Bad stats are hard to kill

Dr Eric Crampton
Insights Newsletter
16 August, 2019

It is the hope that this time it will be different that really kills you. Sisyphus at least knew his labours were futile and could resign himself to the task of forever pushing the boulder uphill.

A decade ago, Matt Burgess and I went through BERL’s headline-grabbing $4.8 billion measure of alcohol’s social cost. Sir Geoffrey Palmer, heading the Law Commission’s alcohol legislation review, had cited the gap between BERL’s measure of social cost and alcohol tax revenue to justify imposing tighter controls.

But BERL’s figure bore no resemblance to normal notions of external costs. BERL even counted $700 million in drinkers’ expenditures on alcohol as a social cost.

There were substantial problems in BERL’s tally of almost $1.5 billion in labour costs such as wages not earned because of alcohol – only a tiny fraction of any corrected figure could be considered ‘external’. Worse, the standard figures on the intangible costs of premature mortality include the cost of any reduced output. BERL was was double-counting by adding both costs.

When BERL measured health costs, it ignored any conditions, like cardiovascular disorders, where alcohol reduces net health costs. When BERL measured crime costs, a prisoner saying alcohol contributed at least ‘some’ towards his incarceration was enough to count all the costs as being due to alcohol.

Our critique of the figure also received a fair bit of attention at the time, with BERL’s number generally considered by economists as on par with toxic waste. Our corrected figure was around a fifth of BERL’s headline, and much closer to the alcohol excise take.

We had finished pushing the rock uphill.

Last year, BERL provided what appears to be an inflation and population adjustment to the old figure: $7.8 billion per year. I had to push the rock uphill again, with a piece in Newsroom reminding everyone about the problems in the tally.

The figure has been all over the headlines again this week as part of uncritical journalistic promotion of a renewed prohibitionist campaign. Shoulder to the rock, one more time.

Knowing it to be futile, Sisyphus could at least make a game of the futility – timing how long it took to push the rock up the hill one day, and then trying to beat that record.

Maybe I should just give up hope that journalists will ever care about getting the numbers right when they are on a temperance crusade.

Stay in the loop: Subscribe to updates