Arnold Kling  

Climate Science and Macroeconomics

PRINT
Price Discrimination, Again... The Financial Crisis Narrative...

Nick Rowe compares and contrasts climate science with macroeconomics. After listing some similarities, he turns to differences.


D1. Macroeconomists are trying to explain people; climate scientists aren't. People are harder to explain. In particular, people's behaviour depends on what they expect to happen in future, so we have leads as well as lags in our causal modeling. That makes our job harder.

D2. I think we have more useful data. If we want to look at the effect of money on inflation, we can look at different countries, as well as different historical episodes. If they want to look at the effect of CO2 on temperature, they only have world data. And we have lots of natural experiments in historical memory. That makes our job easier.

Thanks to Mark Thoma for the pointer. Read Rowe's whole post.

The main reason I am a climate skeptic is (D2). My reasoning is very simple. Macroeconomists do not have enough data to verify hypotheses. (See my lost history paper.) Climate science has even less data. Therefore, climate science is even less reliable.

Before you protest that climate scientists have hundreds of years of data and many observation points, read the "lost history" paper. The point is that the information content of a seemingly large data set can in fact be quite low. Macroeconomists spent a good part of the late 1970's and the 1980's coming to terms with this (although some economists did not come to terms with it to the extent that I think is warranted). I think that climate scientists are unwilling to come to terms with it. That is why I am willing to challenge their expertise, even though I have only a superficial knowledge of the theories involved.

My sense is that another common feature of macroeconomics and climate science is that protagonists resort to bullying and ad hominem attacks relatively more often than in microeconomics or in other scientific fields.

[UPDATE] In my days as a macroeconometric model jockey, I often used "add factors" to make the equations fit the data better. But I never used them to distort the data. I disagree with those who think that "climategate" is a typical scientific brouhaha. This is at least one standard deviation away from normal academic behavior.]


Comments and Sharing





COMMENTS (13 to date)
Rich writes:

Tyler Cowen shared his thoughts on this subject a year or so ago too.

stephen writes:

The funny thing is, is that, climate science is just now getting to the same macro models now 40 years old. With similar results.

http://falkenblog.blogspot.com/2009/04/do-global-climate-models-contain.html

I had to laugh when I read your hypothetical retort: "climate scientists have hundreds of years of data". In geological terms this is like one financial quarter in last 100 years.

TDL writes:

And isn't much of the "hundreds of years" of data actually reconstructed as opposed to directly observed, further compounding the problems?

Regards,
TDL

Bill N writes:

Strictly speaking, the available data limits the ability to validate the theories. That is, they cannot compare models to experiment under multiple conditions. It is most interesting when a model can predict behavior under new conditions.

To a limited extend they can validate using known conditions from the past. Post-diction, however, is pretty easy to accomplish by adjusting your model and parameters until you get the answer you expect.

A potentially more serious problem is verification that the computer models actually behave as intended. Engineers and scientists are quite clever, but are no better at coding computer models than other professionals. Arguably they are somewhat worse than programming professionals. For comparison, Microsoft XP shipped with about 2 defects per 1000 lines of code and that is a factor of 3 better than industry norm. Do you want to place bets on how many defects reside in those climate codes and what their effect might be?

David C writes:

Uncertainty about climate change is an argument in favor of action, not against. That's because including more of the outlier estimates of the costs of carbon emissions increases the damage of climate change. Even if the estimates are off base, then that's still a far cry away from suggesting that carbon emissions have no impact on the environment. That's on a par with concluding that the buying and selling of government bonds and t-bills has no effect on the economy. Once somebody accepts that carbon emissions have an impact, then one must also take into account all reasonable measures of that impact equally. If one assumes that the standard measures have low accuracy, then you have to increase the propability that outlying estimates are accurate, and decrease the probability that mean and mode estimates are accurate. These outlying estimates increase the costs of climate change, they don't decrease it.

http://www.economics-ejournal.org/economics/journalarticles/2008-25/view

Snorri Godhi writes:

I disagree with those who think that "climategate" is a typical scientific brouhaha. This is at least one standard deviation away from normal academic behavior.

That implies the assumption that scientific fraud has a distribution converging to Gaussian. In fact, climategate is better described as a Black Swan.

In my days as a macroeconometric model jockey, I often used "add factors" to make the equations fit the data better.

A rather extreme view is that you can do whatever you want with the data, as long as you explicitly say, in the paper, what you did. Then, it is up to the reviewers and readers to decide if what you did is legitimate. Fraud is when you do not say what you did (and there are shades of fraud, of course).

manuelg writes:

> My sense is that another common feature of macroeconomics and climate science is that protagonists resort to bullying and ad hominem attacks relatively more often than in microeconomics or in other scientific fields.

