A new mechanism to consider when measuring climate impacts on economies

[A shorter (and more heavily copy-edited) version of this post was published in EARTH Magazine, read it here.]

My paper Temperatures and tropical cyclones strongly associated with economic production in the Caribbean and Central America was recently published in the Proceedings of the National Academy of Sciences. Because the paper is a little technical, here is a presentation of the results that everyone should be able to understand.

Following countries over time, years with higher than
normal temperatures during the hottest season 
(Sep-Oct-Nov) exhibit large reductions in output across  
several non-agricultural industries.
Central finding:
Economic output across a range of industries previous thought of as "not vulnerable to climate change" respond strongly to changes in temperature.  The data suggest that the response is driven by the direct human response to high temperatures: people generally are less productive and tire faster when it's hot.  This impact, which appears to be quite large, has not been factored into any previous estimates for the global cost of climate change.

Governments and organizations around the world are trying to figure out how much money we should spend to avoid climate change.  The answer isn't obvious.  On the one hand, climate change seems ominous and we'd like to spend lots of money to avoid it. But on the other hand, if we spend money on avoiding climate change, we can't spend it on other important things. For example, imagine that the United Nations has a million dollars it can spend. Should it spend it on building solar panels or building schools?  Both are clearly important. But if we want to get the most "bang for our buck," we need to figure out what the benefits of these two types of investments are.

A whole research industry has sprung up around the cost-benefit analysis of preventing climate change.  How much money should be spent to prevent climate change by investing in more expensive low-carbon technologies? Who should pay for it and when should they pay for it?  A tremendous amount of intellectual machinery has been applied to this problem by many extremely smart people.  The basic approach is to build models of the world economy-climate system and try to see what happens to the climate and the economy under different global policies.  These models are used by governments around the world to determine what they think the best climate policies are and how much they should spend on the problem.

However, there is something of a dark secret to this approach: we don't really know what will happen to us if the climate changes.  We have a fairly good grasp of how much it might cost to implement different energy policies. And we've learned a lot about how different energy policies will translate into global climate changes.  But when it comes to figuring out how those climate changes translate into costs to society (both financial and non-monetary), we end up having to do a lot of guesswork.

It's unfair to say we know nothing about the costs of climate change, but what we understand well is limited to certain types of impacts.  For example, we have been doing extensive research on the possible agricultural impacts for years. We've also done studies for a lot of the health impacts.  But most research stops there.  For example, we only are beginning to learn about the effect of climate on people's recreation and perceived happiness.  We're also only beginning to learn about the effect of climate on violence and crime.  We know a lot (but not nearly everything) about the effect of climate on ecosystems, but we don't really understand how ecosystems affect us, so we still can't estimate this impact on society. The list goes on.

Now we know a lot about climate impacts on health and agriculture because people have studied those impacts a lot.  Why did we study those kinds of impacts so much? I'm not sure. Maybe because the importance of climate on health and agriculture is obvious (eg. my plants on the windowsill died after just two days of this summer's heat wave).

The fact that we only really understand agricultural and health impacts of climate change is very important in the cost benefit analyses I mentioned earlier.  When governments are trying to figure out the best policies, they add up the known costs of preventing climate change and they add up the known benefits of preventing climate change.   If the costs outweigh the benefits, then that suggests we shouldn't spend much money to stop climate change.  But there is a natural asymmetry in this comparison between costs and benefits: we know all (or most of) the costs but only know the health and agricultural benefits.  So when we add up the costs of energy policies, the numbers tend to look very big.  But when we add up the known benefits of those policies, we add up the health benefit and the agricultural benefits, but we have to stop there because we don't know what else will be affected by climate change.  Maybe it shouldn't be surprising that many cost-benefit analyses find that climate change is not worth spending a lot of money on.

But what we know about climate impacts in non-health and non-agricultural sectors is slowly improving.  In a 2009 working paper, Dell, Jones and Olken did something very simple and got very surprising results.  They compared the economic output of countries over time with year-to-year changes in the weather of those countries.  They found that in poor countries, small increases in the annual average temperature of a country lead to large drops in economic output of that country.  The approach sounds simple, right? It is.  But the results are startling because they found such a large effect of temperature. They estimate that a 1C increase in average temperatures decreases a poor country's gross domestic product (GDP) by 1.1% in the same year. To get a sense of how big this effect is, recall that the economy of the Unites States shrank by 2.4% in 2009 and people are upset about the state of the economy.

