decade, liquor prices have increased more inAlberta than in British Columbia. Current retailprices are similar in the two provinces, on aver-age, but show considerable variability betweenstores in Alberta. If the unit tax system had main-tained relative tax revenue, wholesale costs wouldbe considerably higher in Alberta today than theyare. These lower taxes have dampened the upwardpressures on wholesale price, limiting retail priceincreases. The interesting thing about liquor isthat a low price is not (or should not be) an ob-jective in the responsible control of alcohol con-sumption. Public welfare is higher when price ishigh, consumption is low, and revenues obtainedcompensate for the social costs.

The government implemented the privatiza-tion of the retailing arm of the ALCB using theargument that competition would bring downprices while the government obtained the samerevenue through the ‘provincial mark-up,’ (excisetaxes) on alcohol.

It is true that in general the lower the pricefor any consumer product, the better off individu-als and families will be, all other things constant.However, an objective of low prices for alcoholicbeverages is not obviously a good thing. Becauseof the relationship between alcohol consumptionand numerous social ills (and their costs), the pric-

Sobering Result 29

ics literature as negative externalities or externalcosts. The concept of elasticity has also been ap-plied to the relationship between changes in liq-uor prices and changes in the social ills associatedwith liquor use. Various studies have shown thespecific price elasticities of demand and the cor-relation with reduced alcohol-related problems.46

Studies in many different countries also show awell established link between the price elasticityof demand for alcohol and violence. These stud-ies show the effects of liquor prices on violence.For example, one international study shows a 1%increase in price decreased instances of robbery.19%, assault .25%, and sexual assault .16%.47

Similar to Health Canada’s estimates of so-cial costs in Canada cited previously (Section 3),the Alberta Alcohol and Drug Abuse Commis-sion (AADAC) estimated the social costs of alco-hol consumption in Alberta at $749 million in1992 (the most recent study). This study enu-merates an extensive list of the causes of socialcosts:48

Health Care: general and psychiatric hos-pital treatment; ambulance services; resi-dential, non-hospital, and ambulatory careincluding physician fees and other profes-sional services; prescription drugs; othercosts such as special rehabilitation equip-ment. Law enforcement: police; courts;corrections; probation; customs and excise.Other: damage resulting from fires and traf-fic accidents. Substance Abuse Prevention,Research and Training Administration forTransfer Payments: social welfare; worker’scompensation; health and life insurance;pensions; sick leave. Workplace: EmployeeAssistance Programs; other health promo-tion programs; drug testing. Substanceabuse also exacts a considerable toll in termsof morbidity and mortality. In 1992, theuse of alcohol, tobacco, and illicit drugsresulted in 3,092 deaths (21% of total

mortality for the province), 249,052 pa-tient days in Alberta hospitals, and an esti-mated 59,785 potential years of life lost.Tobacco accounted for more than half ofall deaths (2,344), hospital days (171,228),and potential years of life lost (35,531).Alcohol abuse resulted in 666 deaths,80,976 patient days in hospital, and 20,765potential years of life lost, and illicit druguse accounted for 82 deaths, 6,848 hospi-tal days, and 3,439 potential years of lifelost.

For these reasons, low prices for alcoholic bever-ages have not usually been a policy objective ofgovernments. Pricing of alcohol is a major policyinstrument in the control of liquor use. High re-tail prices serve two purposes: first, high pricesreduce use, and second, the considerable tax rev-enue obtained (the difference between retail pricesand wholesale costs) compensate for some of thedamages caused by alcohol consumption.

It is worthwhile to compare these estimatedsocial costs with the revenue obtained (in 1992)through taxing alcoholic beverages. The estimatedsocial costs in Alberta equaled $749 million, whileonly $431 million was obtained in provincial al-cohol tax revenue (profit). Clearly, in 1992 liq-uor taxes were already insufficient, according tothe ADDAC estimates.

5.3 Optimal Pricing with ExternalSocial Costs

Figure 5.1 illustrates the basic theory of the im-plications when an external social cost exists inthe consumption of some product. The demandor marginal private benefit (MPB) curve showsthe benefits to individuals measured by their will-ingness to pay. The supply or marginal privatecosts (MPC) of producing the product indicate


the quantities that would be produced at alterna-tive prices. The private market equilibrium quan-tity and price is shown as Qe, Pe. When the ex-ternal costs (MEC) are added to private costs(MPC), the marginal social costs (MSC) relation-ship is created. The optimal quantity and price inthis situation is shown as Q*, P*. This analysisillustrates that, when there are external social costsin the private consumption of a product, the op-timal price (P*) is higher than the market price(Pe), and the welfare-maximizing optimal quan-tity (Q*) produced and consumed is less than thequantity (Qe) a free market would achieve.

Low prices for alcoholic beverages are not (orshould not be) an objective in the responsiblecontrol of alcohol consumption. Public welfare ishigher when price is high, consumption is low,and revenues obtained are used for compensationof the social costs. However, this does not meanthat an inefficient distribution and retailing sys-tem, with high costs, is desirable. The governmentshould maximize alcohol tax revenues subject toany pricing policy. From the point of view of Fig-ure 5.1, a higher price and lower consumption ofthe product is desired because of external costs.

The retail distribution (supply) costs (MPC)should still be kept as low as possible so that, allthings constant, the tax rate to achieve the opti-mal output and price, Q*, P* is as high as possi-ble. This issue will be considered more fully inthe section on efficiency.

5.4 Price History

The advent of the privatization of liquor retail-ing, coupled with the changes in the way excisetaxes on liquor are assessed, has created quite vari-able pricing of alcoholic beverages in Alberta. Thissection will review the price changes over time.

As with most products, Statistics Canada sur-veys prices on alcoholic beverages on a continu-ous basis, using the consumer basket determinedby the Family Expenditures Survey. Using thechange in the consumer price index (CPI) for al-coholic beverages, Figure 5.2 shows the inflationrates for alcoholic beverages for all of Canada andfor Alberta specifically. As would be expected,these two series track very closely. The exceptionoccurs around the time of privatization (1994)where initial price increases in retail prices spiked

Figure 5.1 The Effect of an External Cost

on the Private Market for a Product

Sobering Result 31

and then fell in 1995. Retail prices were offset in1995 by a decrease in the wholesale price througha reduction of the excise tax rates (see Table 4.1).

Figure 5.3 uses the CPI for alcoholic bever-ages and ‘all other goods’—the CPI for all itemsexcluding alcoholic beverages. This figure com-pares the rise in prices in Alberta calculated asinflation rates—changes in the CPI per year. In

the late 1980s, there was considerably higher in-flation in alcohol than in other products. Since1990 the two rates have tracked fairly closely. Thespike for alcoholic beverages in 1994 upon priva-tization and the drop in the rate in 1995 are notreflected in all other goods. This supports the viewthat privatization initially increased retail prices,which were then offset by a decrease in wholesale

Figure 5.2 Alcohol Beverage Inflation Rates

-2%

0%

2%

4%

6%

8%

10%

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Year

Infl

ati

on

Rate

Alberta- Alcohol

Canada-Alcohol

Figure 5.3 Alberta Inflation Rates 'Alcoholic Beverages' and 'All Other Products

-2%

0%

2%

4%

6%

8%

10%

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Year

Infl

ati

on

Ra

te

AB Alcohol

AB all other


prices, due largely to a drop in the tax rates. Theliquor industry appears to have then absorbed thetax rate decrease and again tracks with the ‘all oth-ers goods’ inflation rate.

In 2001, however, there was a considerableincrease in retail prices for liquor in Alberta abovethat of all other goods. This rise is prior to the taxhikes initiated in April of 2002 (Table 4.1). Thesharp increase in the 2002 year end data reflectsthe increases in the tax on alcohol implementedin April. As this tax increase will be included inthe retail prices over time, we are likely to see asignificant rise in prices again in 2003.

In order to observe the volatility of alcoholicbeverage retail prices, the difference between theCPI for alcohol and for ‘all other goods’ for bothCanada (including Alberta) and Alberta is de-picted in Figure 5.4. If there was no differencebetween alcohol and other goods, then the serieswould be coincident with the zero axis. ForCanada the difference is slight. Alberta, however,shows considerable variation in price changes foralcoholic beverages over all other goods. The dataillustrate that in Alberta after privatization (1994)liquor prices rose dramatically well in excess of

other goods, and then the rate came down gradu-ally as the new flat rate tax was reduced, both inabsolute and relative terms. Proportionally, theprice increases of alcoholic beverages were mod-erated by the flat tax which reduced the relativetax revenues to the government, with its offset-ting effect on the rising retail prices of liquor. Fig-ure 5.4 also shows that retail liquor prices in Al-berta have taken off above the national average in2001, and this is prior to the flat tax rate increasein 2002.

