Arnold Kling  

Global Warming Question

My Rejoinder to Michael Tanner... Global Warming Question, Con't...

What accounts for the relationship between total world GDP and world average temperature? For background, here is some data on GDP, from chapter 5 of Brad DeLong's Macroeconomics textbook (I did the multiplication to get the last column).

yearworld population (billions)GDP per capita (thousands of dollars)World GDP (trillions of dollars)

So, at a world level, we have been spewing out GDP at a much higher rate in the past ten years than we did even 25 years ago, and at a way, way higher rate than in the early 20th century.

I looked for data on global temperatures in this century, and I wound up, for better or worse, at this Wikipedia page. What it shows, I think (the scale on the graph is not explained clearly enough for a layman like me), is that temperatures rose somewhat linearly from 1900 to 1940, took a break from 1940 to 1980, and then rose at a comparable linear rate from 1980 to the present.

The reason I bring this up is this: if I look at the world GDP data, and I assume that emissions of greenhouse gases track world GDP, then the accumulation of greenhouse gases in the past 10 years should totally swamp anything up to 1980. So, I would think that we would see either no temperature change prior to around 1990 or else an enormous spike going on right now. I don't see any reason why temperatures should be following a straight line up from 1900 to today.

Another way to put this is that I think that if global warming is caused by humans, then we ain't seen nothing yet. I don't see how we can use the past as any guide to the future. That is not an effort to deny global warming--I just think in designing policies going forward we're taking a crapshoot any way you look at it.

Some possibilities:

a) The Global Warming alarmists are being way too conservative, perhaps because they have not grasped how fast GDP has gone up. In fact, temperatures are in the process of spiking now and will spike a lot in the next ten years--we'll see something worse than what the so-called worst-case scenarios project.

b) Emissions of greenhouse gases don't really track with world GDP. We're only spewing out a few more tons of greenhouse gases, even though world GDP has shot up. (But see the more link, below.)

c) There is something about the relationship between greenhouse gases and temperature that I am missing. That is, there is a mathematical way to reconcile a low level of emissions in 1900 and a high level of emissions in 2000 with a rate of temperature increase that is approximately the same.

I'm open to any possible answer, but if I had to guess right now, I think I would go with (b). But I would have expected a reasonably close relationship between greenhouse gas emissions and GDP. If that relationship is weak, then I wonder if there are aggregate variables that do track well with greenhouse gas emissions.

So I welcome pointers to papers that speak to this issue.

Of course, William Nordhaus is the go-to guy on this. I think figure 3-3 in this chapter gives me some relevant data. The blue column labeled "W" shows that world CO2 emissions relative to GDP have been declining at an average rate of 1 percent per year since 1975. But that only makes the global emissions rate in 2000 2.1 times the rate in 1975 instead of instead of 2.6.

It still suggests to my layman's way of thinking that a graph of temperature change over time should not even be able to show global warming as a 50-year phenomenon or a 100-year phenomenon. The temperature increases for the latest 10 years should be a huge multiple of any increase that occurred prior to 1950.

Comments and Sharing

COMMENTS (10 to date)
Rob writes:

A recent episode of NOVA discussed Global Dimming. The gist is that particulates in air pollution reduce the amount of sunlight reaching the earth. One theory is that this reduced heat input has been balanced by the increased heat retention caused by CO2. Now that particulate emissions are being / have been reduced without a corresponding reduction in CO2, it is thought that Global Temperatures are poised to skyrocket.

I bring this up as this theory may explain why Global Temperatures have not tracked with GDP in the way that you described.

Bob Hawkins writes:

The official explanation is that cooling due to anthropogenic aerosols has countered most of the warming due to anthropogenic CO2. (It's always about aerosols versus CO2. Thirty years ago, during the Global Cooling crisis, the idea that warming due to anthropogenic CO2 could counter most of the cooling due to anthropogenic aerosols, was rejected.)

Unfortunately, our actual understanding of the effects of aerosols is almost nil. A recent paper (The nature of light-absorbing carbonaceous aerosols, M. O. Andreae, A. Gelencsér, ) points out that, with the discovery of "brown carbon" aerosols, we are not even sure that anthropogenic aerosols cause net cooling.

spencer writes:

I suspect you need to find some way to adjust the gdp data for stages of development. Countries go through stages where in the early stages of development -- like China or India now -- they do not worry to much about pollution. But as they become wealthier -- like the US and Western Europe -- they start to reduce pollution so the volume of pollution created by a given increase in gdp starts to fall.

