Models For A Model Carbon Policy


We’ve heard an awful lot about models recently. We’re talking economic models here — which are essentially complicated spreadsheets. These are the models that economic forecasters use to predict the future shape of the economy, and the potential impact of various policy changes, such as a carbon tax.

Wayne Swan spent much of the week trumpeting unreleased Treasury modelling figures that proved the economic benefits of imposing a carbon tax. At a National Press Club address, Swan told the assembled media that "by 2020, national employment is projected to increase by 1.6 million jobs, while at the same time growth in domestically-produced pollution slows".

On Wednesday on Lateline, Swan argued that the modelling showed that "you can have strong growth, strong income growth, and strong jobs growth and still reduce carbon pollution substantially".

The modelling turned out to be a bit more controversial than Swan intended. The Greens’ Christine Milne wanted to know why the Treasurer was announcing confidential modelling to the Press Club when she hadn’t even seen it on the Multi-Party Climate Change Committee. Milne called it an "act of bad faith" and sniped that Swan was "stepping outside the agreed process."

Close watchers of the carbon tax debate will recall that the Government has set up a committee with key cross-benchers to work through the details of the carbon tax, and that some of the crucial inputs to the committee process are a series of special reports from the Treasury and the Productivity Commission on the international picture of carbon abatement policy.

Yesterday, the Productivity Commission’s report was handed down. Although complaining that they didn’t have enough time to do a thorough analysis, the Productivity Commission nonetheless examined (pdf) "more than 1000 carbon policy measures" in 10 of Australia’s key trading partners, including the US, UK, South Korea, Germany, Japan, India and China.

The Commission’s finding that a carbon price through an emissions trading scheme was the cheapest way to reduce carbon pollution echoed the findings of nearly every significant policy investigation into the issue undertaken in this country, including the Shergold Report prepared for John Howard, as well as the work of Ross Garnaut.

As the Productivity Commission states on page 30 of the report:

"Overall, key insights to emerge are that the European Union ETS has driven relatively low-cost abatement, where it has induced switching from coal to gas-fired electricity generation. Policies supporting renewable energy sources are more expensive, reflecting the higher costs of large-scale renewables production and particularly small-scale solar technology, which was found to be very expensive in all countries examined."

That’s broadly in line with the experience in Australia, where informed observers have long agreed that the really big impacts on carbon abatement will come from phasing out coal-fired electricity generation, which accounts for a whopping three-quarters of Australia’s electricity. In contrast, as the New South Wales government has recently discovered, small-scale solar rooftop subsidies have consistently been found to be very expensive ways to reduce carbon emissions, even if they make householders feel good.

The Productivity Commission report is another piece in the jigsaw puzzle that is slowly being assembled by the Government and the minor parties as they negotiate a carbon policy.

Throughout the debate, modelling has been a key weapon for both sides, with industry using a series of dubious reports to issue doomladen forecasts of tens of thousands of job losses, while green groups and the government have used their own modelling to proclaim a sunny future for jobs growth driven by huge investment in renewable energy industries.

Who you believe depends upon which side of the debate you fall into, and what you think about the models themselves. Showing his usual talent for irresponsible audacity, Abbott is today using the Productivity Commission’s report to claim that "the carbon tax ultimately spells death for the coal industry". By contrast, Tony Windsor is arguing that the report shows that "the most cost-effective way of dealing with the issue is through some sort of pricing mechanism, to price carbon".

Indeed, the very validity of the models being employed was raised by economist and Reserve Bank board member Warwick McKibbin this week, when he pointed out that the Treasury models didn’t take unemployment into account.

What? The Treasury doesn’t even model unemployment when looking at a carbon price? According to McKibbin, no. "I guess my surprise is when the Treasurer’s talking about employment impacts, I hope that they’re using a model in-house which doesn’t assume full employment," McKibbin told the ABC’s Ticky Fullerton. "Most of the climate models do assume full employment so there can never be an impact on jobs of a change in carbon taxes."

Although we can’t know for sure, because the Treasury methodology has not been released, McKibbin is probably right. Most so-called "computable general equilibrium" models actually don’t take unemployment into account — as this excellent paper on the history of equilibrium models by the New School’s Benjamin Mitra-Kahn explains (pdf). That’s because they’re based on quite unrealistic assumptions about the state and nature of a modern economy, assumptions which are generally adjusted to make sure the numbers add up correctly rather than because they bear any resemblance to the complex nature of contemporary economic reality.

And if anything highlights the complexity of the carbon debate, McKibbin’s intervention is it. We can’t really take McKibbin’s word as gospel here, because he’s pushing his own barrow. A well-known modeller himself, McKibbin has been using his high profile and, dare we say it, his public position as a Reserve Bank director, to spruik his own business interests — as a provider of economic modelling.

"I am getting a lot of interest in my views on how the world works and my computer models which are being used around the world, are getting a great deal of attention because they are producing insights which seem to be coming true," he told Fullerton later in the Lateline Business interview, making sure to get in a plug on taxpayer-funded airtime.

In fact, the insights of economic models have a decidedly dubious track record. As no less an authority than John Maynard Keynes observed in the 1940s, the uncomfortable truth is that economies are highly unpredictable. We don’t know what the Australian economy will look like in 20 years, let alone in 2050. Economic forecasters — including the Treasury — regularly get their predictions wildly wrong, as sceptical economist Paul Ormerod pointed out over a decade ago in his book The Death of Economics. For instance, throughout the boom years of the mid-2000s, Treasury regularly under-estimated how much tax revenue the Australian government would accrue. It then failed to predict the global financial crisis (along with most of the western world).

What we do know is the world will be much warmer. Climate models, which are based on rigorous physical observations backed up by the fundamental laws of physics, have turned out to be far more powerful predictive tools than their economic counter-parts. As a result, we are debating minor issues about the effect of a small tax on carbon while missing the bigger picture of what now looks likely to be a world that’s four degrees warmer.


Like this article? Register as a New Matilda user here. It’s free! We’ll send you a bi-weekly email keeping you up to date with new stories on the site. And you can like New Matilda on Facebook here.

Want more independent media? New Matilda stays online thanks to reader donations. To become a financial supporter of NM, click here.

Ben Eltham is New Matilda's National Affairs Correspondent.