5.5 Current Prices

The previous section showed the effect of privati-zation on the changes in alcohol beverage pricescompared to all other goods using StatisticsCanada data. A previous study (West) comparingprices over time found similar results in the vola-tility of prices. Rates of change in prices over time,however, do not tell us the current state of retailliquor prices in Alberta. Nor do they tell us thevariability of prices over geographical space. Manycomparison price surveys are being conducted tomake one point or another. More rigorous work

Figure 5.4 The Difference in CPI between Alcohol Beverages and All Other Items

-1

0

1

2

3

4

5

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Year

CP

I D

iffe

ren

ce

Alberta

Canada

Sobering Result 33

surely needs to be done if these comparisons areto be of any use and are to be taken seriously.

Taken together, most of these comparisons—for example, between Alberta and British Colum-bia—find that some products are lower in pricein one jurisdiction while others are more expen-sive. This result is likely caused by the differenttax methods and the prevalence of variable pric-ing in Alberta, including sale prices on specificproducts. Alberta has a flat or unit tax system,while British Columbia has an ad valorum (orpercentage of cost) tax. This difference will auto-matically increase the price of higher-cost prod-ucts in the British Columbia market over Alberta.For example, beer, which is a low cost productper volume of alcohol, is generally cheaper in Brit-ish Columbia than in Alberta, while spirits andhigher-cost wines are more expensive in BritishColumbia. An anomaly sometimes occurs withliqueurs, where in Alberta the tax is lower, butthe price is often higher (because of a larger per-centage markup by the retailer). This is likely dueto the relative price inelasticity of demand. Sim-ply put: the retailer can mark up the price moreon such items, as the price does not affect the

quantity bought as significantly as for, say, beer.Sometimes a sale price is below the wholesale costand is used as a loss leader to entice customers,where their overall expenditure once in the storemay be greater than the cost of the same productsif they were bought from another retailer at regu-lar prices.

A comparison of price changes between Al-berta and British Columbia is illustrated in Fig-ure 5.5, which tracks the CPI (1992 = 100). Pricesin Alberta were traditionally lower than those inBritish Columbia in the early 1990s. The figureshows that, after retail privatization and the changeto a flat tax in Alberta, price increases for alco-holic beverages sold in stores exceeded those inBritish Columbia. What sold for $10 in 1992 ineach province cost $13.26 in Alberta and $11.53in British Columbia in 2002. Tax rate reductionsin Alberta have not been sufficient to keep priceincreases below those in British Columbia. Thereasons for this will be discussed in Section 6.

A small price study was conducted, initiallyto test the hypothesis that prices rise as one movesout from the urban core to the rural periphery. Amodest basket of products was selected, one that

Figure 5.5 CPI, Alcohol Beverages Sold in Stores Comparison of Alberta and British Columbia 1992=100

100

105

110

115

120

125

130

135

140

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Year

Co

ns

um

er

Pri

ce

In

de

x

AB Alcohol

BC Alcohol


would represent an average consumer’s purchases.This basket included beer (one dozen each of do-mestic high-volume national producer, and me-dium- volume Alberta producer); spirits (one 750-ml bottle each of rum, whiskey, and vodka; wine;one 750-ml bottle each of French red, Germanwhite, USA white, and Australian red; and one750-ml bottle of liqueur. The results of this sur-vey are shown in Table 5.1.

The survey did not support the hypothesisthat prices are higher as one moves out from theurban centre. But it did show a considerable vari-ability in product prices. The average (mean) costof the basket was $184.92 in Alberta. For com-parison, the cost of the same basket in BritishColumbia was $183.93, an insignificant differ-ence.49 There was greater variability within theurban centre, primarily due to transitory sale priceson some products coincident with the time of thesurvey.

This small survey suggests that there is con-siderable variability and volatility in prices in Al-berta. The total basket cost had a considerablerange from a low of $170.01 and a high of$201.72. The variance in individual productprices is also considerable (as measured by the co-efficient of variation). Beer, with 50% of dollarsales volume, has the least variance in prices acrossstores. Product knowledge and tastes are much

more likely to be certain for beer. Spirits hadgreater variance in prices than beer, but againproduct knowledge and experience is high. It iswine, however, where the large variances show up,likely due to the lower product knowledge on thepart of consumers.

There was no correlation between basket costand distance from the urban core, nor was thereany correlation with the basket cost and thenumber of close competitor stores. In the ruralareas, although there may be less competition,some costs are lower: for example, lower retailspace rents. Also, the pressure exerted on propri-etors who are local residents to be perceived ashaving ‘fair’ prices offsets the opportunity ofcharging higher prices.

Prices vary considerably more in Calgary thanin rural communities. Price competition is strong-est in Calgary, where discount stores and sales onparticular items can mean that prices vary greatly.At the same time, a consumer must be highlyaware and willing to incur costs in time if he orshe is to achieve any benefit from this price com-petition. As well, the advantages of store prolif-eration are lost if one has to cross town in orderto get the best price.

From these statistics in Table 5.1, we can geta sense of the price range and variability in Al-berta. The Pearson’s coefficient of skewness in

Table 5.1 Summary Statistics on Price of Ten Alcohol Beverage Products Surveyed January 2003

Product AlbertaWholesale Mean Price

StandardDeviation

Coefficientof

Variation

Pearson’sCoefficient of

Skewness

BritishColumbia

Price 1 $16.29 $18.78 $0.58 3.07% -0.1033 $17.80Beer 2 $16.55 $19.12 $0.62 3.23% 0.1066 $21.10 3 $19.48 $23.20 $1.12 4.81% -0.6011 $20.50 4 $19.53 $23.27 $1.01 4.35% -0.2671 $20.50Spirits 5 $22.89 $27.03 $1.53 5.65% 0.0732 $26.00 6 $13.60 $15.07 $1.71 11.35% 0.0429 $18.00 7 $7.11 $9.17 $0.91 9.88% 0.7150 $10.00 8 $9.66 $11.24 $1.03 9.19% -1.9238 $14.99

Wine

9 $7.12 $10.06 $1.16 11.53% 0.0112 $9.04Liqueur 10 $23.10 $27.98 $2.08 7.42% 0.4535 $26.00Total Basket Cost $155.33 $184.92 $183.93

Sobering Result 35

Table 5.1 indicates the degree to which the pricedistribution is not symmetrical. A positive numberindicates skewness to the right (more prices abovethe mean than below), and a negative numbermeans skewness to the left (more prices below themean than above). As a general rule of thumb, ifthe Pearson’s coefficient is less than .5, the distri-bution is symmetrical. Only product 8 has a high(negative) skewness. This suggests that the pricedifferences for the most part are normally distrib-uted: that is, most prices (~95%) fall within onestandard deviation of the mean. For example,product 1 has a mean price of $18.78 and 95%of stores sold at a price between $18.20 and$19.36. This compares to the BCLDB price of$17.80.

The survey shows that prices, on average, arenot dissimilar in the two provinces of Alberta andBritish Columbia. The Alberta Liquor Store As-sociation acknowledges this: “[The association]

feels neither the association nor Commission canany longer claim the province has the most com-petitive liquor prices, aside from those of high-end unique brands.” 50 Whereas British Colum-bia has the same prices throughout the province,set by the British Columbia Liquor DistributionBranch (BCLDB), Alberta retailers are free to setprices at whatever level they choose. This free-dom and the types of competition that haveevolved in Alberta lead to considerable variationin prices between products and stores. This pricevariability can create an opportunity for motivatedconsumers to obtain considerable price advantageif they are prepared to incur the costs of researchand travel. For the majority, it is more likely tocause considerable consumer confusion, and mostpeople will frequent a specific store on the basisof some other factor, such as the convenience ofthe location.


6 The Evolving Market for Alcohol

erages, usually measured in terms of outlets percapita or outlets per square kilometer. Availabil-ity also depends on hours of operation and daysopen. In Alberta, retail liquor stores, includingoff-sales, can remain open until 2:00 a.m. 364days of the year.

