Thursday, July 14, 2011

Don’t base policy on crapulous alcohol statistics

Alcohol. The anti-alcohol lobby say just one drink increases your risk of cancer, and news yesterday was that cigarette-warning-style labels will start appearing on bottles of booze. The social costs of alcohol are often cited as an additional reason to crack down on it. Here, Dr Eric Crampton casts a sceptical eye over how that social cost is measured

If I told you that surfing cost the Australian economy a billion dollars and that we consequently should make life jackets compulsory, you could be forgiven for thinking that the number represented some real cost to the community; perhaps the cost of rescuing surfers caught in rips or medical care for those injured in accidents.

But if you found out that the vast majority of that figure was the combination of surfers’ expenditures on their boards and the costs of holidays they took heading up to Yallingup, you might think twice about endorsing the policy recommendation. And you might wonder a bit why anybody would have thought those costs could matter for policy.

And so it is with the often-cited social costs of alcohol. In 2008, Macquarie University’s Professor David Collins and the University of Queensland’s Associate Professor Helen Lapsley estimated that alcohol consumption cost Australia some $15 billion for the 2004/5 fiscal year.

The figure proved popular, being cited dozens of times on Australian television and radio, more than 100 times in newspaper articles and op-eds and it’s been referenced on a dozen occasions in Australia’s Parliament by nine MPs. With one large number proving popular, others like Professor Robin Room went on to produce even larger figures.

But what isn’t widely understood is that these papers use a method that, like the surfing example, mostly counts the costs drinkers impose upon themselves. This “Cost of Illness” method varies considerably from standard methods of economic analysis. There are many ways of producing a cost figure, but once we stray too far from standard economic method, we cannot interpret the produced figure as being economically meaningful.

There has emerged a cottage industry of academics and consultants using this Cost of Illness method to produce large figures for Ministries of Health from Canada to Australia. When the method counts over a billion dollars that drinkers spend on their own alcohol in Australia, but dismisses that those drinkers just might have enjoyed having a drink, the practice borders on advocacy.

In its final report which was publicly released in May last year, the Henry Tax Review argued that alcohol tax is a blunt tool for addressing harms caused by alcohol, but that tax can address these spillover costs where marginal social costs otherwise exceeds the price of alcohol.

But methods that include private costs as public provide a distorted picture of the appropriate level of taxation; the Henry Review urged caution about relying on these methods used by Professor Collins and Associate Professor Lapsley. As the $15 billion figure includes a large proportion of costs which fall on the drinker, the figure is meaningless from an economic and policy perspective.

Together with Matt Burgess of the Institute for the Study of Competition and Regulation and Brad Taylor from Australian National University, I have had a close look at the $15 billion figure and compared it to the figure that would have emerged had more standard economic method been used in its construction.

The end result was that policy-relevant annual costs of alcohol consumption in Australia were no more than $3.8 billion. To provide perspective for examination of spillover costs, the total tax take from alcohol (excluding GST) for the same financial year was $4.1 billion.

In other words, alcohol consumers contributed more in tax than they cost others through the health system and traffic accidents. The new $3.8 billion figure is not the result of cost-benefit analysis – more work would need to be done to quantify the benefits of drinking. Rather, it is the figure that emerges when we apply standard economic method and count only the costs tabulated by Professor Collins and Associate Professor Lapsley that truly count as spillovers.

Unfortunately, political discussion of alcohol’s sometimes tragic costs is too often clouded by emotion. It’s not that we do not care when a drinker falls ill or, worse still, loses his or her life.

But if we are going to count the costs that some drinkers impose upon themselves through their consumption, a fair analysis would also have to count against those private costs the benefits that drinkers enjoy from a night out or from a quiet drink at home. Or, it would not present costs drinkers impose upon themselves as being costs to society.

On Thursday, I will be presenting to the Australian Conference of Economists in Canberra as part of the Conference’s annual “Dodgy” awards. I’ll be arguing that preventative health suffers from the worst application of economic analysis to policy, Professor Collins and Associate Professor Lapsley’s report and how it’s been treated in policy discourse will help me to make that case.


Doubts over Nestle hypoallergenic baby formulas

BABIES who drink hypoallergenic formula milk are no less likely to develop allergies when they are older than other children, a study has found.

The finding by researchers in Melbourne raises questions about whether parents should ignore health guidelines recommending whey-based hypoallergenic formulas for babies considered at risk of developing allergies.

Tests were carried on more than 600 babies with a family history of allergies to see if giving them hypoallergenic formula helped prevent them from developing conditions including eczema, asthma and allergic rhinitis when they were older.

The babies were given either a partially hydrolysed whey (pHWF or hypoallergenic) formula made by Nestle, cows' milk or soy formulas after their mothers stopped breastfeeding.

Each had skin-prick allergy tests - for milk, egg, peanut, dust mite, rye grass and cat dander - when they were six, 12 and 24 months old. Follow up assessments were then made when they were six or seven years old.

But the researchers found those babies who drank the hypoallergenic formula were just as likely to develop allergies as children who drank the cows' milk or soy formulas.

Adrian Lowe, a research fellow at the Murdoch Children's Institute who led the study, said the findings contradicted widespread advice on hypoallergenic formula and suggested breastfeeding could be more beneficial in preventing allergies.

"A number of guidelines recommend that when babies are being weaned from breastmilk, if the parents are worried about allergies, they go on to these hypoallergenic formulas," he said. "But our study showed it doesn't really make any difference.

"Families at high risk of allergies should be encouraged to breast feed for the many known benefits associated with breast feeding, and if parents need to move on to formula that that decision need not be based on the belief that one formula will reduce the risk of allergies compared to another."

Allergic diseases have doubled in western countries in the past quarter-century. The most common conditions in children are food allergies, eczema, asthma and hay fever. The Australasian Society of Clinical Immunology and Allergy estimates that up to 40 per cent of children in Australia and New Zealand are affected by allergy disorders.

It recommends that mothers who cannot breastfeed and who are worried about their babies developing allergies use hypoallergenic formula in the first four to six months of their baby's life.

Nestle, which sponsored the study, defended the effectiveness of its hypoallergenic formula, saying several studies including a large trial carried out in Germany, had shown significant benefits. "There have been at least eight different meta-analyses conducted and all conclude that HA is recommended for reducing the allergy risk in high-risk infants," a spokeswoman said.

"Having said that, while partially hydrolysed whey-based infant formula may benefit children who are not breastfed compared to cows milk formula, Nestle firmly believes that breast feeding is always best for babies."


1 comment:

Wireless.Phil said...


Wait to you see the article from 2010 in the right column at the page below: "Light to Moderate Drinking Linked to Fewer Heart Problems in Male Bypass Patients, Study Finds (Nov. 15, 2010)"

Low amounts of alcohol have different effects on left and right ventricles of the heart

ScienceDaily (July 15, 2011) — Few studies have examined the acute effects of alcohol on myocardial or heart function.