"The global annual temperature for combined land and ocean surfaces for 2007 is expected to be near 58.0°F and would be the fifth warmest since records began in 1880. Some of the largest and most widespread warm anomalies occurred from eastern Europe to central Asia.
Including 2007, seven of the eight warmest years on record have occurred since 2001 and the 10 warmest years have all occurred since 1997. The global average surface temperature has risen between 0.6°C and 0.7°C since the start of the twentieth century, and the rate of increase since 1976 has been approximately three times faster than the century-scale trend.
The greatest warming has taken place in high latitude regions of the Northern Hemisphere. Anomalous warmth in 2007 contributed to the lowest Arctic sea ice extent since satellite records began in 1979, surpassing the previous record low set in 2005 by a remarkable 23 percent. According to the National Snow and Ice Data Center, this is part of a continuing trend in end-of-summer Arctic sea ice extent reductions of approximately 10 percent per decade since 1979."
--National Climactic Data Center, Preliminary Annual Report
I am worried about climate change. In one respect, I may be more worried than other people. I am worried because I have very little confidence that we know what is causing it. One of my fears is that we could reduce carbon emissions by some drastic amount, only to discover that--oops--it turns out that climate change is being caused by something else.
I am not a skeptic about the rise in average temperatures. Nor am I skeptical that the amount of carbon dioxide in the atmosphere has been increasing. However, I remain skeptical about the connection between the two.
My question is this:
what are the most persuasive reasons for believing that the rise in temperature is due to increased atmospheric carbon dioxide?
What I am looking for is evidence that I can use to overcome my skepticism. My view of climate change is that we have about three data points--an increase in temperatures from 1900-1940, and slight decrease from 1940-1970, and a recent increase. There are a lot of variables that could affect climate, and I wonder how we can be confident about our understanding of the process, given that we have only those three data points to work with.
I certainly am open to an argument that we have more data points to work with. I am just trying to explain where I am coming from.
Other Scientific Propositions
It seems to me that with other scientific propositions, I do not rely on a scientific consensus for proof. I am persuaded by other evidence. For example, I do not have a deep comprehension of the relationship between matter and energy. However, I find the atomic bomb a reasonably persuasive piece of evidence that Einstein was onto something.
In economics, many of us believe that economic institutions matter. We believe that prosperity is more likely in societies with systems that protect private property, encourage trade, and so forth. We do not know exactly which systems work best, a point which I made in Cracking the Code of Prosperity and which Dani Rodrik makes in his book, One Economics, Many Recipes. But many of us believe that institutions are a major factor.
If I were asked to supply the most persuasive reasons for believing that institutions matter in economics, I would site the following:
1. East Germany's decline relative to West Germany under Communism. For example, Japp Sleifer writes,
"Before the Second World War the East German economy had the signs of a blossoming landscape. At that time per capita national income amounted to 103 per cent of West Germany, compared to a mere 31 per cent in 1991. In the industrial sector labour productivity dropped from 91 per cent of the West German level in 1936 to merely 31 per cent in 1991. East Germany is a case of an economy that was relatively wealthy, but lost out in relatively short time."
A similar point could be made about North and South Korea. In both German and the Korean cases, a reasonably homogeneous society was split in two, with one half adopting a Communist system of central planning and the other half adopting a more market-oriented system. The outcome was that economic performance was better in the market-oriented system.
2. Workers in low-income countries can increase their wages dramatically by moving to high-income countries. See Lant Pritchett's book, Let My People Come. If it were simply a matter of American firms preferring low-wage Mexican workers to high-wage American workers, capital would be flowing across the border to Mexico. However, the same capital and labor is much more productive here than there, and this likely reflects better institutions in the United States.
3. Economic liberalizations tend to increase growth. We have seen this in China, in India, and under Margaret Thatcher in the UK. However, the evidence is not overwhelming. Skeptics point out that the liberalization in India was not particularly dramatic. Indeed the more dramatic liberalizations in some Latin American countries and in post-Communist Russia were not very effective.
Overall, the evidence that institutions matter is not as devastating, so to speak, as the atomic bomb. Still it is evidence, and I am willing to present the evidence and let individuals make up their own minds.
Recently, New York Times columnists David Leonhardt dubbed Shannon Brownlee's Overtreated the best economics book of 2007. I also spoke highly of it in my list of the year's best books and in a longer review.
A major thesis of Brownlee's book is that Americans undergo a large number of unnecessary medical procedures. I agree with this thesis. Evidence in its favor includes:
1. Studies by Dartmouth economist Jack Wennberg and colleagues showing that there is wide variation in intensity of the use of procedures across Medicare regions, with no difference in outcomes.
2. The RAND health insurance experiment gave similar patients different levels of health insurance coverage. Those who had more coverage elected to undergo more procedures--with little difference in outcomes.
3. Recently, economist Amy Finkelstein studied the impact of Medicare on health spending. She found a large effect on spending--but little effect on outcomes.
Change my Mind
Health care is an issue on which I have changed my mind. Before I began working on my book, my inclination was to believe that at some point we will see America's extravagant health care spending translated into better outcomes. Now, I have my doubts. I suspect that a lot of procedures are done for institutional and emotional reasons, without tangible medical benefits.
I am willing to change my mind about the role of carbon emissions in causing global warming. However, I would like to know what evidence other people find most persuasive. I would like to assess this evidence, given what I know about statistics and modeling. Feel free to leave your argument in comments, emails, or blog posts--use "Arnold Kling" in your post and my ego-surfing will pick it up. I assume that one of the reasons that people believe the carbon emissions story is that the evidence for alternatives is fairly weak. However, if someone thinks that he or she has good evidence for an alternative, feel free to let me know.