But such a commitment would not have been credible. What emerged in Paris was in fact better, because the negotiators were able to agree on practical steps in the right direction. Individual countries pledged to limit their emissions in the near term, with provisions for future monitoring and periodic reviews of targets.

This is far better than setting lofty goals for the distant future while giving little reason to think that they would be met. The important thing is to get started.

In four key respects, the agreement is a good one for those who regard global climate change as an important problem and want to take feasible steps to address it.

First, and most important, participation is comprehensive, with 188 countries offering individual commitments, called intended nationally determined contributions (INDCs). In the past, only rich countries were expected to reduce their greenhouse-gas emissions; developing countries were explicitly spared that within the UN Framework Convention on Climate Change.

That had to change, partly because it is in developing countries, not the advanced economies, that emissions are growing the fastest. Furthermore, countries such as the US would not agree to limit their emissions if they feared that carbon-emitting industry would simply migrate to developing countries.

Second, the agreement includes a process of future assessment and revision of targets. Every five years, the parties will take stock and renew the commitments. Targets can be adjusted in light of future developments to be more or less aggressive, (probably more, if the scientists’ predictions are borne out). Negotiations on the INDC revisions are to begin in 2018, even though the first set of targets is scheduled to take effect in 2020.

Third, the Paris deal takes steps towards transparency in monitoring, reporting, and verifying countries’ progress. Starting in 2023, countries are to report every five years on compliance with their emissions targets. The US and Europe had to push China and India to agree to this. But without transparency, the INDCs would not be credible.

Fourth, the agreement contains mechanisms to facilitate international linkage, including scope for residents of rich countries to finance emissions reductions in poor countries. This is important because it is cheaper to pay a poor country to refrain from building new coal-fired power plants than it is to shut down an existing plant in a rich country. And achieving the first period’s INDCs at low cost will be an important determinant of countries’ willingness to take further steps in future.

Achieving more aggressive environmental goals, particularly limiting warming to 1.5ºC, or zero greenhouse-gas emissions in the second half of the century, would of course be desirable in terms of minimising the risk of disaster scenarios. In fact, the first INDCs, by themselves, are nowhere near enough even to limit warming to 2ºC (the global goal that was agreed in Cancún in 2010).

But proclaiming ambitious targets is very different from achieving them. It is almost besides the point that the economic cost of pursuing a goal of 1.5ºC would be very high. In any case, leaders can’t make credible commitments 35 years into the future. And the plan needs to be credible if it is to influence a business decisions made today.

Some developing-country leaders may be displeased for another reason: the figure of $100bn in finance from rich countries does not appear in the legally binding body of the agreement. The rich countries did admit their moral responsibility to help small island states, for example, cope with “loss and damages” from sea-level rise. But they rejected demands for formal acceptance of legal liability.

This was a reasonable outcome in a difficult situation. Rich countries can’t deny that their past emissions have inflicted harm on the world. In a domestic legal system, an entity whose land was, say, flooded would have a claim to compensation from the entity that had caused the damage. But sovereign countries are not operating in such a system. The $100bn in finance has always seemed problematic. The developing countries fear that the rich countries won’t deliver the money, at least not cash; and they are right. The rich countries fear that such “reparations” would disappear into the pockets of local elites; and they, too, are right. So it is better not to make promises.

In such ways, the Paris agreement ensures both fairness and efficiency. Achieving it was a daunting challenge, and more challenges lie ahead. But the negotiators’ success in converging on a plan that offers hope of practical progress is an unambiguous triumph.

Jeffrey Frankel is professor of capital formation and growth at Harvard.