Policy matters more than targets

26 April 2019

Australia's election campaign should be focused on the policies needed to tackle climate change, not the size of emissions targets, Warwick McKibbin writes.

The debate on climate change policy mistakenly focuses on the size of emission reduction targets. This is not the key issue. The key issue for climate policy is to reduce greenhouse gas emissions as much as possible (consistent with the body of science) at lowest economic cost.

Two countries with the same emission reduction targets can have very different economic costs. As shown in a 2019 Brookings study I co-authored on modelling the Paris Agreement, countries face costs from climate policy due to many factors including the structure of the economy, economic growth between now and the target years and the extent to which income is based on the endowment of fossil fuels.

We found that 80 per cent of the cost to Australia of the Paris Agreement comes from the actions of other countries. This is primarily due to reducing demand for fossil fuel exports from Australia. The Brookings report updates two earlier reports that I produced for the Department of Foreign Affairs and Trade in 2015 on the impact of the cuts in carbon dioxide that countries were proposing in the lead up to the Paris Negotiations.

The world has changed in the past four years, but the basic insights of the two earlier reports are still valid.

The first 2015 report focused on action by countries other than Australia. The second 2015 report explored the impact on Australia of different proposed target paths of emissions in Australia of 16, 26, 35 and 45 per cent below 2005 levels by 2030.

The actual costs of climate policy depend on where the economy would be in 2030 without policy intervention. That depends on assumptions of economic growth between 2015 and 2030 but also the amount of technical innovation, particularly in the use of renewables in electricity generation.

The uncertainty around this projection is large. The studies made assumptions about population growth and productivity growth and the model calculated projections for a wide range of economic variables. We assumed an efficient policy across the four targets. Given this assumption, the larger the reduction in emissions, the bigger the costs.

While the costs were significant - being 0.4 per cent of GDP for the 26 per cent target and 1 per cent of GDP for the 45 per cent target - they are a small relative to the Australian economy that was projected to grow by over 2 per cent per year over the period. The Australian economy did not plunge into recession as a result of any of the targets considered.

We then explored the importance of different assumptions. This is a key reason why a model is useful. It is not to calculate a precise cost. We don't know the precise cost in an uncertain world, but a model enables the estimation of a plausible range of costs under current knowledge and clear assumptions.

A key lesson from using models to evaluate policy is to NOT design the policy on the assumption that the numerical projections from a model are correct. The future is inherently uncertain and even if the model is the true depiction of the world, the assumptions fed into the model about the future might be widely wrong.

Believing forecasts was part of the mistake made in the design of fiscal policy under the Howard government where the China boom was expected to last forever or the design of the Rudd/Gillard climate policies where the world was assumed to have a high global carbon price by 2015.

Several key assumptions were tested in the studies. We found if renewables prices fell more quickly than assumed then the costs would be significantly lower - up to 30 per cent lower under reasonable assumptions. We found that if Australia could buy credits for overseas emissions reductions (called international units) at a substantially lower price than cutting emission in Australia, it was possible to reduce the costs of any target.

For example, a 45 per cent reduction target partly satisfied by buying international units, could make the costs similar to the 26 per cent reduction target without buying international units. If the world price was higher than the Australian cost, then these units would not be purchased and this policy would not change the costs.

However, it is an important safety valve for the Australian policy in case domestic costs are higher than estimated. To rule it out is bad policy. It is reasonable for Australia to pay its fair share to reduce emissions wherever it is cost-effective - either in Australia or in other countries.

Australia needs a comprehensive policy to deal with climate change. Piecemeal symbolic announcements do not make a comprehensive policy.

An emissions target is an essential element to guide expectations, but it is not the most important part of climate policy. What is most important are the actual policies. These need to be economy-wide because greenhouse gas emissions are economy-wide. To be low cost, they need to have a clear carbon price. They need to have wide support because as shown in the failure of energy and climate policy over the past decade, the cost of political uncertainty is likely to be larger than the costs of a well-designed climate policy. The policies also need to be scalable. If the global community pursues even deeper cuts, then the Australian policy may need to ramp up. Piecemeal policy will collapse under pressure.

The debate that should be the focus of this election campaign is not the size of the emission target but the comprehensive set of policies that will be implemented.

 This piece was first published in the Australian Financial Review.