Because the effect found by Dell et al. is so large, many people have been skeptical that it represents something real (note from my own unpublished work: I can corroborate their results using different data sets from the ones they used).  To check these results further, in 2010, Jones and Olken tried to looking for a similar effect in the exports of these countries and found that they also responded strongly to temperature changes.  Do people believe the general result yet? I'm not sure.  But part the skepticism seems to persist because its hard to know why poor countries should be so strongly affected by temperature.  One reason for this is that it's very hard to know what mechanisms are at work when one is only looking at macro-economic data.  Further, thinking of ways in which temperature affects economics this strongly and systematically across countries seems to be hitting the limits of many peoples' imaginations. This is where my study comes in.

There are two widely discussed reasons that Dell et al. may have found a large effect of temperature on economic output that might explain their findings:

(1) Maybe agriculture is so important to the economies of poor countries that temperature-induced changes in agriculture cause large macro-economic changes in the country.

(2) Maybe temperature doesn't really matter, but there is something else in the atmosphere (or ocean) that is correlated with temperature which matters a lot, but is left of our the analysis in Dell et al.

(A) An example hurricane that I model using LICRICE. (B) The
average amount of energy released by hurricanes and tropical storms
in a year, calculated using LICRICE. (C) The yield for the banana
crop in Guadalupe [green] and the cyclone energy released in
Guadalupe [orange].  It certainly looks like cyclones are important
for local economies (when cyclone energy is high, banana yields
tend to drop sharply). This is why I account for them in my analysis.  
The first point seems completely plausible.  But it doesn't necessarily agree with the findings of their study on exports, where exports of non-agricultural goods also decline when temperatures rise.

The second point is harder to think about.  We know that lots of phenomena are correlated with temperatures in the atmosphere (rainfall, droughts, hurricanes, tornadoes, clouds, blizzards, winds, humidity, pollution, etc).  Dell at al. try to account for changes in rainfall, but can't account for these other phenomena.

The design of my study deals with these two issues.  First, to deal with (1), I look at output in several types of industries (eg. construction, transport and communication, manufacturing, etc.). This allows me to compare the effect of temperature on agriculture to the effect of temperature on other industries.  Because agriculture is actually a relatively small part of the economy in the countries I study, any changes in agriculture due to temperature changes should have a relatively small influence on the other larger portions of the economy. [Note: This approach requires that I use a slightly different economic data set from Dell et al., but using a second dataset is useful since it confirms that the results of Dell et al. are not just an artifact of the data they are using.]

To deal with (2), I do two things. First, instead of looking at all countries in the world, I only look at a small group of countries that are climatically similar.  Then, because they are similar, I can figure out what the major types of phenomena might be correlated with temperature so I can account for them in my study.  In the Caribbean and Central America, hurricanes have a major influence on economics, so I spend a lot of effort trying to account for them. (I control for them using output from my LICRICE model [Limited Information Cyclone Reconstruction and Integration for Climate and Economics], which I won't describe here).

Even when I account for the economic impact of hurricanes and rainfall, I find that output in non-agricultural industries responds to annual changes in temperature.  The overall response I observe (a drop of 2.4% in GDP per 1C) is more than twice as large as the response found by Dell et al., which many people felt was "too large to be real"  (there are many technical reasons why they may have underestimated the effect, but I won't discuss them here).

When I compare the effect of temperature on agriculture and the effect on other industries, I actually find that the effect on other industries is larger. For example, the effect of temperature on output in wholesale, retail, restaurants and hotels is -6.1% per 1C, but it's only -0.8% per 1C for agriculture.  But the difference is even more stark is you compare the total number of dollars lost in agriculture compared to non-agricultural production: for every $1 in agriculture lost to increased temperature, $29 is lost in non-agricultural industries!  Now, it might still be possible that changing temperatures affected non-agricultural production only through its effect on agriculture [point (1) above], but it seems pretty unlikely given that the effect on agriculture would have to be amplified 29 times by some sort of internal dynamics of these economies.  What seems much more likely is that temperatures influence economic activity through some other (more direct) mechanism.

The estimated effect of temperature
changes only affects the year in which
the temperature actually changed (t).
As one might expect, production in
a given year is not affected by the
temperature the year before (t-1) or
the year after (t+1).
One plausible mechanism is the direct effect of temperature on human productivity.  Engineers who design air conditioning systems, ergonomists, our National Institute for Occupational Health and Safety and the United States Military have studied this effect for at long time and have a fairly established literature (there is even an International Journal of Hyperthermia), although it seems to have been completely passed over by all previous cost-benefit analyses of climate change.  At first, it may seem odd that economists haven't accounted for the effect of high temperatures on healthy people, since we are all very familiar with the effect: when its hot, we get tired easier, we work more sluggishly and we often try to work less so that we don't feel so hot or don't sweat so much (however, this issue is analyzed carefully in a recent working paper by economists Matt Neidell and Josh Graff Zivin).  But in our defense, it may not be obvious to us exactly how sensitive we are to high temperatures for two reasons: (1) most intellectuals probably do most of their work in climate-controlled environments, so our own output is relatively decoupled from temperatures outside and (2) most people are generally quite bad at assessing changes in our own productivity in response to subtle environmental changes.  (There are several famous laboratory studies where people are actually asked to assess changes in their own productivity due to environmental stressors; and it turns out that we consistently underestimate how much we are affected.)

