The headline-grabbing figure in the updated report, The Growing Cost of Obesity in 2008, is that the cost has tripled to $58billion, up from $21billion three years ago. This is a scary number and it's meant to make us think that the burden obesity is placing on taxpayers is about to bankrupt the health system. This number massively inflates the real cost of obesity to Australian governments.

The $58billion figure is compiled using the accounting tricks economists and lobby groups regularly use to make an alleged problem seem bigger than it is, and justify governments spending more taxpayer dollars on the alleged solution. The purported increase in the cost of obesity is not linked, as media reports have implied, to a verifiable increase in the number of obese Australians.

The Diabetes Australia report cites a rise in the obese proportion of the population from 16 per cent in 2005 to 17.5 per cent in 2008. But this isn't proof that the ''epidemic'' is spreading throughout the suburbs of Australia and escalating the cost of obesity. This is simply a revised approximation of the prevalence of obesity, based on what is widely recognised to be very poor data.

The real reason for an apparent rise in costs is that the original calculations are adjusted for inflation. The report also used a new methodology which links obesity with a higher percentage of the disease burden associated with Type2 diabetes, cancer, coronary disease, and osteoarthritis. But this does not mean that the real cost of obesity is now $58billion a year.

A whopping $49.9billion is accounted for by estimating the cost of lost wellbeing the cost of healthy life years lost by obese individuals owing to premature disability and death. A figure like this would never appear on a company's financial statement, because it monetises an opportunity cost that has no explicit monetary value. Estimating the non-financial cost of obesity is meaningless in relation to public policy. It's an imaginary cost.

It's a similar story in relation to the remaining $8billion of financial costs. About 44 per cent, or $3.6billion, is attributed to productivity losses another non-monetary cost.

Carer costs total $1.9billion and Dead Weight Losses from government transfers, which includes welfare payments and forgone government tax revenue, total $727million. Contrary to what we are led to expect, this leaves just $2billion in health system costs.

When the total financial costs are broken down, half are borne by individuals or families. Only $3.2billion is borne by the Commonwealth and State governments.

This is a considerable sum, but reassuring nevertheless in the context of alleged $58billion bill.

The bottom line is that the majority of the costs of obesity are private, not public, costs. The cost of obesity to Australian governments is small in the context of total government spending about 1 per cent of the $300billion Commonwealth budget alone. The health costs of obesity are more significant in our $60billion public health system.

But these costs still represent only 3 per cent of health spending. This puts into perspective the perennial complaint that only 2 per cent or so of government health expenditure is currently ''invested'' in prevention.

Despite the evidence, the Rudd Government has promised to spend millions more to fight obesity, for which it is unlikely to get value for money. Study after study has shown that anti-obesity initiatives fail to induce the long-term dietary and exercise changes required for obese people to achieve significant weight loss.

So let's hope the House of Representatives Committee inquiry into obesity, chaired by Steve Georganas, has its wits about it. It should reject the inflated guesstimates of the cost of obesity and focus on the real cost to the public purse.

Jeremy Sammut is a research fellow at the Centre for Independent Studies.