Table 6.1 compares the total government re-tail liquor and agency stores for each province andterritory. The data for Alberta shows private liq-uor and general merchandise stores, but does notinclude approximately 600 off-sales licensees. Aswell, the table does not include private stores inprovinces other than Alberta. For example, TheBeer Store is the primary distribution and saleschannel for beer in Ontario, with 433 retail storesin 2002. And in British Columbia in addition tothe 368 government and agency stores there are290 private cold beer and wine stores (now fullylicensed for all liquor, including spirits), and 110private manufacturer /VDQ stores (2002), and26 consignment agency stores.52

The advent of privatization has dramaticallychanged the market structure of liquor retailingin Alberta. Prior to the privatization of liquor re-tailing initiated in 1993/94, Alberta had 310 to-tal retail stores. Subsequently, AGLC has licensedany business that meets minimal conditions andhas expectations of local authority approval. ByJanuary 2003, there were 983 retail stores.51 Whyso many stores now, and what is the consequence?There are two reasons for the large increase in thenumber of private retailers of alcoholic beveragesunder privatization. Entrants can open a smallstore with minimal capital costs and easy licens-ing—easy entry; and these entrepreneurs believethat liquor is an easy industry in which to makeprofit.

6.1 Physical Availability

The physical availability of alcohol refers to theprevalence of retail outlets that sell alcoholic bev-

Table 6.1 Total Government and Agency Stores (Alberta private)Year

Province or Territory1995/96 1996/97 1997/98 1998/99 1999/00 2000/01

Newfoundland and Labrador 96 95 100 103 108 109Prince Edward Island 17 19 19 19 19 19Nova Scotia 101 102 101 101 102 102New Brunswick 106 120 118 118 117 118Quebec 494 497 491 495 467 622Ontario 682 685 693 702 709 708Manitoba 225 225 222 219 220 220Saskatchewan 271 271 270 272 268 268Alberta 637 674 807 826 826 863British Columbia 361 363 371 377 389 388Yukon 7 7 7 7 7 7Northwest Territories 6 6 6 6 6 6Nunavut 0 0Canada (total), excluding AB 2366 2390 2398 2419 2412 2567

Sobering Result 37

Figure 6.1 illustrates the proliferation of re-tail stores in Alberta compared to Ontario, Brit-ish Columbia, and the Canadian average storeson a per capita basis. In 2000-2001, Alberta hadalmost three times the national average of storesper capita and more than a third greater than thenumber in British Columbia. The number ofstores in Alberta continues to climb.

A major reason for the social concern aboutthe unrestricted availability of alcoholic productsis the fact that, other things being equal, increasedaccess to alcohol will increase alcohol abuse bythe population groups at risk, even if the level ofconsumption by the majority of the populationremains more or less unaffected. Just who reallyshops for liquor at 2:00 a.m., anyway?

Notwithstanding the previous macro perspec-tive on crime rates discussed in Section 3, Laxeret al documented many empirical studies that gen-erally support the hypothesis that unrestricted freemarkets increase alcohol consumption while pub-lic monopolies restrict consumption. For exam-ple, Gruenewald and Millar have documented theresearch linking the availability of liquor prod-ucts to consumption and consumption to crime,

violence, and automobile crashes in Americanjurisdictions.53 “Current research suggests that dif-ferent degrees in physical availability are directlyrelated to differences in rates of alcohol consump-tion and related problems.”54

A major objective of the creation of ALCB in1924 was to control the sale of alcoholic bever-ages. ALCB directed its efforts over the years toserving Albertans in a socially responsible way. Isit possible for a private market to go against itsown self-interest in maximizing sales and profitand limit sales for socially responsible reasons?The concept of availability should also includethe willingness to sell to minors and intoxicatedpersons even when these sales are illegal but inthe economic interest of the retailer.

6.2 Market Structure or IndustrialOrganization

Economists study a given industry using theoreti-cal models to guide their analysis.55 These modelsrange along a continuum, with monopoly at oneend, through various forms of imperfect compe-

Figure 6.1 Liquor Outlets (Ontario and British Columbia adjusted for private outlets)

0

5

10

15

20

25

30

35

1995/96 1996/97 1997/98 1998/99 1999/2000 2000/01 2001/02

Year

Liq

uo

r S

tore

s/1

00,0

00 p

op

ula

tio

n

Alberta

British Columbia

Canada

Ontario


tition, including oligopoly (a few firms) and mo-nopolistic competition, to the other end of thespectrum with perfect competition. Monopoly isa single seller with complete control of the mar-ket and barriers to prevent others from enteringthe market. The other extreme, perfect competi-tion, includes a large number of firms, easy entryinto the industry, and (importantly) each firmprovides a very similar (identical) product. Per-fect competition results in the situation where eachfirm is a price taker and has no market power.Real world situations generally fall into the mid-range of this continuum where varying degreesof market power and competition exists.

Monopoly can occur ‘naturally’ if there arelarge economies of scale or through multiple-out-put production (economies of scope) such thatonly one producer (seller) could achieve the low-est possible cost of production given the size ofthe market. Otherwise, a monopoly can only ex-ist through government protection or ownership,or illegal anti-competitive behaviour. The eco-nomics literature explains the inherent inefficiencywhen a private monopoly maximizes its profitthrough its ability to set price above cost, and re-stricts the quantities of the product on the mar-ket. The society incurs a welfare loss because lessis produced and a higher price is charged thanresource scarcity would require. This clearly canhappen only when the monopoly is private andthe firm’s objective is profit maximization. Publicownership of the monopoly, or regulation of aprivate natural monopoly, changes the firm’s ob-jective from one of profit maximization to thepublic objective, often output maximization witha price equal to cost (including a normal profit)where welfare is maximized.56

The reason for public control through nearmonopoly ownership of liquor retail outlets wasneither to maximize profit nor was it essentiallyfor consumer convenience. And the objective wasnever to maximize output at the lowest possibleprice such as was the case for telephone service

(Alberta Government telephones) when it waspublicly-owned. Public ownership of liquor re-tailing was put in place to limit the distributionof liquor in a reasonable and socially responsiblemanner and to ensure the implementation andcollection of excise tax on alcoholic beverages.However, the public retailing of liquor through amonopoly was able to take advantage of anyeconomies of scale available, and was able to planfor the number, size, location, and opening datesand times of retail stores in a socially responsiblemanner.

It appears that the Alberta government im-plemented liquor retail privatization on the ideo-logical belief in competition: Adam Smith’s ide-alized state of ‘invisible hand’ perfect competi-tion where a large number of firms producing es-sentially the identical product with no cost ad-vantage compete on price, and the market real-izes a price equal to the lowest cost of produc-tion. The government also assumed the privatesector, through a competitive market, could meetconsumer desires and respond better to more lo-calized market conditions, while also achievingthe lowest retail prices. There are two things tonote about this. First, this is a major change inattitude toward liquor as a product—treating asif it were just another innocuous consumer good.Second, it assumes that a competitive (read: per-fectly competitive) industry would result wherethe lowest cost of production (retailing cost)would be achieved and prices would fall to thiscost.

6.3 The Alberta Liquor Retail MarketStructure

Liquor retailing in Alberta is currently not a per-fectly competitive industry. It is a‘monopolistically competitive’ industry, one inwhich there are a large number of firms, as inperfect competition, but each firm produces a

Sobering Result 39

product that is differentiated from that producedby its competitors. Many other service industriesare also examples of monopolistic competition.The current structure is still evolving and the in-dustry is experiencing considerable change as theeconomies of scale are exploited and chains get afirmer hold on the market. As chains develop, themarket will move towards oligopoly where a fewlarge firms compete. At present, the restrictionthat wholesale prices, and particularly delivery andtransportation costs, be uniform to all vendorswherever they are located has limited the degreeof economies being realized.

The number of firms that make up amonopolistically competitive industry can be veryhigh, as it is relatively easy to enter and exit theindustry. As a result, no one of them is largeenough to dominate the market. In the liquorretail business in Alberta, there are over 1,580 classD licensees, with over 900 retail stores. As an in-dication of the size of the typical firm, there is anaverage of about four workers per store, with av-erage annual sales less than $1 million each. Thismeans the average outlet accounts for only about1/1000th of total industry sales.

6.4 Product Differentiation

Unlike perfect competition, in which firms pro-duce an identical product, in a monopolisticallycompetitive industry each firm’s product is slightlydifferent from every other’s, therefore each firmhas limited power as a price setter, and firms en-gage in considerable non-price competition. Theconsiderable variability of liquor prices in Alberta,the increasing differentiation of stores and prod-ucts, and the increasing prevalence of advertis-ing, all support the view of the industry asmonopolistically competitive.

