Skip to Content
drug information
Matters of Substance

Counting the cost

Tuesday, September 1, 2009

Are arguments for an increase in alcohol taxes being based on substandard economic research? Or are opponents of a hike misusing an economics paper to make their point? David Young examines the controversy over the BERL report on the costs of alcohol.


Before the Law Commission made any official recommendations on the future of New Zealand’s liquor laws, public debate had already started.

The independent provider of advice to Parliament on laws that need reforming, updating, or developing seldom attracts mainstream criticism. Its careful recommendations are generally backed by considerable intellectual heft.

However, the Business Roundtable’s Roger Kerr and National Party-aligned blogger David Farrar are among those who criticised the Law Commission, after President Sir Geoffrey Palmer suggested in a speech that liquor tax hikes could be up for discussion. Kerr and Farrar have based their criticism on the arguments from Canterbury University economist Eric Crampton.

The Law Commission was asked by the Labour Government to review all alcohol laws last year.

Palmer – Minister of Justice when the Sale of Liquor Act 1989 was passed – said, “The central issue is whether the pendulum has swung too far in the direction of liberality.”

In his speech on 24 April in Nelson, he made it clear the Law Commission will find that it has.

Palmer noted – as many experts in drug harm have pointed out – that alcohol would be classed as a Class B drug under the Misuse of Drugs Act 1975 if it were treated on its merits. He said harm minimisation must be the prime object of any new law, balanced with the need for any regulatory controls to be efficient and effective.

Palmer outlined the “dramatic findings” of a research paper commissioned by the Ministry of Health and the Accident Compensation Corporation, Costs of Harmful Alcohol and Other Drug Use.

He said the report – written in March by private firm Business and Economic Research Ltd (BERL) – found that the total social cost of alcohol and drug misuse for 2005/06 was calculated at $6.881 billion ($4.794 billion was attributed to harmful alcohol use alone), up to 50 percent of the social costs could be avoided and half of all alcohol was consumed in a harmful manner.

Palmer concluded, “We have sufficient evidence to consider whether some of the costs isolated in the BERL report should be internalised to the liquor industry. I doubt that such a proposition will be met by great enthusiasm, but it does seem to me that the taxpayer should not be asked to shoulder as much of the burden as is currently being met from public funds.”

It appears this comment in particular attracted the attention of Crampton, who says he was surprised to see Palmer using the BERL report. From his first look, Crampton had concluded, “It couldn’t really be used for policy analysis.”

Crampton began criticising the BERL report and its use by the Law Commission on his popular economics blog, Offsetting Behaviour. He wrote a commentary on the matter in The Press, and eventually, with fellow economist Matt Burgess (Research Associate at the Institute for the Study of Competition and Regulation), published a 42-page referenced review of it. This effort was not funded by any outside organisation, but was enough to attract attention from reporters, bloggers and pundits.

What is the BERL report, and why has it become so controversial?

The Ministry of Health’s National Drug Policy Team Leader, Chris Laurenson, says BERL was asked to measure “in economic terms, the aggregate costs to New Zealand society… of the abuse of alcohol and other drugs.

”BERL was asked to provide costs of alcohol from the perspective of wider society and from the perspectives of government and business.

It was asked to look at what the Ministry calls ‘tangible’ costs: “the cost of consequences to the health and welfare systems; productivity consequences in the workplace and the home; crime, law enforcement and criminal justice; road accidents; fires; environment; research and prevention.”

It was also asked to examine ‘intangible’ costs: “the cost of loss of life, pain and suffering and the costs of consequences to the wellbeing of family/whänau.”

The Ministry of Health suggested BERL use accepted guidelines for estimating the costs of substance abuse that have been endorsed by the World Health Organization, which it did. The report was also subject to peer review from the co-authors of the framework.

Much of Crampton’s criticism is technical. He claims BERL has committed basic mistakes like double-counting and using multipliers inappropriately. BERL rejects that criticism.

