Sir

The papers by Malte Meinshausen and colleagues ('Greenhouse-gas emission targets for limiting global warming to 2 °C' Nature 458, 1158–1162; 2009) and by Myles Allen and colleagues ('Warming caused by cumulative carbon emissions towards the trillionth tonne' Nature 458, 1163–1166; 2009) suggest that society could limit global warming below the widely discussed goal of 2 °C by adopting a cumulative budget for carbon emissions. Although they do underscore the difficulties, their prescriptions are only marginally relevant for policy design.

Solving the carbon problem needs international coordination. Success depends on many factors, but paramount is the credibility of promises that governments make to each other through international agreements. The trouble with the Kyoto treaty was that for pivotal countries, notably the United States, the promises were not credible. Correcting that error is a central aspect of negotiations before the climate summit in Copenhagen in December.

Credible promises will make most countries willing to do even more: a cycle of cooperation could unfold. Essentially, all successful international regulatory regimes evolve this way, starting with modest promises that, if kept, create confidence and credibility for greater efforts later on.

The problem with long-term cumulative targets such as those Allen advocates is that they cannot readily be codified into anything governments will find credible. They lack immediacy for policy if governments decide to leave costly actions to their successors.

This is partly why Kyoto's 'budgets' lasted only five years (2008–12). Nobody thought that was long enough, but it was expected to force action to smoke out credibility. (In the United States, alas, the effort failed.)

Global, cumulative emission budgets are nothing new. But they will never gain traction because a government must translate them into something it can control, such as shorter-term emission targets; it can implement these through 'cap-and-trade' schemes or other kinds of tangible policy effort, such as carbon taxes or regulatory programmes. At best, broad cumulative budgets are a general guide for policy. At worst, they distract the debate from what governments can actually achieve.

Your special issue of 30 April 2009 on 'The coming climate crunch' is also a timely reminder that the 2 °C target is a political delusion. Nobody knows what is safe — in part because the climate will be sensitive in unknown ways (as Meinshausen's paper shows) and also because safety depends on circumstances. There is no simple relationship between what governments can actually control and abstract goals such as a set limit to warming. Real outcomes might be plagued by interactions that doom the planet to warming of 2 °C (or more), whether or not emissions are cut. Even with a big dose of luck, the effort needed to get to 2 °C would be heroic, as Allen and colleagues indicate, and probably far beyond what real governments can achieve.

Being neither achievable nor safe, the target is becoming dangerous. The new papers are a reminder of how wrong-headed such goal-setting has become.