The key characteristic in a monopolistic com-petitive industry is product differentiation. Dif-ferentiation occurs in a number of ways. The pre-

dominant differentiation of retail outlets is loca-tion. Convenience of location is arguably the mostimportant difference. For example, the proxim-ity to a major grocery store has always been a greatadvantage. The grocery chains themselves aremoving quickly to exploit this advantage and willlikely come to dominate the market. Other im-portant characteristics create differences betweenoutlets.

There are now over 17,000 SKUs on thewholesale list. Even the largest of retail stores canstock only a fraction of these products. Thisplethora of products creates an opportunity forconsumer confusion. Just the particular selectionfrom this number of products differentiates thestore. Some stores stock as few as 400 items. Asall licensees must pay cash for their stock, theyare therefore motivated to keep the items on theshelves restricted to those that turn over frequently.

Store decor has become a point of differen-tiation. Some stores present a professional attrac-tive environment. Obviously this involves costs.Others, such as deep discount stores, keep theirstore appearance and decor costs to a minimum,presenting a rather shabby public front. There arestores that have invested in obtaining consider-able expertise in wines and differentiate themselvesby offering wine-tasting events, and courses onwine selection. Some operators distinguish them-selves by offering delivery service. Opening timewas extended to 2:00 a.m. but most stores find ittoo costly to remain open this late (early). How-ever, some stores differentiate by staying open rightup to this mandated closing time. Other storeshave frequent-user discounts. Some have joinedthe Preferred Alberta Liquor Stores (PALS) pro-gram which offers air miles with purchases. Thereare currently 108 stores enrolled in this popularreward program. The Calgary Cooperative Asso-ciation extends its member benefits to liquor storepurchases, as does Safeway.

The Real Canadian Liquor Store also differ-entiates by having a lower price per unit if four or


six units are purchased at one time. A small storeowner says in frustration: “Certain customerscome in to buy their regular six-pack, but whenthey have a party they go to the Real CanadianLiquor Store to get the discounts. I am not goingto stay in business just for this convenience.”

Product differentiation is clearly the most im-portant difference between monopolistic compe-tition and perfect competition. What are the con-sequences?

Product differentiation means an individualliquor store faces a downward sloping demand.Since each firm’s product is a close, but not per-fect, substitute for every other firm’s product, de-mand is quite elastic, but not perfectly so. This isin contrast to a purely competitive market whereeach firm produces essentially the same productand therefore faces the same market price. Thischaracteristic is the ‘monopolistic’ aspect of mo-nopolistic competition and, coupled with the lowbarriers to entry into the industry, determines theparticular outcomes of this form of market struc-ture.

In a monopolistically competitive industry,firms have the same kind of freedom to enter intoand exit from the industry as they do in a per-fectly competitive industry. In Alberta, potentialliquor store operators face little difficulty in ob-taining a license. Licenses are not transferable, norcan they be sold, so that no economic rent residesin holding a license, as is the case, for example, intaxi licenses. Any limit on the number of stores,location, and size has been left to the municipalgovernments. Freedom of entry and exit meansthat there are no barriers to competition in amonopolistically competitive industry. This makesfor keen non-price competition among firms andensures that in the long run no monopolisticallycompetitive firm is able to make a greater thannormal profit.

Since each firm produces a similar but some-what different product compared to every otherfirm in the industry, there is an incentive for each

firm to play up the difference in its product inorder to boost its sales. This non-price competi-tion takes many forms. Anything that may serveto distinguish a firm’s product from that of itscompetitors in this way might be tried, so long asthe firm feels that the cost of such promotionalactivity is more than made up for by the resultingincrease in sales. Advertising of liquor in localnewspapers, shoppers’ guides, the Internet, andother media outlets is becoming common. A par-ticularly insidious form of this is the u-haul signat the side of the roadway beckoning you to comein and try this or buy that. Lately the CalgaryHerald has displayed a person wearing a large beercan (Budweiser) for a head attempting to attractcustomers to get a discount on beer. This non-price competition raises costs.

The wide variability in prices now found inAlberta demonstrates the degree of differentiation.For instance, the more a firm’s product is similarto that of other firms in the industry, the morewilling consumers will be to switch to that firm’sproduct if it lowers its price, and away from thatfirm’s product if it raises its price. Thus, the greaterthe degree of similarity, and therefore substitut-ability, between a firm’s product and those of itscompetitors, the greater will be the change in itssales resulting from any given change in price. Inthe extreme case of perfect substitutability, thefirm’s demand curve is perfectly horizontal, or in-finitely elastic, and we are in a perfectly competi-tive environment. Alternatively, the more prod-uct differentiation there is among firms in theindustry, the less elastic will be the individual firm’sdemand curve. Convenience of location as themost important differentiation means, for exam-ple, that one will be willing to pay a higher pricethan he or she might be able to obtain, if the lo-cation of the store is handy. A monopolisticallycompetitive firm is able to affect its level of salesby changing its price; it is in this respect similarto a monopoly.

Sobering Result 41

6.5 Excess Capacity—Too ManyRetailers

Monopolistic competition creates an environmentallowing for an inefficient number of outlets,where each firm has higher costs than necessary,but most still manage to obtain sufficient revenueto cover the costs of operation (barely) and stayin business. As one owner of a small ‘mom andpop’ operation in Calgary stated: “I hope to makeit through this year, but it’s not easy.” Excess ca-pacity is the problem with a monopolistically com-petitive industry. There are far too many outletsretailing liquor in Alberta, with far too little traf-fic on average. This excess capacity pushes up unitsales costs well beyond what would occur in anefficiently run monopoly or oligopoly marketstructure. As another store owner expressed it, “Ilike to work hard when I go to work. Here I cansit for hours before one customer comes in thedoor.” This excess capacity creates higher costs.The costs of non-price competition and idlenessoccur even though wages for hired staff have fallenwell below one-half the wage and benefits foundpreviously in the unionized ALCB. The costs ofexcess capacity are keeping retail prices relativelyhigh even as the percentage return to governmentin taxes is falling.

The welfare consequences of this are not with-out controversy. There is a trade-off going on withmonopolistic competitive markets: variety in themarket with excess capacity and higher prices(costs) versus more standardization and lowerprices (costs). Chamberlain and Robinson areboth attributed as founders of the theory of mo-nopolistic competition. Chamberlain took a posi-tive view to the diversity of products and oppor-tunities for choice that monopolistic competitioncreated.57 Robinson, in contrast, stressed the wasteand unnecessarily higher costs incurred by theexcess capacity.58 Whatever one’s point of view asa consumer about the benefits, there is no ques-

tion the costs are considerably higher. It appearsthat the government has wanted low retail pricesand has been willing to reduce the tax rates inorder to accomplish this, thus masking some ofthe inherent higher costs of the market structurethat have resulted since privatizing retail liquorsales.

6.6 The Growth of Chain Stores

The market might achieve greater efficiencies overtime. A current restraint on efficiency from amarket perspective is the wholesale cost and trans-portation structure. As well, the current require-ment that all entrants create stand-alone facilitiesprevents grocery chains from achieving the costadvantages of offering liquor products in theircurrent retail space. This has limited the oppor-tunity for firms to realize some cost advantagesthat might be attained by size or location. Thishandicaps the industry from achieving greaterconcentration, a move toward a more oligopolisticmarket structure. There are, however, some econo-mies of scale to be had and the chains are begin-ning to realize them. These economies includeadministrative advantages, advertising, and mini-warehousing.

The same administrative functions can beconducted for a large range of output (sales).Tracking, ordering, invoicing, paying the bills,etc., all have to be done whether you have onesmall (400-item) store or 10 large stores. Thesefunctions cost much less per unit for the largeroperation. When these functions are integratedinto an existing grocery store with much the sameadministrative tasks, the cost per unit is reducedeven more. Advertising, especially in the large dailynewspapers, is expensive. Small stores cannot af-ford to do it. The larger the chain, the more itcan spread the costs of this advertising over all ofits outlets as long as the advertised products andprices are the same at each location.


Mini-warehousing—i.e., using one large storeto feed other smaller operations—is one way toreduce costs. For example, this allows for case-lotpurchases of an expensive product and splittingout single bottles in order to expand offerings ineach particular store in a chain. A single store can-not often afford to purchase and stock a case (theminimum) of expensive products.

