Published in The Australian Financial Review, Tuesday 29 March 2011

The carbon debate – like most political debates – is shaped by what people want to hear. Politicians tend to focus on the parts of the pudding we want to eat: more jobs and lower cost of living. But there are few free lunches, let alone magic puddings, in real life.

Choosing the best policy needs clear-eyed analysis about all of the real costs and benefits of the alternatives.

Carbon pricing inevitably means that some existing jobs will go. The whole point of carbon pricing is to restructure the economy to emit less carbon. Firms that innovate to produce goods and services with less carbon will reduce their relative costs, gain market share and add jobs; those that do not innovate will ultimately shed employees.

Similarly, standards of living will be lower in the short run. Spending more to reduce emissions, but produce the same amount of electricity, reduces productivity by definition.

Unfortunately, doing nothing is a very bad option. Reducing emissions is not going to be painless: the trick is to design policy to minimise the pain.

In any country, job losses cause pain. Governments are inevitably under political pressure to protect existing jobs. But attempts to shield existing jobs from carbon pricing – like most government assistance to protect existing jobs – will come at the price of less visible jobs and living standards elsewhere.

To take one of the most politically charged examples, preserving the 5000 jobs in Australia’s aluminium smelting industry will in effect cost about $160,000 per job per year. The “free” permits proposed for the industry will ultimately raise the costs that others have to pay to reduce emissions.

These other costs and jobs are more dispersed across the economy, and are harder to identify, but these jobs have families too. Of course, carbon pricing with free permits is not useless reform. It is just substantially more expensive for the general community than it needs to be. Governments can respond to the political pressure by pointing to the “green jobs” created as we reduce emissions.

These will no doubt offset some of the job losses in emissions-intensive sectors.

But at the end of the day, there is no magic pudding: with a net increase in the costs of production, average living standards across the economy will inevitably grow more slowly for a time. The most important policy question is how to minimise the total cost to all Australians.

One alternative is to pay directly for reductions. This avoids the immediate pain of imposing a new and visible tax. But taxpayers must ultimately pay for the reductions through reduced services or higher taxes. Even if the funds are identified out of “budget savings”, those savings could instead be used to increase living standards through lower taxes or better services. The community still pays, whether government taxes emissions or pays for reductions.

At a more sophisticated level, direct payment for emission reductions, unlike carbon pricing, avoids the churn of money from consumers to producers to the government and back through compensation payments. However, government funding to reduce emissions creates a substantial industry. In a direct payment scheme, government must identify potential reductions, tender for these, and then check on delivery. As with other rebate-type schemes, there is a real risk that taxpayers will pay for actions that would have happened anyway. Living standards ultimately bear these “deadweight” costs, in addition to the cost of real adjustment to cut actual emissions.

Is there any way for Australians to get some magic pudding?

We might get a bite or two if Australia quickly set up low-carbon-technology industries to export to the rest of the world. However, this is unlikely given the rate of progress elsewhere. Emerging markets (particularly China) registered 3000 low-emission-technology patents in 2006, almost half the total of the developed world, and this percentage is increasing rapidly. As other countries ramp up their response to global warming, it is getting harder for Australia to reach the forefront of any low- emissions industry.

There might be another bite of magic pudding in the push towards energy efficiency. Many people buy a cheap fridge even though it will cost more to run.

Carbon pricing, and publicity around carbon emissions, may result in people making wiser capital decisions that increase efficiency, reduce emissions, and cost less in the long run. But we cannot expect to reduce emissions to our targets purely through efficiencies.

Using the revenue from carbon pricing to reduce inefficient taxes, as recently suggested by Ross Garnaut, might provide a full slice of genuine magic pudding. Carbon pricing is a relatively efficient tax: it is cheap to collect and difficult to avoid because the levy will only be paid and accounted for by fewer than a thousand firms.

The tax burden is ultimately shared widely across the economy, roughly proportionate to consumption, because much of the tax cost will be passed on through commodities such as electricity and steel that are small inputs to a wide variety of goods and services. And the tax encourages taxpayers to avoid behaviour that we want to deter – emitting carbon. For these reasons, it is a “good” tax relative to many existing taxes such as stamp duty. If it replaced them, productivity would be higher.

Of course, we might be able to get the same productivity dividend by replacing these inefficient taxes with higher income, consumption or land taxes. However, the history of the past few years suggests we could be waiting a while.

Theory and practice both suggest that a straightforward carbon price is likely to deliver at lower cost to us all. And if we take the opportunity to replace some inefficient taxes, we might just get a taste of magic pudding.