Okay, so perhaps workers are less productive when it gets hot, even though this point was passed over by earlier economic studies.  Is there any way we can check this against the data?  Unfortunately, the data I'm examining is total output for an entire country, so we can't actually observe what's happening to individual people.  But there are two things that can be done.

First, we can check that the industries that are affected by temperature should be ones where humans (as opposed to machines or crops) play an important role in production. The industries that are most affected by temperature in the Caribbean and Central America appear to be "wholesale, retail, restaurants and hotels", "services", "transport and communication", and "construction".  The industries less affected are "agriculture" and "mining and utilities." "Manufacturing" seems to fall somewhere in the middle.  So it certainly looks like the industries that rely most heavily on humans are the ones most strongly affected.

Second, we can check that the structure of the response in these industries matches what we know about human productivity in the laboratory.  From laboratory studies of human performance under thermal stress (eg. putting typists in a room and counting how much/well they type at when the temperature of the room is adjusted) we know that small changes in temperature matter more when a worker is already in a hot environment (see figure below, panel B).  For example, we know that increasing the temperature from 24C to 25C doesn't affect workers as much as changing the temperature from 30C to 31C.  This kind of pattern can be looked for in the data, and it turns out that it actually shows up: it's the hottest days in the hottest season that are driving the observed reductions in output (see panel A below).

(A) The estimated response of production to daily surface temperatures during the hottest season (colors are the same
as the first figure in this post). (B) The response of individual productivity to temperatures as observed in over 150
previous laboratory-based studies.  You might notice that the axes on the two graphs don't have the same units, but
the two orange crosses in both graphs roughly correspond to one another.

Perhaps the most important take-away message from this analysis is that temperatures impact economies through a mechanism that has been completely omitted by previous economic analyses: human physiology and behavior respond to high temperatures.

The second take-away message is the effect might be large, maybe even very large, but more research will be needed to figure out exactly how important this mechanism is.  My analysis only examined the Caribbean and Central America, so it is difficult to know exactly how much the results should be applied to other regions.  The physiology of humans around the world does not vary dramatically (insofar as thermal regulation goes), but environmental conditions of working environments do.  So we should be careful when applying this idea elsewhere.

If the human response to thermal stress is included in cost-benefit analyses of climate change, how much will it change prescriptions for energy policy?  It's hard to say without doing more analysis.  But implementing global energy policies to transition to lower-carbon technologies is thought to cost less than 2% of GDP (even for the most aggressive policies proposed). So if economic losses on the scale of 2.4% of GDP per 1C (my estimate for C&CA) exist in other poor and hot countries around the world, including this response could have a substantial impact on our understanding of what the most cost-effective response to climate change might be.
All figures are courtesy-of and copy-written by Proceedings of the National Academy of Sciences.


  1. An excellent read, Sol! I will forward this to my colleagues!

  2. Really exciting. On the other hand, I think people may argue that the decrease in productivity from places getting too hot will be offset by the increase in productivity in places that were previously "too cold."

  3. very nice. i think the key is that human beings have learned to cope in (or optimized around) a relatively static version of their environment (whether hot or cold, to the above comment). this is something others have not successfully addressed (and which is likely why people have actively avoided the temperature topic altogether so as to avoid the neo-colonial implications). i think a great contribution you could make is driving home the point that it is the *change* that is important...perhaps even regardless of the direction or the starting point. (im thinking construction in new york city. whoa, did somebody say wtc? wouldnt that be something...)

    id be surprised if there isnt some good research out there on temperature change thresholds in the zone of physiological stress...at least as it affects other animals. that way you might be able to avoid questions about the ability of humans to adjust to what are still relatively gradual increases in temperature.

    quite thought-provoking. congrats!

  4. Talking with my dad about the paper he notes:

    "Back in the 70s when most of the government buildings still weren't air conditioned in New York City, they had a temperature-humidity index that was announced all day. Once it hit a certain number the city [note: he worked in a library at the time] just sent everyone home because they knew no one would be able to work anymore anyway."