Individual Heterogeneity in Alcohol Consumption: The Case of Beer, Wine and Spirits in Australia*

Abstract

This paper examines the effects of socioeconomic and demographic factors on Australian individuals’ participation in beer, wine and spirits consumption using unit‐record data during 1991–2001. A trivariate probit formulation allows for participation in the three alcoholic beverages to be modelled jointly accounting for correlation via unobserved personal characteristics. Own and cross price elasticities are estimated for both unconditional and conditional participation probabilities. While the three beverages are commonly considered closely related economic goods, we find that they relate to rather heterogenous population groups and the correlations via the unobservable characteristics are low in magnitude.

The effect of price increases on predicted alcohol purchasing decisions and choice to substitute

“…While research has shown that associations exist between socio-demographic characteristics, the type of alcoholic beverage consumed, and the consumer behaviours exhibited, less is known about how these factors influence the decision to substitute one beverage for another (Ramful & Zhao, 2008). Some evidence suggests that the choice to substitute one beverage type for a cheaper alternative is affected by drinking patterns: heavy drinkers appear to be more responsive to price increases to their preferred beverage than moderate drinkers Wagenaar, 2010a).…”

The Demand for Alcohol in Thailand

“…Occupation has been identified as one of the key determinants to explain the likelihood of alcohol expenditure being observed. In Australia, Ramful and Zhao () classified individuals into employed, unemployed, studying and other. They found that Australians who worked and studied had a greater likelihood of spending on wine, but Australians who studied had less likelihood of consuming spirits.…”

Wine Export Shocks and Wine Tax Reform in Australia: Regional Consequences Using an Economy-wide Approach*

“…As of early 2009, that seemed a reasonable characterisation of the type of shock the industry would face in 2009-10. Our choice to not allow cross-price effects is vindicated by the recent estimates by Ramful and Zhao (2008), using a trivariate probit formulation, which allows for the three alcoholic beverages to be modelled jointly to account for correlation via unobserved personal characteristics. Specifically, the second export demand shock assumes there is a 10 per cent decline in super premium wine export demand -less severe 10 The extent to which preference independence applies for different types of alcohol has been debated in the literature.…”