Chart In Focus
Post El Niño Cooling Means a Bottom Soon For Bond Yields
McClellan Financial Publications, Inc
Posted Nov 12, 2020
Nov 5, 2020
A peak in global average temperatures in 2015-16 is foretelling a bottom for long-term bond yields. But to understand that point, I need to explain this week’s chart.
It shows an inverted plot of Moody’s Aaa corporate bond yield, with proxy data for the years before Moody’s began that series in 1919. Also in that chart is a plot of the HadCRUT4 global average temperature series, published by England’s Hadley Climate Research Unit, at the University of East Anglia. One key adjustment is that the temperature data plot is shifted forward by 5 years in order to reveal how global temperature fluctuations tend to show up later in interest rates.
The top chart is a really long-term chart, going back all the way to the beginning of the HadCRUT4 series in 1850. It is tough for the climatologists to construct meaningful temperature data series before then, in part because decent thermometers did not widely exist. If you can look past the noise in both plots, you can see the larger trend directions match up pretty well. More on that below.
We can even see the correlation between the plots when we zoom in closer on just the past few years:
This chart employs the same two adjustments; the plot of Aaa yields is inverted, and the HadCRUT4 temperature series plot is shifted forward by 5 years. There is not a perfect correlation, which means that temperature fluctuations do not explain the entirety of price movements. But the correlation is really quite good, most of the time.
It got a little bit off track in 2000, when the Fed was pumping money into the banking system because of worries about the Y2K problem (remember that?), and thus helping to fuel the Internet Bubble. But shortly afterward the two plots got back on track again. One can find other moments in history when the correlation went off track in other ways, but the bond yields eventually get back on track and follow the message of the temperatures data.
Why that matters right now is that we are coming up on the 5-year anniversary of the 2015-16 peak of global temperatures that resulted from the strong El Niño cycle, which brought a big spike in temperatures, and accordingly also led “experts” to make predictions of the whole planet melting, and other assorted flavors of doom for all humanity. And 5 years later, bond yields are falling (rising in this inverted scaling) just as this model says should be happening. But soon we will get to the 5-year echo point of the end of that temperature trend, and the cooling which happened after that 2015-16 El Niño peak means that bond yields should start rising again.
Why does this happen? My understanding is that the relationship involves agriculture, since higher global temperatures typically result in greater crop yields, and thus lower food price inflation. That flows through eventually to lower inflation on other types of prices, and thus lower interest rates.
Here is a chart comparing the HadCRUT4 temperature series to U.S. corn crop yields:
There is no time offset employed in this chart. The two plots have both been rising for much of the 20th century, in part because better science on fertilizers and strains of corn seeds have helped boost production. But higher temperatures have helped, and so have higher CO2 levels, since CO2 is an essential nutrient for growing plants. Higher CO2 levels also help plants to more efficiently use water, so droughts are not as significant of a problem as they used to be. If you look closely at the chart, you can see that the cooling periods since the 1960s have been associated with dips in the annual corn crop yield figures.
The implication of all of this is that if we enter a cooling period, as the earth is scheduled to, that is going to mean a dropoff for crop yields, and an associate rise for interest rates, after the 5-year lag. That cooling period has arguably already started, in the years since that 2015-16 El Niño peak.
There is a 60-year cycle that shows up in interest rates data, going back as far as those data reliably exist.
That cycle says that interest rates were supposed to have bottomed in 2010, and started a 30-year rise to the next peak due in 2040. Interest rates are thus 10 years overdue for that upturn (just like they were in 1900). So why is it happening, that rates are staying low despite what this cycle says?
The answer is that temperatures have not cooled down like they are supposed to. Climatologists know that this same 60-year cycle exists in a variety of climate data, going back to before the modern temperature measurement era, using proxy data. So if temperatures do not cool down, for whatever reason, then crop yields don’t wane, and interest rates can stay low. But temps and rates can only borrow from that bank for just so long before they have to start paying it back, and soon we will be at that inflection point.
Look for rising rates from 2021 to 2024, and likely in the years beyond.
Editor, The McClellan Market Report
Subscribe to Tom McClellan's free weekly Chart In Focus email.
Copyright ©1996-2020, McClellan Financial Publications. All Rights Reserved.