Phantom numbers

Today’s Herald reports that the NSW Treasury has done its own estimates of the costs of achieving various targets for carbon emissions.

The NSW Treasurer, Michael Costa, said it would cost $430 billion to reduce greenhouse gas emissions by up to 80 per cent as outlined by Ross Garnaut, and the sharp reductions proposed would impose a substantial cost on the economy. The Rudd Government has consistently said its target for 2050 is 60 per cent. The 80 per cent figure was predicated on other nations making similar cuts. In a hard-hitting speech to a business audience last night, Mr Costa said that based on NSW Treasury modelling, the proposed sharp reduction in greenhouse gas emissions would wipe 4 per cent off the size of the Australian economy over the next 20 years. This is significantly larger than the 1 per cent economic impact signalled by earlier reviews such as that by the British economist Sir Nicholas Stern.

Now, Michael Costa’s is not necessarily the most credible voice in this debate. In the first place, he is notorious as the highest ranking global warming denialist in political power. This is not on the basis of any knowledge of the science, as he candidly admits, but rather on the basis that there are power stations and coal mines in his electorate. Secondly, he intends to privatise the electricity sector and wants the prospective owners to be guaranteed carbon credits, which will raise the sale price in a world where commercial prospects for coal-fired power generation are gloomy. He will do anything or say anything — and it wouldn’t surprise me if he candidly admits this as well, in the Brechtian theatre of NSW politics — to persuade people that the burden of mitigation will be onerous.

But setting this aside, what are we supposed to make of these numbers?

Stern stressed that there was a wide range of estimates from different models of the cost of reducing emissions by 60%-75% by 2050; but the one that got the most publicity was ‘1 percent of GDP by 2050’ This implies a reduction in the annual growth rate by a minuscule 0.025 (one fortieth) of a percentage point over the next 42 years, say from 2.5% to 2.475% as illustrated in the table. But the Herald story refers to ‘the next 20 years’.

Supposing for argument’s sake that the NSW Treasury actually meant 4% (four times higher than Stern’s 1%) by 2050 — the same horizon — rather than 2028, then their estimate would in fact be similar to the one in the report of the Allen Consulting Group, which assumed a reduction in the annual growth rate by 0.01%. (There was some pointless controversy about this chez Quiggin a while ago.) Unlike Stern’s, this was an estimate for Australia specifically, and though the costs were higher than that well-publicised Stern figure, they were still surprisingly modest.

But if they really meant 4% by 2028, that implies a reduction in the average annual growth rate of more like 0.2%. In that case, as the table shows, the cumulative losses would indeed amount to about $430 billion. They would, that is, if the discount rate chosen was 2%, which is definitely on the low side, at least by the standards of Treasury modeling. Of course, the only reason to make this calculation is that it seems like a very big number, and presumably Costa thinks he can scare somebody with it. But unless he does a similar calculation with respect to the cost of global warming under Business as Usual, it’s all pretty meaningless.

Perhaps we’ll eventually find out what the scenarios were. I suspect that the reason Costa didn’t supply any details is that the exercise was pretty unsophisticated. But I always wonder whether journalists reporting this kind of thing have any idea what the numbers in their stories mean, or any expectation that the readers will.

These figures are based on the assumption that current GDP is $1000 billion. The growth rate of 2.5% under Business as Usual is just illustrative, based on the long run average, and ignoring the economic costs of the current emissions trajectory.