We examine the impact of privatization of retail sale of alcohol in Alberta, Canada, between 1985 and 1995 on mortality rates from suicide. Privatization took place in three stages: The opening of privately owned wine stores in 1985, the opening of privately owned cold beer stores and the selling of spirits and wine in hotels in the rural area in 1989-90, and finally privatization of all liquor stores in 1994. Interrupted time series analysis with Auto Regressive Integrated Moving Average (ARIMA) modeling was applied to male and female suicide rates to assess the impact of the three stages of privatization. The analyses demonstrated that most of the privatization events resulted in either temporary or permanent increases in suicide mortality rates. Other alcohol-related factors, including consumption levels and Alcoholics Anonymous (AA) membership rates, also affected suicide mortality rates. These analyses suggest that privatization in Alberta has acted to increase suicide mortality rates in that province.
Key words: Alcohol, suicide, privatization, Alcoholics Anonymous, unemployment.
Many Western jurisdictions had publicly controlled alcohol retail systems in the middle portion of the 20th century. However, by the latter part of the century many were considering moving away from public control of alcohol retailing to private control, while other jurisdictions privatized some or all of the alcohol retailing system (Her, Giesbrecht, Room & Rehm 1998; Mann, Rehm, Giesbrecht et al. 2005). Proponents of privatization point to the potential economic benefits that may occur. These include funds received from the sale of previously state-owned retail outlets, the sale of licences to sell alcohol and the cost savings resulting from the reduction in staff and facilities required for alcohol sales (Beverage Alcohol System Review Panel [BASRP] 2005). However, others point to the likely impact of privatization of alcohol sales on consumption and health problems (e.g., Mann, Rehm et al. 2005). One analysis in Ontario estimated that alcohol consumption would increase by between 10% and 20% if Ontario's government-controlled alcohol retail system were fully privatized (Her et al. 1998). Studies of privatization of sales of alcoholic beverages in the United States indicate that availability and consumption increased; Wagenaar & Holder (1995), in a review of the American literature, found increases in consumption ranging between 13% and 150%. Increasing alcohol consumption can be expected to increase alcoholrelated morbidity and mortality (Bruun, Edwards, Lumio, Mäkelä, Pan, Popham et al. 1975; Room, Babor & Rehm 2005), and government control of retail sales of alcohol has been recommended as a key public health measure to control alcohol-related damage in society (Babor, Caetano, Casswell, Edwards, Giesbrecht, Graham et al. 2003).
In the Canadian experience, moving from public to private control of retail outlets has occurred in the provinces of Alberta and Quebec (Demers & Fournier 2006; MacKenzie & Giesbrecht 2006), and in Ontario and other provinces has been proposed on several occasions (Giesbrecht, Stoduto & Kavanagh 2006). In Alberta and Quebec the numbers of alcohol outlets increased substantially (Her et al. 1998). In Quebec some divergence in evidence on consumption changes has been seen (Her et al. 1998), with the most recent analyses finding evidence for modest but significant increases in some forms of consumption (Trolldal 2005b). In Alberta evidence has been found for a significant increase in consumption of spirits, particularly at a time when consumption was decreasing in other parts of the country (Flanagan 2003; Trolldal 2005a). As well, government tax levels have been reduced several times since demonopolization of retail outlets, and this has been attributed to pressure from the alcohol retailers (Flanagan 2003).
In their review of the literature on privatization of alcohol sales, Her, Giesbrecht, Room & Rehm (1999) noted that "there is relatively little evidence on the effects of demonopolization, or privatization, directly on alcohol-related harms" (p. …