But he also makes broader criticisms – and it is these that have been taken up by Kerr and Farrar, who oppose higher liquor taxes.

Crampton says the much-touted finding of a $4.8 billion ‘cost’ of alcohol is exaggerated and meaningless.

To put this figure into some context: in 1996, economist Nancy Joy Devlin found the cost of alcohol-related harm varied from $1.4 to $4 billion (although she stated she was looking at “a relatively narrow range of alcohol-related effects”). A year later, Brian Easton looked at social and economic costs and reached an eye-watering figure of $16 billion.

Crampton’s complaint is that BERL’s number is too high because it included many private costs in its study, which do not hurt society itself, so should have been ‘netted out’ to make the results meaningful.

One example he cites is reduced labour productivity as a result of alcohol consumption. Drinking too much will diminish an individual’s output at work. Crampton says that the reduction in productivity should be seen as a private cost, because an individual is less likely to get a promotion or a salary rise.

BERL argues that lower productivity should be seen as a social cost, because lower output will impact on colleagues, and even on society through lost taxes. It points out that “a computer does not keep writing by itself when you have a sick day. Nor may resources be freed up for others to use if a person turns up to work hung-over.”

Crampton criticises the inclusion of other costs that BERL has tallied, including the value of forgone unpaid household work, the production and distribution costs of alcohol and reduced output because people are away from work.

He also takes issue with BERL’s definition (which follows World Health Organization categories and Australian alcohol guidelines) of harmful drinking as the consumption of 1.8 pints of beer or more and BERL’s categorisation of all drug use as harmful.

BERL assumes that all consumption it defines as harmful is irrational. In economic terms, this means drinkers and drug users are incapable of realising that the personal costs from their activity are higher than the benefits. The economists also assume that those engaged in harmful behaviour obtain no benefits (such as enjoyment) from their actions that would counteract or reduce the costs.

In Crampton’s view, even a harmful drinker gains benefits – such as enjoyment – from their first few drinks that would roughly balance the excess costs imposed by later drinks.

Crampton’s arguments are similar to those he used in a review in the Medical Journal last year about Des O’Dea’s report on tobacco taxes. Arguing that smokers gain personal benefits from their activity, he commented that even though “one may suffer adverse health consequences from smoking, this does not mean that the smoker didn’t enjoy smoking” and said that the benefits enjoyed by smokers should not be subject to substantial discounting.

He also has concerns about the Drug Harm Index, published last June by BERL for the New Zealand Police. He hasn’t subjected it to a similar review, but notes that “it counted costs of drug enforcement regimes as a cost of harmful drug use rather than as a cost of prohibition and counted no benefits of consumption to users.”

He says he would be “surprised if the results of a thorough ‘fisking’ would not be similar to the results on alcohol, other than that there are no offsetting tax revenues in the case of prohibited substances.”

In the case of the BERL report, Crampton argues that a “policy-relevant report” should either count all the costs and weigh them against benefits (a cost-benefit study, which is a much bigger, more expensive task than the consultants were actually set) or count only the external portion of those costs, to identify the impact on society.

Crampton cites approvingly a report written by Felicity Barker of Treasury in 2002, which did what he argues the BERL report should have done: netted out the costs that fell on the individual.

Barker’s conclusion was: “In 1999/00, the amount of revenue collected from the tax on alcohol was $580 million. This is near the mid-point of the estimated bound of the external tangible costs of alcohol. Thus, the current rate of excise tax can be justified on externality grounds.”

Crampton himself attempted to net out what he sees as the privately borne costs in the BERL report. This involved reverse-engineering BERL’s numbers. He concluded this reduced the total costs of alcohol by 40 percent to $2,955.1 million, of which the “policy-relevant” net external costs of alcohol (basically crime, healthcare costs and road crashes) amount to just $146.3 million, or less than 5 percent of BERL’s $4.8 billion figure.