Laxer et al predicted in 1994 that the indus-try would likely evolve in this way: “Specifically,the current system of small, independent retailersis temporary, and that alcohol retailing in Albertawill in future be conducted by groceries and otherlarge chains.”59 Table 6.1 shows the current stateof chains in Alberta. At present, chains constituteabout 13.5% of all stores, but operate primarilyin the urban centres where they are a larger per-centage of the local market. This trend is just be-ginning to take on steam. The Calgary Coopera-tive Association has been expanding into liquorretail quickly. Safeway has only recently enteredthe market and will surely expand. IGA stores arenow considering entering the industry and havemade offers to existing liquor stores near theirstores. Willow Park has recently purchased at leasttwo other competitors and will be converting these

to their chain soon. As Willow Park is adding twostories to its flagship store for administrative of-fices, we can expect an ambitious plan of expan-sion.

6.7 Wholesale Costs

Increased prices and reduced government tax rev-enue are not due solely to the cost inefficienciesfound in the resulting monopolistic competitiveretail industrial structure. A partial explanationof these results is found in increases in wholesaleprices. These increases are due to both the in-creased costs of the legislative/enforcementmethod of regulation and to the increased landedcosts of the manufacturers.

If Alberta had an efficient distribution sys-tem, it should be able to obtain approximatelythe same percentage return on net sales as doesBritish Columbia. Table 6.2 shows the operatingaccounts for Alberta and British Columbia for theyear ending March 31, 2002. Alberta received a35.7% return from sales, while British Columbiareceived 40.8%. This latter amount does notcount the social services tax revenue from liquorsales in British Columbia where final prices in-

Table 6.2 The number and percent of chain storesCompany Number Percent of storesCalgary Co-Operative Association 11 1.21%Greyhound Canada 4 0.44%Liquor Barn 3 0.33%Liquor Depot 23 2.53%Liquor World 15 1.65%Lucky Liquor Store 4 0.44%O K Liquor Store 6 0.66%Olympia Liquor Cold Beer and Wine 3 0.33%Patterson Liquor Stores 3 0.33%Royal Liquor Merchants 3 0.33%Safeway Liquor Store 5 0.55%Spirits of Belmont 9 0.99%The Real Canadian Liquorstore 19 2.09%Tops Liquor 3 0.33%Willow Park 12 1.32%Total 123 13.52%

Sobering Result 43

clude the social services tax (PST). If we addedthis revenue to the British Columbia government’stake, we would have to compare Alberta’s 35.7%to 47.5% in British Columbia.

In the years prior to privatization, the ALCB(when the retailing of alcoholic beverages was stilla part of their operations) obtained the same per-centage returns as the BCLDB has in 2002. TheALCB reported returns for 1990 through 1993,at 40.52%, 40.85%, and 41.68%, respectively.60

The 2002 wholesale prices under AGLC mustinclude some hefty operating costs that did notexist in the former ALCB. For example, the costin British Columbia of selling $1.5 billion at thewholesale level was $778 million, or 52%. Albertaspent $887 million to sell $1.38 billion, or 64%.The AGLC has considerably higher costs per $1of sales. The AGLC has to cover the costs of in-spectors and auditors which, compared to a pub-lic distribution system, are higher due to thegreater number of retail stores and the greatervariability in quality of their operations comparedto a publicly-run system. It appears that, had thegovernment of Alberta retained the retail compo-nent through the ALCB, distribution would notcost much more than it does now under AGCL(without a retail component).

In 1992, the ALCB’s total operating expense(in current dollars) was $83,451,000. This figure

included wages and benefits of $53,434,000. Ifthe total operating expense in 1992 is correctedfor inflation and population increases, and withan appropriate expansion in stores, it would havecost $125 million to operate the Alberta liquordistribution system in 2002, including the retail-ing. If we assume an average return of 15% (thisincludes GST on the retail markup) in the pri-vate retail market in Alberta, it cost more than$200 million in 2002, in addition to the higherwholesale costs created by the new regulatory re-gime.

Not all of the increased wholesale costs canbe attributed to regulatory inefficiencies. The gov-ernment recently conducted a review of the liq-uor mark-up structure in Alberta. It is worth quot-ing at length from this review:61

Manufacturers or liquor suppliers set thelanded cost of their product at their discre-tion. Curiously, the landed cost for identi-cal beer products varies across provincialjurisdictions; in some cases the variation isdramatic. In effect, the landed cost of thesame product sold in another province maybe higher or lower than the landed cost inAlberta. The influence of the landed coston the final retail price of beer productshould not be underestimated.

Table 6.3 Sales, costs, and profit data Albertaand British Columbia (1000s)Description British Columbia AlbertaGross sales $1,792,877Operating expenses $231,859Net sales (wholesale) $1,561,018 $1,379,629Cost of sales (less PST) $777,704 $887,075Social Services Tax (PST) $105,044 N/AGross profit $678,270Capital costs* $41,597Profit (net income) $636,673 $492,554Percent profit/Net sales 40.8% 35.7%Percent tax/Net sales 47.5% 35.7%*Enhancements to owned retail operationsSource: AGLC and BCLDB 2001-2002 annual reports


Table 6.4 Landed costs of a few popular products in four provincesPRODUCT AB BC SK ON

Labatt Blue (6-pack cans) $5.13 $4.18 $5.07 $7.84Molson Canadian (6-pack cans) $5.13 $4.18 $5.09 $7.84Corona (6-pack bottles) $6.06 $4.97 $4.98 $6.72Crown Royal (750 ml) $11.20 $8.40 $8.88 $9.30Smirnoff Vodka (750 ml) $8.07 $6.65 $7.09 $7.27Le Piat D’or Red (750 ml) $4.28 $3.55 $3.66 $3.20

The following table [Table 6.4] givesthe landed costs of a few popular productsin four provinces in November 2002.

The Commission has also observedthat an adjustment in the liquor mark-updoes not always translate into an equal ad-justment in the product’s wholesale price.For example, all previous adjustments inthe beer mark-up since privatization of liq-uor retailing have been downward adjust-ments, or a decrease in the mark-up rates.There have been situations where a decreasein the liquor mark-up rate for beer was metby an equal increase in the landed cost ofvarious beer products. Thus the intent of alower mark-up, specifically a decrease in thewholesale price, may be effectively negatedby a manufacturer’s/supplier’s increase inthe landed cost of the liquor product. Con-sequently, there is no decrease in the whole-sale price of the product, the price paid byliquor retailers for the liquor product be-fore it reaches their shelves.

Oligopolist liquor producers appear to be uti-lizing their market power in order to set prices.In defence of the manufacturers, the change to a

private market has increased the number and fre-quency of visits that distributors have to make tothe numerous individual stores, thus increasingthe marketing costs incurred by the distributors.Some of these increased costs have been passedon to the retailers and their customers in the formof increased landed costs, thus pushing up whole-sale prices.

These numbers very much emphasize the ex-tra costs inherent in the development of a mo-nopolistic competitive liquor retail industry in Al-berta. Public liquor control boards that own andoperate retail stores clearly have lower operatingcosts. The numerous privately-owned retail out-lets are fragmented and have multiplied rapidly,while public monopoly stores were open for fewerhours and had an integrated and centralized dis-tribution network and operations. Individual pri-vately-owned stores have higher capital and bor-rowing costs, and their inventory must be paidfor in cash. Couple this with the larger numberof stores and the costs increase considerably. Thegovernment has given up the greater control ofthe alcohol distribution system that public retail-ing ensured, and consumers have lost consider-able welfare with higher prices and less public rev-enue.

Sobering Result 45

7 Conclusions

the public’s objectives concerning liquor products.The objective of private firms is to sell product.The public’s objective is to minimize the abuse ofalcohol through the limit and control of the saleof liquor, in particular to prevent the sale to un-derage consumers and the intoxicated. Addition-ally, socially responsible marketing would educatethe public about such dangers as drinking anddriving and fetal alcohol syndrome. A publicly-owned and controlled system of distribution doesnot have this inherent incompatibly of incentives.

Because of the incentive incompatibilities in-herent in the private markets for liquor, the gov-ernment has had to implement greater regulationand appeal to the criminal justice system to amuch greater extent in order to maintain publicobjectives around alcohol. This, of course, addscosts to liquor distribution, some of which areincurred in the ministry through inspectors whileothers are shifted to police departments.

