How much will climate change impact the economy? I don’t know (Part I)

The default position for a true scientist is an exceptionally humble one.  They may come to their vocation with theories about the way the world works, but they will only retain those that can be supported through observation and experimentation.  The theories they hold will only ever be tentative: the scientist will perpetually try to find evidence to debunk ideas they believe to be true so that the stock of their own knowledge, and that of humanity, forever expands.

But it all starts with “I don’t know.”

What true scientists will not do is make unfounded assertions in the absence of evidence or make leaps of logic, like saying that if X causes Y then Z is inevitable.  They will use methods that have been carefully validated and whose properties have been thoroughly studied.  If the scientist develops a new method they will take steps to demonstrate its robustness and reject the technique if it is found to be wanting.

Which brings me to the current debate between physical climate scientists and economists.  It always shocks me when esteemed climate scientists, many who have conducted painstaking empirical work to support their view, leave their principles behind when they wade into the realms of finance and economics.  They seem to think that in these fields it’s okay to use untested methodologies, take leaps of logic, and make unfounded assertions.

This discussion relates to the extent to which climate change will cause economic damage and, subsequently, how much this will impact financial stability.  On this question it is extremely easy to form a range of theories and counter theories to explain why global warming may be highly detrimental to the human experience, and thus to productivity.  It is just as straightforward to hypothesize how these outcomes might be mitigated – or even seized for economic advantage – by the rat-cunning of humans who are able to adapt to the world they experience.  Pre-existing disparities in the distributions of income and wealth, both between individuals and nations, and associated power imbalances further complicate this picture.

The easy answer is to say “I don’t know” and let fate decide whose theories ultimately apply.  But the seriousness of the underlying issues dictate that we move on from mere theories and at least attempt to put some empirical meat on the bones.  If the evidence is too weak to confirm or refute any particular theory we must respond appropriately.  Our best model of the likely future impact of climate change may be of poor quality, but it can only be replaced if we offer new empirical estimates that can be shown to be superior.

The best estimate of the economic impact of climate change – or at least the one I will use as a straw man –  is that offered by William Nordhaus in a paper from 2018 and critiqued at length in an article by Keen et al from 2021.  Many of the points developed in the Keen et al paper were also used in a more digestible industry paper published in mid 2023   In Nordhaus’ own words climate change “leads to a damage of 2.0 percent of income at 3°C, and 7.9 percent of global income at a global temperature rise of 6°C.”  Nordhaus’ article takes great pains to state the assumptions on which the analysis is based, points out its many flaws and difficulties, offers avenues for future research and makes clear that the estimates are, at best, tentative in nature.  In other words, Nordhaus doesn’t know but he’s trying very hard to find out.

My aim here is primarily to critique Keen et al’s response to Nordhaus’ estimates and explain why their analysis is fundamentally unscientific in nature.  The key point being that they don’t put up a credible alternative to Nordhaus’ model.  As such, we have no way to determine whether the alternative proposed by the community of climate scientists (or a subset thereof) should take its place as the current best estimate of climate economic impacts.

They start by discussing the existence of tipping points and bemoan the fact that economists like Nordhaus ignore their potential damaging impact in the development of GDP projections under various warming scenarios.  But even if you fully accept the most dire descriptions of such phenomena provided by climate scientists, it remains difficult to see how these could be incorporated into an economic study that at least attempts to be rigorous and empirical in nature.  To do this you would have to be able to observe past tipping points, identify their relationship to observed economic outcomes – even if the relationship is a weak one – and then construct the relevant projections.

Of course Nordhaus has considered tipping points as a speculative exercise, but Keen et al disagree with the magnitude and probability of associated damage assumed by Nordhaus.  I don’t think the latter’s speculations can be received with any real confidence, but I don’t think Keen et al’s can be either.  Until we actually observe a tipping point, we will not know whether humans have the capacity to adapt to changing conditions and thus broadly maintain their standard of living.  

Keen et al point out that a range of studies “find that lethal temperature levels could affect areas currently home to 74% of the world’s population by 2100.”  This means that those living in such areas, and who lack the power to migrate, face catastrophic risks due to climate change.  This fact alone should prompt urgent action by all and sundry to curb their emissions.  But this does not change the fact that this group of people account for only a small share of global income and wealth and thus exert little impact on global GDP outcomes.

For the rest of humanity, scientists suggest that adaptation or migration to more hospitable climes will be possible even in a 6°C degree world.  Given this, the notion that overall economic damage will be limited cannot be ruled out using this data.

I obviously have several more comments on this topic and am already over 1000 words.  Let’s stick a fork in it and I’ll pick this up again next week.