|
Article Excerpt
Increased availability of alcohol has had a tendency to result in increased consumption, which in turn usually affects alcohol-related problems. Availability of alcohol is usually divided into two main categories: economic availability in terms of disposable income and beverage prices, and physical availability in terms of outlet density, age limits, and days and hours of sale (Bruun et al., 1975; Edwards et al., 1994; Single, 1988). The relationships between availability, alcohol consumption and alcohol-related harm have been key issues in countries where alcohol policies have been formulated and implemented, primarily in parts of North America and in the northern part of Europe.
The hypotheses regarding the effect of changes in economic availability on alcohol sales are based on traditional microeconomic theory. The premise is that lower prices of a commodity and higher disposable income among the inhabitants lead to greater sales, all other factors being equal (Pindyck & Rubinfeld, 1989). This hypothesis has been confirmed in many studies of the demand for alcohol, though the results differ from study to study (Godfrey, 1989; Osterberg, 1995).
The effects of changes in price and income on the demand for a commodity are often measured in terms of price and income elasticities. These elasticities are mostly expressed in percentages that indicate how much a 1% increase in price or income affects sales. Several studies have been made regarding the effect of changes in prices of alcohol and disposable income on sales in Canada. In these studies, estimated price elasticities have ranged from -0.30 to -1.77 for spirits, from -0.50 to -1.78 for wine, and from -0.03 to -0.33 for beer (Johnson & Oksanen, 1974, 1977; Johnson et al., 1992; Lau, 1975; Quek, 1988). Estimated income elasticities have ranged from 0.10 to 0.99 for spirits, from 0.00 to 1.13 for wine, and from -0.02 to 0.48 for beer (Johnson & Oksanen, 1974, 1977; Johnson et al., 1992).
Two time-series studies concerning price and income elasticities on the provincial level in Canada have been made. Adrian & Ferguson (1987) studied the period 1958-1981 and Johnson et al. (1992) studied the period 1956-1983. Both studies found that the estimates regarding price and income varied from province to province, but the results presented in the two studies differed as well. However, the estimates in both studies were not very reliable, since the study periods were quite short (23 and 27 years, respectively).
In general, the ideas regarding the effects of changes in physical availability of alcohol on sales are based on a theory similar to the one regarding economic availability. Apart from the price of a beverage, there are so-called transaction costs associated with the purchases of alcohol, such as the distance to a store, opening hours, possible queueing time, and so forth (Babor et al., 2003). The assumption is that the more inconvenient it is to purchase alcohol and the higher the transaction costs, the less alcohol will be bought.
Studies of the relationship between outlet density and alcohol sales have been relatively few in number (see Babor et al. (2003) for an overview). Some of the studies have been cross-sectional analyses (Harford et al., 1979; Parker, Woltz & Harford, 1978). In these kinds of studies it has often been very difficult to determine the specific role played by the number of outlets, because the jurisdictions also differ in terms of many other determinants of alcohol consumption. However, when Gruenewald, Ponicki & Holder (1993) conducted a time-series cross-sectional study regarding the relationship between off-premise outlet density and sales of alcohol in a number of states in the U.S., their results showed that outlet density had significant effects on sales in most of their analyses.
Most studies have found no effects of changes in the number of on-premise outlets on consumption. In most countries, on-premise consumption accounts for a minor part of the total consumption. Consequently, if there was a positive relationship, changes in the number of on-premise outlets was normally not large enough to affect total consumption (Brook et al., 1981). However, several studies have found a significant effect of changes in the number of on-premise outlets on incidence of drunk driving and criminal violence (Blose & Holder, 1987; Norstrom, 2000).
Some studies have been made regarding the effects of privatization of retail sale monopolies on the alcohol market. The increasing number of outlets after privatization has been the main issue, even though privatization also implies changes in other kinds of availability. In the U.S. several states privatized previous monopolies of alcohol sales, primarily of wine, in the 1970s and 1980s. These states were Maine, Idaho, Iowa, Washington (fortified wine), Virginia, Alabama, Montana and New Hampshire. After sales of a particular beverage had been privatized, they increased in most cases (Holder & Wagenaar, 1990; Macdonald, 1986; Mulford, Ledolter & Fitzgerald, 1992; Wagenaar & Holder, 1995). In 1967 in Sweden and in 1969 in Finland, grocery stores were given permission to sell medium-strength beer. Following this change the total consumption of beer increased in both countries (Makela, Rossow & Tryggvesson, 2002).
According to several studies, changes in days of sale have had effects on sales of alcohol, but effects of changes in hours of sale have been ambiguous (Babor et al., 2003). However, a number of studies have found that changes in age limits on the purchase of alcohol have had significant effects on the number of fatal traffic crashes involving young drivers (Wagenaar, 1983; Whitehead, 1980; Williams et al., 1983).
In Canada each province has had its own publicly owned retail sale system for off-premise sales, and the wholesale trade of alcohol, since the prohibition era ended in the 1920s. There have been a few exceptions in which off-premise sales were...
|