Ironically, having shifted to a private marketfor liquor distribution, the government has handi-capped this market in its ability to achieve mar-ket efficiencies. These efficiencies have been lim-ited by the control on the wholesale distributionand transportation costs—limited in order to havea ‘level playing field’—and the restriction requir-ing stand-alone outlets. Of course, in competi-tive markets for other products, one of the majorfactors in achieving market efficiencies is the free-dom to find less costly mechanisms of retailing.For good or for bad, for example, the traditionalcorner store has all but been eliminated byfranchised chains such as 7-Eleven and Mac’sConvenience Stores.

The private retail liquor market has evolvedinto one where there is considerable inefficiency

The retail liquor industry in Alberta has beenevolving over the past decade following the changefrom a government monopoly to a competitiveprivate market and the change from an ad val-orem tax to the unit or flat tax mark-up. Compe-tition was touted at the time as a means to lowerprices and improve consumer service and conven-ience, while retaining public safety through regu-lation and enforcement and maintaining publicrevenues through control at the wholesale level.

The jury is still out on the effects of privati-zation on social issues. Evidence on the links be-tween alcohol consumption and social ills is over-whelming. Absolute alcohol consumption hasbegun to climb in Alberta only since 1997. Ag-gregated crime statistics show major crimes arefalling in Alberta, as they are everywhere in NorthAmerica, largely due to demographic shifts, par-ticularly an aging population. However, thoroughstudy of the use of alcohol and criminal behav-iour at the micro level needs to be conducted inorder to see what changes are occurring due tothe greater availability of liquor since privatiza-tion of liquor retailing. This study would need tofocus on particular problem areas, such as theurban city centres and low-income residentialareas. A major study is under way by the Centrefor Addiction and Mental Health in Toronto onthe levels of alcoholism and other social ills asso-ciated with alcohol use.

From a public policy perspective, the intro-duction of a private market in the distribution ofalcoholic beverages is controversial and question-able. The incentive mechanism in the privatemarketing of any product is profit maximization.This incentive, which works very well with mostprivate products, is incompatible with many of


in the form of excess capacity, duplication, andredundancy, particularly in urban centres. Thisinefficiency generates considerable higher costs ofretailing, even though wages are at one-half com-pared to other jurisdictions. As prices are compa-rable to those found in British Columbia,62 al-though with considerable variation, the tax rev-enues returned to government are much lower inAlberta. This means the privatization effort hasbeen supported and subsidized by the governmentthrough a reduction in the tax share of the finalretail price. It is this tax reduction that has in ef-fect prevented prices from escalating due to thecost increases caused by excess capacity.

Even though the market is currently handi-capped, there are sufficient economies of scalefacilitating the expansion of retail chains. Thefuture direction appears to be moving in favourof the large grocery chains where Safeway, TheReal Canadian Superstore, the Calgary Coopera-tive Association Ltd., IGA, and others dominatethe market. These chains are more favourable fororganized unions to negotiate for a share in theproductivity gains and cost efficiencies realizedthrough size. It is expected that an industry shake-down is imminent that will likely reduce thenumber of stores per capita, increase efficiencies,and increase wages and benefits in the industry.

The movement from a government monopolyto a private market has resulted in an inefficient,monopolistically competitive market which hasthe questionable advantage of a degree of greaterconvenience with the large number of stores openlonger and later hours. However, easy access toliquor, a potentially damaging product, was not apublic objective for the distribution of liquor.Prices have increased, but not to the degree theymight have only because the share taken as gov-ernment revenue through excise taxes has fallen.

It has been suggested in some studies that in-creased convenience has lowered transportationcosts. This may appear to be true, at least in the

urban centres, but each store has a limited selec-tion in a considerably differentiated market, andthe large variability in prices may actually increasecosts when price and product search costs areadded in. Additionally, it would be rare to expectthat the average consumer would make a trip sin-gularly for liquor. It is more likely that a consumerwould combine his or her liquor shopping with atrip to the grocery store, or with some other needsuch as the dry cleaners, drug store, etc. It is thisbehaviour in fact that makes liquor store locationnear a grocery store so attractive in the first place.

Many more rural and small towns now haveliquor stores, suggesting greater convenience forthese people. However, the former ALCB storesprovided uniform prices, a higher selection, andlower prices for rural consumers than market fac-tors alone would allow. Small stores carry less stockbecause of physical limitations, and owners can-not afford to have considerable capital tied up instock. The trade-off is less selection for more con-venience in small towns and rural areas.

In the final analysis, the policy changes im-plemented in Alberta in 1993/94 have consider-ably diminished the government’s ability to im-plement public control of liquor distribution. Ad-ditionally, the government has downloaded theconsequent problems to municipal jurisdictionswhere, for example, the City of Calgary is grap-pling with appropriate controls on the retailingof liquor. The public welfare has been reducedwith the loss of tax revenue. The willingness toobtain lower tax revenue has masked the ineffi-ciencies in the retail and wholesale system bymoderating price increases. Greater product vari-ety has increased consumer benefits to some de-gree, yet consumers have also lost with the higherprices of alcoholic beverages. Along with greateravailability, liquor consumption is again on therise in Alberta, along with the associated ill-ef-fects research indicates is likely to follow.

Sobering Result 47

Endnotes

14 Statistics Canada, Spending Patterns in Canada 2000,Catalogue no. 62-202-XIE, Minister of Industry,2002.

15 James A. Johnson, Ernest H. Oksanen, Michael R.Veall, and Deborah Fretz, “Short-Run and Long-Run Elasticities for Canadian Consumption of Al-coholic Beverages: An Error-Correction Mechanism/Cointegration Approach”, Review of Economics andStatistics, February 1992, v. 74, iss. 1, pp. 64-74.

16 Statistics Canada, Your Guide to the Consumer PriceIndex, Catalogue No. 62-557-XPB96001, Ministerof Industry 1996.

17 Alberta Gaming and Liquor Commission, 2001-2002 Annual Report, 2002.

18 World Health Organization, Global Status Report OnAlcohol, Geneva: 1999, p.9.

19 See for example, K. Makela, R. Room, E.Single, P.Sulkunen, and B. Walsh, Alcohol, Society, and the StateI, Toronto: Addiction Research Foundation, 1981.

20 Carol Oliver, Heather White, and Dr. MargaretEdwards, Fetal Alcohol Syndrome: a hopeful challengefor children and families and communities, A Reportto Health Canada: June 4, 1998.

21 Marc Eliany, editor, Alcohol in Canada, Ottawa:Health and Welfare Canada, 1989.

22 Thomas Babor, “Alcohol, Economics and the Eco-logical Fallacy, Toward an Integration of Experimen-tal and Quasi-experimental Research”, in Eric Sin-gle and Thomas Storm, editors, Public Drinking andPublic Policy: proceedings of a symposium on observa-tion studies held at Banff, Alberta, Canada, April 26-28, 1984, Toronto: Addiction Research Foundation,1985.

23 Health Canada, Results of Canada’s Alcohol and OtherDrugs Survey (CADS), November 17, 1995.

24 Statistics Canada, Juristat— Catalogue no. 85-002,Vol. 21, No.9, October 2001.

25 I have requested data on alcohol related crime fromthe Calgary Police Service. The CPS is currentlycompleting a study of its own to be released in thenear future. I am relying on data included in West(2002) and in The Canadian Taxpayers Federationreport, Ending the Prohibition of Competition, the casefor competitive liquor sales in British Columbia, March2002.

1 Howard Palmer, Alberta: A New History, Edmon-ton: Hurtig Publishers, 1990.

2 The Minister of Municipal Affairs responsible forthe ALCB.

3 Statistics Canada refers to the provincial revenuefrom alcohol sales as profit; technically it is excisetax revenue. The amount transferred to governmentgeneral revenues was never exactly the same amountas the profit in a given year.

4 Alberta Liquor and Gaming Commission, Liquorin Alberta — Quick Facts, January 2003.

5 Alberta Liquor and Gaming Commission, “Beforeand After Privatization”, http://www.gaming.gov.ab.ca/who/before_after.asp.

6 Douglas West, “The Privatization of Liquor Retail-ing in Alberta”, Public Policy Sources Number 5, Van-couver: Fraser Institute, 1997.

7 Gordon Laxer, Duncan Green, Trevor Harrison, andDean Neu, Out of Control: Paying the Price For Pri-vatizing Alberta’s Liquor Control Board, Ottawa: Ca-nadian Centre for Policy Alternatives. Septem-ber 1994.

