David Kreutzer, an economist at the Heritage Foundation, is one of my favorite guys in the climate change debate. (I invited him to the IER conference on carbon tax that I organized.) Today he sent me his recent article in the Daily Signal reacting to the NOAA report that 2015 was the second-hottest US year on record (since 1895), a fact that produced obvious reactions from various carbon tax advocates.
Out of curiosity I clicked the links through to NOAA’s actual data set, and remarked to David that if we rank the years from 1895-2015 from hottest to coolest, then the top 5 years include 1998 and 1934, which is a bit odd if you think CO2 is the big driver.
That made me relate the issue to monetary and price inflation. Most economists would agree that the main driver for rising consumer prices is a rising quantity of money, especially over long stretches of time. It would be inconceivable that if you ranked the years 1895 through 2015 in terms of the level of prices (not the annual rate of price inflation, mind you, but the level), that two of “the most expensive” years would be 1998 and 1934.
At this point I really wanted to waste some time, so I decided to generate some graphs to illustrate my point. First let’s do monetary vs. price inflation. If we go from 1919 through 1992, it looks like this:
Note that in the above, I chose the monetary base (rather than M1, M2, or some other potential aggregate) because that’s the series FRED had furthest into the past.
It’s a pretty good fit, isn’t it? Of course, it’s not perfect, and if you calibrated a computer model with the above, you would’ve been caught flat-footed when the huge expansionary of the monetary base since 2008 didn’t generate comparable increases in CPI:
(Ideally I would have kept the tight fit of the two lines up through 1992, and then had the blue line explode upward while the red continues its gentle increase, but I couldn’t figure out how to make FRED do that.)
So, if we want to make an analogy with the climate debates, we could imagine some poor economists who had been warning of price inflation being perplexed by the above. They might speculate that the excess reserves went into the ocean, and would show up eventually in higher consumer prices. (That was Cliff Asness’ awesome tongue-in-cheek response to Krugman.) Yet even so, given the pretty tight correlation in the past, and the unusual nature of events since 2008–not to mention investors thinking that perhaps that monetary base will eventually be sucked out of the system–you can see why the inflation hawks would not think their worldview had been utterly smashed.
Let’s turn now to the issue of atmospheric carbon dioxide concentrations and (U.S.) temperatures. Here’s the chart relating the two, though I caution you that scientists only have direct observations of atmospheric CO2 from 1959. Prior to that, they estimate it using ice core samples. In any event, here you go:
In case you’re wondering why I started at 1904, the answer is simple: Excel didn’t recognize it as a valid date if I went back to 1903.
Perhaps I’m biased, and you can try putting it in Kelvin to see if that changes things, but that doesn’t look like nearly the no-brainer connection as we had with money and CPI. For example, the U.S. average temperature in 2013 is about the same as it was in 1910, even though CO2 concentrations were up almost a third. Average temperature in the 1930s was 0.3 degrees higher than in the 1980s, even though average CO2 was 12 percent higher in the 1980s.
Before closing, let me reassure my scientifically literate readers that I have read lots of the IPCC material. I understand full well how a climate scientist who is alarmed about anthropogenic climate change would explain the above. But my point is that even though in the past I have whimsically made a comparison between people (like me) who warned of large consumer price increases and those (like Krugman) warning of large temperature increases, I think the people in my camp have been much more reasonable about it.