There is probably some good data on this, but I do not know where to find it. I would look at the OECD sources first.

Victor writes:

It would be great to have your analytical mind turn to this topic. Obviously read the IPCC reports ( I couldn't find the info. I recalled during a 30 minute lunch break investigation spurred by your post. But effectively, buried in any projection for global warming are estimates for (a) future world GDP, (b) future energy-intensity of future GDP, (c) future technological improvements. This is also true for retrospective matching to actual data.

My reading from several years ago left me with the sense that their economic arguments were sufficient for me to discount the human welfare impact of the climatological arguments.

A quick overview of some of possible criticisms are summarized at this link (which I wouldn't take at face value, just a Google, but some of the arguments struck me as plausible from my years-ago reading): .

As to your question, specifically, world temperatures supposedly lag CO2 levels by quite some time (maybe 80 years). Therefore, massive increases in GDP today, if that causes massive increases in CO2 emissions, will impact climate in the latter half of this century. Or thus say the GCM's (Global Climate Models). As peak oil hits -- and the necessary transition from oil in the 21st century is dealt with in the IPCC reports if my memory serves, yet another unknown variable -- the relationship between GDP and CO2 is also predicted to change.

That's probably quite enough blather from a 5-year old recollection on my part. Feel bad I'm not entirely up to speed on this ATM.

But I encourage you or anyone to read the IPCC reports and disentangle these issues for yourselves. If you've had some economics training, you'll see tons of issues that need to be addressed but seemingly are ignored. More than enough fodder for an educated person to really dig into, whether you believe the sky is warming or not. And doing so would, I'm sure, help to pull this debate out of the ignorant mire it is in today. I look forward to whatever you produce on the subject.

tom writes:


Click on the link below and look a quarter of the way down the website. There is a graph titled "Global Temperature and Atmospheric CO2 over Geologic Time". According to this chart, there doesn't appear to be a strong correlation (let alone causation) between CO2 and global temperature. And the chart shows that "the Late Ordovician Period was also an Ice Age while at the same time CO2 concentrations then were nearly 12 times higher than today-- 4400 ppm". Hmmm.

Another source showing the historical CO2 levels (See Figure 1).

Max writes:

There are two reasons for me not to judge your idea of linking GDP.

First, if human growth is also echoed in anthropogenic CO2 emission (more people means more CO2 breathed out!), then global warming will not be prevented by crippling economy.

Second, you assume that GDP and CO2 output are at least partly proportional, but I must disagree. Especially since the awareness to pollution from factories has increased, measures to eliminate CO2 emission in industry have also increased. So, I don't think we can directly correlated this. On the other hand, major CO2 "pollution" stems from airplanes and car traffic, so maybe there is a correlation between amount of people and GDP (meaning more car and plane rides).

As a student of engineering I must say that the industry is not the major cause of CO2 emissions, because it restricts itself pretty harshly (especially since the rise of the environment movement and the regulatory state).
The regulatory costs for not complying with environment regulation are even more costly than to buy those inefficient (economically speaking) devices like filters or alternate energy sources (gas instead of coal power).

So, perhaps those regulations have already reduced the temperature to a steady almost linear gradient?

mobile writes:

d) The natural variation in global temperature swamps the effect of CO2 emissions, at least at the prevailing emission levels for most of the 20th century. The changes in global temperature from, say, 1000 to 1900 have nothing to do with world GDP and anthropogenic CO2 output (which are approximately zero), so they probably have little to do with the temperature in 1900-2000.

Jody writes:

To me, the most obvious implication of that data is that over the last century global temperatures have been primarily driven by something other than CO2 emissions.

Maybe temperature is primarily driven by that big ball of gas in the sky.

Mr. Econotarian writes:

Here is CO2 versus US GDP:

Interestingly, CO2 per dollar of GDP is decreasing, and if trends continue, total US CO2 ouptut would begin to fall in 2014 and fall back to 1990 levels in 2018.

Fatih writes:

There is also one more link for US Data showing an economic snapshot on GDP and CO2.

Comments for this entry have been closed
Return to top