8 Douglas West, The Privatization of Liquor Retailingin Alberta, prepared for the Centre for the Study ofState and Market, Faculty of Law, University of To-ronto, 1995; Public Policy Sources Number 5, FraserInstitute, 1997; Fraser Institute Digital Publication,January 2003.

9 Clive Seale, editor, Researching Society and Culture,Thousand Oaks, Calif.: Sage Publications, 1998.

10 Health Canada, Results of Canada’s Alcohol and OtherDrugs Survey (CADS), November 17, 1995.

11 Table 2. Recorded per capita consumption of purealcohol (litres) per adult 15 years of age and over in1996, World Health Organization, Global StatusReport On Alcohol, Geneva: 1999, p.10-13.

12 Table 3. Trends in recorded per capita consumptionof pure alcohol, (litres) per adult 15 years of age andover between 1970-1996. World Health Organiza-tion, Global Status Report On Alcohol, Geneva: 1999,pp.13-16.

13 Statistics Canada, Your Guide to the Consumer PriceIndex, Catalogue no. 62-557-XPB, Minister of In-dustry, 1996.


26 By John J. DiIulio, Jr., “Broken Bottles Alcohol, dis-order, and crime”, Brookings Review (Pages 14-17),Spring Vol. 14 No. 2, The Brookings Institution,1996.

27 Centre for Addiction and Mental Health, Best Ad-vice: Reducing The Harms Of Alcohol Related Colli-sions, Toronto: Addiction Research Foundation, July2002.

28 Statistics Canada, Juristat — Catalogue no. 85-002,Vol 19, No.11., November 1999.

29 Centre for Addiction and Mental Health Best Ad-vice, Reducing The Harms of Alcohol Related Colli-sions, Toronto: Addiction Research Foundation, July2002.

30 Global Television, program on underage liquor storecustomers, aired in November 2002.

31 Alcohol Related Birth Injury (FAS/FAE) ResourceSite, “Retailers asked To Support Fetal Alcohol Syn-drome Campaign”, From Vendor Magazine, Winter1999.

32 British Columbia Liquor Distribution Branch, 2001-2002 Annual Report, June 2002.

33 Calgary Herald, Tuesday, March 18, 2003.34 For example, in 1994 the LCBO estimated liquor

smuggling accounted for $500 million and illegalliquor manufacturing for over $300 million, in NuriT. Jazairi, The Impact of privatizing The Liquor Con-trol Board of Ontario, Toronto: Mimeo, September1994.

35 Excise Act Review, A Proposal for a Revised Frame-work for the Taxation of Alcohol and Tobacco Prod-ucts, Revenue Canada, February 1997.

36 The Retail Sales Tax in Canada, Canadian Tax Paper77, A.J. Robinson, Canadian Tax Foundation, 1986

37 A.C. Pigou, The Economics of Welfare, London:Macmillan, 1930.

38 Health Benefits and Risks of Moderate Alcohol Con-sumption Sandra Goatcher, Policy Background PaperPrepared by Policy and Planning for the Alberta Al-cohol and Drug Abuse Commission, May 2002.

39 The Ramsey rule (or inverse rule) results from theanalysis of the excess burden of a commodity tax.Excess burden is the net lost welfare from a tax. Theconsumer loses, with a tax , an amount greater thanthe revenue the government gains—the differenceis the excess burden. Excess burden can be mini-mized if each commodity tax level is specific andinverse to the price elasticity of demand. For exam-ple, a good with low elasticity (inelastic), i.e. thequantity is not responsive to price, could have a hightax and not have much excess burden. A price elas-

tic commodity would have a high excess burden thegreater the tax imposed. Thus to minimize total ex-cess burden each good should have a specific tax rateapplied according to its price elasticity.

40 Leisure is a ‘good’ that one enjoys when they do notwork. It can only be taxed implicitly by taxing at agreater rate goods complementary (consumedjointly) to leisure. This concept is called the Corlett-Hague rule.

41 Excise Act Review: A Proposal for a Revised Frame-work for the Taxation of Alcohol and Tobacco Prod-ucts, Revenue Canada Tax Services, Department ofFinance, Ottawa, February 1997.

42 Excise Act (R.S. 1985, c. E-14) Updated to Decem-ber 31, 2001.

43 Excise Tax Act (R.S. 1985, c. E-15 ) Updated to De-cember 31, 2001.

44 A.J. Robinson, The Retail Sales Tax in Canada, Ca-nadian Tax paper No. 77, Ottawa: Canadian TaxFoundation, 1986.

45 James A. Johnson, Ernest H. Oksanen, Michael R.Veall, and Deborah Fretz, Short-Run and Long-RunElasticities for Canadian Consumption of AlcoholicBeverages: An Error-Correction Mechanism/Cointegration Approach, Review of Economics andStatistics, February 1992, v. 74, iss. 1, pp. 64-74.

46 For example see: Sara Markowitz, MichaelGrossman, Alcohol regulation and violence towardschildren, Cambridge, Mass: National Bureau of Eco-nomic Research, 1998.

47 Sara Markowitz, Criminal violence and alcohol bev-erage control: evidence from an international study,Cambridge, Mass: National Bureau of Economic Re-search, 2000.

48 The Costs of Substance Abuse In Alberta, December1996. Adapted from the report: Eric Single et al,The Costs of Substance Abuse in Canada, Highlightsof a major study of the health, social and economiccosts associated with the use of alcohol, tobacco andillicit drugs, Ottawa: Canadian Centre on SubstanceAbuse, 1996.

49 Prices include all taxes (PST in BC only) and de-posit charges.

50 Review of Liquor mark-up structure and related, Find-ings and Recommendations, AGLC, Feb. 20, 2003.

51 Alberta Liquor and Gaming Commission, Liquorin Alberta — Quick Facts, January 2003.

52 British Columbia Liquor Distribution Branch, 2001-2003 Annual Report, June 2002.

53 Paul J. Gruenewald, Alex B. Millar, “Alcohol avail-ability and the ecology of drinking behavior”, Alco-

Sobering Result 49

hol Health & Research World, 1993, Vol. 17 Issue 1,p39.

54 Paul J Gruenewald, Alexander B. Millar, “Access toalcohol”, Alcohol Health & Research World, 1996, Vol.20 Issue 4, p244.

55 See: C. Michael Fellows, Gregory L. Flanagan, StanShedd, and Roger N. Waud, Economics in a Cana-dian Setting, New York: HarperCollinsCollegePublishers, 1993 or C. Michael Fellows,Gregory L. Flanagan, Stan Shedd, Economic Issues, aCanadian Perspective, Toronto: Irwin (McGraw-HillRyerson Ltd.), 1997.

56 See for example Daniel Spulber, Regulation and Mar-kets, Cambridge: The MIT Press, 1989.

57 Edward H. Chamberlain, The Theory of Monopolis-tic Competition, Cambridge: Harvard UniversityPress, 1933.

58 Joan Robinson, The Economics of Imperfect Compe-tition, London: Macmillan & Co., 1933.

59 Laxer et al, pg 11.60 ALCB annual reports, 1991-1994.61 Review of Liquor Mark-up Structure and Related Find-

ings and Recommendations, AGLC, Feb. 20, 2003p66.

62 British Columbia’s retail liquor prices include a pro-vincial sales tax of 10 percent. Therefore, BC liquorprices are effectively 10 percent cheaper. (Otherproducts have a 7.5 percent provincial sales tax withsome exemptions.) Comparable liquor products,under the same market conditions, would be 10percent higher in BC due to this tax.


Bibliography

Alberta Liquor Control Board, Annual Reports, 1981-1994.

Alberta Liquor Control Board, A New Era in Liquor Dis-tribution, the Alberta Experience, December 1994.

Alberta Gaming and Liquor Commission, 2001-2002Annual Report, 2002.

Alberta Gaming and Liquor Commission, “Liquor inAlberta — Quick Facts,” January 2003.

Alberta Liquor and Gaming Commission, Review of Liq-uor Mark-up Structure and Related Findings and Rec-ommendations, Feb. 20, 2003.

Alberta Liquor and Gaming Commission, “Before andAfter Privatization”, http://www.gaming.gov.ab.ca/who/before_after.asp.

Alberta Government Liquor Control, Review of Liquormark-up structure and related, Findings and Recom-mendations, , Feb. 20, 2003.

Alcohol Related Birth Injury (FAS/FAE) Resource Site,“Retailers asked To Support Fetal Alcohol SyndromeCampaign”, From Vendor Magazine, Winter 1999.