So the lack of papers published supporting global cooling or non-human first-order causes of global warming is because of _hurt feelings_? And... let me guess: my fingers typing that last sentence was a form of _bullying_...

The malaise in macro is not surprising: really-existing macro policy funnels money to politically privileged groups. Papers published in macro that are out of step with really-existing macro policy will seem crankish because they have zero chance of being implemented. And - with the possible exception of pure mathematics - all sciences have this problem to a greater or lesser extent. I don't make the rules, I just don't forgive myself from noticing them.

I am sympathetic to the libertarian/economic criticisms of the scientific mainstream of climate science because the longer we defer expensive cures - within reason - the more likely we will have a more attractive menu of cures to choose from. The catch is "within reason". Another sticky point - when libertarians and economists choose arguments of very poor quality, they are less likely to be relevant to people outside of right-wing echo chamber. The United States climate policy will probably have to be attuned to the sympathies of the people who hold US debt - I wish that was predominately other US citizens, but it is not. Again, I don't make the rules, I just don't forgive myself from noticing them.

So, with the science of global warming having a first-order human cause, since I am not a researcher, I have to make a judgement: is it a nonesuch science, where the publishing practitioners cannot be trusted to draw conclusions from their own research, or we have an example of the critics practicing run-of-the-mill obscurantism, like in the fellow "controversies" of evolution, HIV/AIDS, the Jewish Holocaust of WWII, tobacco carcinogenicity, vaccination for public health, and the Twin Towers falling because of airplanes piloted into them and not because of controlled explosive demolition.

There I go with the bullying and the hurt feelings again. As penance, I will pray for the tender souls who wail for mercy in this big bad world.

But I am sympathetic to the libertarian/economic criticisms of the scientific mainstream of climate science, where it manifests as push-back against expensive quackery.

lukas writes:

Given the impact of the human race on climate, doesn't (D1) become less valid by the day? Once the ChiComs start geoengineering on a big scale, that one is definitely over.

Bill S writes:

From Jon Faust's "Whither macroeconomics? The surprising success of naïve GDP forecasts":

"I regularly hear the accusation that economic forecasting is no better than weather forecasting, but this does a disservice to weather forecasters. It is also an unfair comparison: weather forecasters have immense advantages over economic forecasters."

"Returning to the weather analogy, our results would translate something like this. We are forecasting temperature without knowing current temperature and having only a fuzzy estimate of temperature in the recent past. Further, we find that given a good estimate of recent temperatures, we do not know any systematic way to improve our temperature forecast using measures of other variables such as precipitation, barometric pressure, location of the jetstream, etc.
While necessary, forecasting real activity is a nasty endeavour. Our recent research confirms the basic conclusion of earlier work that the Fed's Greenbook forecast is excellent. No model we assess, however, including Greenbook, historically outperforms a naïve forecast based only on the best available estimate of recent GDP itself."

http://www.voxeu.org/index.php?q=node/901

Tim Lambert writes:

Considering the staggering amount of climate data there is, I am staggered that you can, with a straight face, claim that there isn't enough.

As for the crude adjustment for the divergence problem you linked -- you simply don't have a clue what it's about. Do you even know what the divergence problem is? It's not as if there has been some conspiracy to hide it.

Peter writes:

HELLO!! The level of CO2 in the atmosphere is now at its highest level in 800,000 years! Ice cores prove it. Only Alfred E. Neuman (and W. Bush) would say "What, Me Worry?"

Pete

bil. A. writes:

Uncertainty about arresting climate change is an argument in favor of political inaction, not against. That's because including more of the outlier estimates of the costs of limiting carbon emissions increases the damage of arresting climate change. Even if the estimates are off base, then that's still a far cry away from suggesting that politically/coercively limiting carbon emissions have no impact on social welfare. That's on a par with concluding that the buying and selling of government bonds and t-bills has no effect on the economy. Once somebody accepts that limiting carbon emissions has an impact, then one must also take into account all reasonable measures of that impact equally. If one assumes that the standard measures have low accuracy, then you have to increase the probability that outlying estimates are accurate, and decrease the probability that mean and mode estimates are accurate. These outlying estimates increase the costs of arresting climate change, they don't decrease it.

Miguel Bynum writes:

The secret to eternal youth is to be always in awe and to be always be open to learn new things. I'm 35 years old and I've recently learned how to fold 50 paper airplanes found at http://www.paperairplaneshq.com and it was a fun and challenging experience.

Comments for this entry have been closed
Return to top