After seeking comments from BERL and others on mistakes and misinterpretations in his review, Crampton later reworked the numbers, and, instead of a net external cost, came up with a net external benefit from alcohol of $37.8 million.

In an online response, BERL says Crampton fundamentally misinterprets the study’s brief and, on this basis, employs an inappropriate framework for his analysis.

“The project brief was not to assess benefits, nor to provide policy analysis. The brief was to quantify the costs of harmful use of alcohol and other drugs using an internationally recognised method.”

Crampton responds: “It would have been pretty simple for BERL to have added in a section tabulating the external portion of costs while doing all of their other tabulations; it’s been a lot of work to instead reverse-engineer those numbers.”

For their part, the BERL economists criticise Crampton for using a “a nonverifiable argument”: that all drinkers are rational and, thus, if someone consumes a good, the private benefits must equate or exceed the costs.

“We would suggest,” write BERL, “that it is nonsense to argue that a drunk driver who wraps themselves around a power pole has made a fully informed, rational choice that is consistent with their long-term welfare and should be of no concern to society.”

The economists point out, as an example of Crampton’s “worldview”, that he had assumed drinkers would fare worse in the labour market, even in the absence of harmful drinking.

They also drew attention to his argument that “alcohol saves many more lives than it takes and has health benefits well beyond the point where BERL says harm starts and all benefits stop”, which they said was a value judgement that “would not have been appropriate for an independent study such as ours.”

Crampton rejects notions of being influenced by a particular worldview. He says the only “goggles” he uses to examine research are those of an economist.

All of this would simply be a relatively inconsequential, academic argument between economists if it were not for the political dimension: the Law Commission’s use of the BERL report in a speech suggesting higher liquor taxes.

From an economist’s perspective, so-called ‘Pigovean taxes’ are designed to internalise the external costs to society. If something has large social costs, there is a clear economic argument for taxation.

Recall Palmer’s comments in Nelson: “We have sufficient evidence to consider whether some of the costs isolated in the BERL report should be internalised to the liquor industry.”

Crampton claims – in arguments that have been echoed by anti-tax hike commentators – that “an underlying paternalistic argument [is being] covered in the garb of economics and presented as the result of the application of scientific economic method.”

He is annoyed that BERL has not publicly told the Law Commission that the research provided only a pure cost approach and couldn’t usefully be compared with tax revenue.

“While BERL does note in the report that their measure is cost only, it would be pretty easy for someone to miss that bit for all of its talk of ‘net costs’ and welfare.”

Since the public discussion started, BERL has not actually argued for its report to be used as a standard economic case for higher liquor taxes. Nor has it defended the Law Commission’s use of it.

Lead author Adrian Slack has commented: “In terms of forming policy… it gives you some direction on where are the biggest problems, where should we focus” and has noted that the report was not prepared for the Law Commission.

“As with anything that enters the public domain, it is the consumer’s right to interpret it as they see fit and for them to take responsibility for their reaction to it, not for the author to manage their response to it.”

At the Ministry of Health, Laurenson says that the report “makes an important contribution to understanding the cost to society of harmful alcohol consumption”, and it is using the report “to provide advice to Government on alcohol related harm and ways to minimise it.”

Crampton worries that government organisations “appear to commission reports with the purpose of lobbying” and argues that a ministry favouring higher taxes or a tougher regulatory regime would request a report that tallied all social costs, rather than just external costs.

And the Law Commission itself has stayed largely above the fray. Palmer recently told the National Business Review, however, that, on 22 May, he sought advice from Treasury and an independent economist on ‘issues’ in the BERL report. This, he pointed out, was long before the report by the two economists Crampton and Burgess (although it was actually after Crampton’s Press article and blog postings started.)

Palmer has recently expanded on his views about the economic case for liquor tax changes in the Law Commission’s public discussion paper. Its call for higher taxes on alcohol means this debate is only likely to get louder.

  • David Young is a freelance journalist.