Babor, Thomas, “Alcohol, Economics and the Ecologi-cal Fallacy, Toward and Integration of experimentaland Quasi-experimental Research”, in Eric Singleand Thomas Storm, editors, Public drinking andpublic policy: proceedings of a symposium on observa-tion studies held at Banff, Alberta, Canada, April 26-28, 1984, Toronto: Addiction Research Foundation,1985.

British Columbia Liquor Distribution Branch, 2001-2002 Annual Report, June 2002.

Canadian Centre on Substance Abuse, “The Costs ofSubstance Abuse In Alberta, December, 1996Adapted from the report: Eric Single et al, The Costsof Substance Abuse in Canada, Highlights of a ma-jor study of the health, social and economic costs

associated with the use of alcohol, tobacco and il-licit drugs, Ottawa:, 1996.

Canadian Taxpayers Federation, “Ending the Prohibitionof Competition, the case for competitive liquor salesin British Columbia”, March 2002.

Calgary Herald, Tuesday, March 18, 2003.

Centre for Addiction and Mental Health “Best Advice:Reducing The Harms of Alcohol Related Collisions”,Toronto: Addiction Research Foundation, July 2002.

Chamberlain, Edward H., The Theory of MonopolisticCompetition, Cambridge: Harvard University Press,1933.

Cullis, John G. and Philip R.Jones, Public Finance andPublic Choice, 2nd ed., Oxford: New York: OxfordUniversity Press, 1998.

DiIulio, John J., Jr., “Broken Bottles Alcohol, disorder,and crime”, Brookings Review (Pages 14-17), SpringVol. 14 No. 2, The Brookings Institution, 1996.

Edwards G, et al., Alcohol Policy and the Public Good.Oxford medical publication. Oxford UniversityPress, 1994.

Eliany, Marc, editor, Alcohol in Canada, Ottawa: Healthand Welfare Canada, 1989.

Fellows, C. Michael, Gregory L. Flanagan, Stan Shedd,and Roger N. Waud, Economics in a Canadian Set-ting, New York: HarperCollins CollegePublishers,1993.

Fellows, C. Michael, Gregory L. Flanagan, Stan Shedd,Economic Issues, a Canadian Perspective, Toronto:Irwin (McGraw-Hill Ryerson Ltd.), 1997.

Global Television, program on underage liquor store cus-tomers, aired in November 2002.

Sobering Result 51

Goatcher, Sandra, Health Benefits and Risks of ModerateAlcohol Consumption, Policy Background Paper, Pre-pared by Policy and Planning for the Alberta Alco-hol and Drug Abuse Commission, May 2002.

Government of Canada, Excise Act (R.S. 1985, c. E-14 )Updated to December 31, 2001

Government of Canada, Excise Tax Act ( R.S. 1985, c. E-15 ) Updated to December 31, 2001

Gruenewald, Paul J., Alex B. Millar, “Alcohol availabilityand the ecology of drinking behavior”, Alcohol Health& Research World, Vol. 17, Issue 1, 1993.

Gruenewald, Paul J, Alexander B. Millar, “Access to al-cohol”, Alcohol Health & Research World, Vol. 20,Issue 4, 1996.

Health Canada, “Results of Canada’s Alcohol and OtherDrugs Survey (CADS)”, November 17, 1995.

Jazairi, Nuri T., “The Impact of privatizing The LiquorControl Board of Ontario”, Toronto: Mimeo, Sep-tember 1994.

Johnson, James A., Ernest H. Oksanen, Michael R. Veall,and Deborah Fretz, “Short-Run and Long-RunElasticities for Canadian Consumption of AlcoholicBeverages: An Error-Correction Mechanism/Cointegration Approach”, Review of Economics andStatistics, v. 74, iss. 1. February 1992.

Laxer, Gordon, Duncan Green, Trevor Harrison, andDean Neu, “Out of Control: Paying the Price ForPrivatizing Alberta’s Liquor Control Board,” Ottawa:Canadian Centre for Policy Alternatives. Septem-ber 1994.

Makela, K., R. Room, E. Single, P. Sulkunen, and B.Walsh, Alcohol, Society, and the State I, Toronto:Addiction Research Foundation, 1981.

Markowitz, Sara, Michael Grossman, “Alcohol regula-tion and violence towards children”, Cambridge,Mass: National Bureau of Economic Research,c1998.

Markowitz, Sara, “Criminal violence and alcohol bever-age control: evidence from an international study”,Cambridge, Mass: National Bureau of Economic Re-search, 2000.

Oliver, Carol, Heather White, and Dr. Margaret Edwards,“Fetal Alcohol Syndrome: a hopeful challenge forchildren and families and communities,” A Reportto Health Canada: June 4, 1998.

Pigou, A. C., The Economics of Welfare, London:Macmillan, 1930.

Revenue Canada, Excise Act Review: A Proposal for a Re-vised Framework for the Taxation of Alcohol and To-bacco Products, Tax Services, Department of Finance,Ottawa, February 1997.

Robinson, A. J. The Retail Sales Tax in Canada, Cana-dian Tax Paper 77, Canadian Tax Foundation: 1986.

Robinson, Joan, The Economics of Imperfect Competition,London: Macmillan & Co., 1933.

Seale, Clive, editor, Researching Society and Culture, Thou-sand Oaks, Calif.: Sage Publications, 1998.

Spulber, Daniel, Regulation and Markets, Cambridge: TheMIT Press, 1989.

Statistics Canada, Canadian Crime Statistics [annual is-sues] 1997-2001, Catalogue no. 85-205-XIE, Ca-nadian Centre for Justice Statistics, Minister of In-dustry, 1998-2002.

Statistics Canada, Your Guide to the Consumer Price In-dex, Catalogue No. 62-557-XPB96001, Minister ofIndustry, 1996.

Statistics Canada, Juristat — Catalogue no. 85-002, Vol19, No.11, November 1999.

Statistics Canada, Juristat— Catalogue no. 85-002, Vol21, No.9, October 2001.

Statistics Canada, Spending Patterns in Canada 2000,Catalogue no. 62-202-XIE, Minister of Industry,2002.

Statistics Canada, The Control and Sale of Alcoholic Bev-erages in Canada, Fiscal year end for March 31, [an-nual issues] 1997-2001, Catalogue no 63-202-XIB,Minister of Industry, 1998-2002.

Streissguth, A. P., F.L Bookstein., P.D. Sampson, andH.M. Barr. The Enduring Effects of Prenatal Alcohol


Exposure on Child Development: Birth Through SevenYears, A Partial Least Squares Solution (4th edition).Ann Arbour: The University of Michigan Press,1996.

Tirole, Jean, The Theory of Industrial Organization, Cam-bridge, Mass.: MIT Press, 1988.

West, Douglas, “The Privatization of Liquor Retailingin Alberta,” prepared for the Centre for the Studyof State and Market, Faculty of Law, University ofToronto, 1995.

West, Douglas, “The Privatization of Liquor Retailingin Alberta”, Public Policy Sources Number 5, 1997.

West, Douglas, “The Privatization of Liquor Retailingin Alberta”, Digital Publication, Vancouver: FraserInstitute, January 2003.

World Health Organization, Global Status Report On Al-cohol, Geneva: 1999.

By Greg Flanagan

The Alberta Liquor

Retailing Industry

Ten Years after

Privatization

S o b e r i n gR e s u l t :

Parkland Institute is an Alberta research network that examines public policyissues. We are based in the Faculty of Arts at the University of Alberta andour research network includes members from most of Alberta’s Academicinstitutions as well as other organizations involved in public policy research.

Parkland Institute was founded in 1996 and its mandate is to:

• conduct research on economic, social, cultural and political issues facingAlbertans and Canadians.

• publish research and provide informed comment on current policy issuesto the media and the public.

• sponsor conferences and public forums on issues facing Albertans.• bring together academic and non-academic communities.

The Canadian Centre for Policy Alternatives is one of Canada's leadingprogressive policy research organizations. Founded in 1980, the CCPA issupported by its 8,000 individual and organizational members.

The Centre, through its work, seeks to promote economic and social literacyamong Canadians on important issues affecting their lives.

The CCPA publishes reports, books and other publications, including amonthly magazine.

Canadian Centre for Policy Alternatives

Parkland Institute

Canadian Centre for Policy Alternatives Sobering...

Documents

Transcript of Canadian Centre for Policy Alternatives Sobering...