Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
The hypothesis of additive utility (or preference independence) is often applied to the demand for broad aggregates. Recent testing provides some evidence favourable to the hypothesis, thus overturning the older results based on the standard asymptotic tests which are seriously biased against the null in small samples. Using data for seven countries and a variety of tests, this paper shows that preference independence also cannot be rejected for more narrowly defined commodities - beer, wine and spirits. The implications of the results for efficient taxation of alcoholic beverages are also explored.