For years, public health advocates have insisted higher excise taxes will reduce both overall consumption and alcohol problems. Meanwhile, industry members and some researchers have argued that alcohol beverages are relatively “inelastic” (higher prices don’t significantly reduce consumption) and that heavy drinkers/problem drinkers especially are unlikely to be deterred by higher taxes.

A new study of US Government surveys strongly supports the industry’s position. Looking at drinking rate data from the federal Behavioral Risk Factor Surveillance System (BRFSS) surveys from 1984-2009 and federal/ state beer excise taxes (a proxy for overall alcohol taxes), the authors tried to tease out the effects of tax rates among different races and among light, moderate and heavy drinkers. They made three broad conclusions:

• Most important: “For all races/ethnicities, the estimated tax effects on consumption are large and significant among light drinkers (1-40 drinks per month), but shrink considerably for moderate drinkers (41-99) and heavy drinkers (100+).” In fact, the authors wrote: “We cannot reject [that] alcohol consumption of the latter types is unresponsive to tax changes.” Translated, it is very possible, perhaps likely, that tax changes have no effect on moderate to heavy drinkers, those most likely to experience alcohol-related problems.

• “The tax effect on any alcohol use” (whether adults age 18+ drink or not) “is largest among White and smallest among Hispanic” adults.

• Among “existing drinkers. Native Americans and other race/multi-race are most responsive to tax effects while Hispanics least.”

The data are flawed (see below), but the numbers showed that while higher taxes have a negative impact on those drinking up to one drink/day, once drinking exceeds a drink per day, again across races, any negative impact declines markedly. And for some groups, for example whites who consumed 41-99 drinks per month, higher taxes had no negative impact at all.

The authors acknowledge several “important weaknesses” in the study, including: 1) very few and modest state tax changes during the period of study; 2) alcohol use figures did not distinguish between alcohol beverages; 3) the beer tax was used as a proxy for “average” taxes, but drinking preferences may vary across races/types of drinkers.

Economist Victor Fung at Guerilla Economics reviewed the study. He told us: “Unfortunately, rather than using true price data across different alcohol categories, the authors of this report, for some reason limit themselves to examining beer taxes in isolation. Because of this, the elasticity figures that they present in reality only measure differences between states – differences that in many cases have existed since the end of Prohibition. However, when the authors examine the effects across individuals in the same state, they do show how tax changes affect the consumption levels across heavy, moderate and light drinkers. In doing this, the authors show that – as economists would expect – heavy drinkers are much less responsive to tax changes than those who consume alcohol either moderately or occasionally.” Cleary more study is called for, as the authors note. But they also cite a 2009 study that came to virtually the same conclusion. That study also distinguished between two groups of drinkers, “one is significantly responsive to price while the other is not. The responsive group consists of light drinkers…while the moderate and heavy drinkers occupy the unresponsive group.” Ref 1