Why The Yield Curve Is Not Broken
At the beginning of November 2023, I wrote an article titled Now, Where Is Your Recession? and in this article, I already raised a question that still seems relevant at this point: despite expecting a recession, the U.S. economy is still performing strong. The prediction of a recession is often based on the inversion of the yield curve but since the yield curve inverted a long time has passed (over two years since the first sign). In my previous article I wrote the following conclusion:
I personally expect the market to crash to about 3,200 points in the next few months (after we see a small recovery before). And this is only the first target on our way down.
This was clearly wrong, and the recovery lasted much longer than anticipated. And of course it was not a small recovery, but a rally leading the S&P 500 about 28% higher in the last seven months.
And considering the huge rally we saw in the last few quarters in U.S. equities we must focus on two questions that beg for an answer: Is the yield curve inversion still a reliable indicator for a recession? And: Did we just miss out on the buying opportunity of a lifetime as the recession and bear market won’t come? Let’s start by taking a closer look at the yield curve inversion as recession warning signal.
For starters, no indicator is perfect and looking only at one indicator is always risky. Nevertheless, I would argue that the inverted yield curve is a good and reliable indicator to predict a recession in the United States. In my case, I would look at two metrics for the inverted yield curve:
I usually look at both and when both metrics are below zero, we can define the yield curve as inverted.
And we can mention 3 reasons why the inverted yield curve is a good indicator:
And as we can see in the chart above, both metrics are clearly below zero indicating another recession and giving a clear warning. The problem is that the United States economy does not seem to be in a recession so far and the time since the warning signal was given is extremely long raising the question if we made a mistake. Let’s look at the data since 1980 and the six recessions that were correctly predicted by the inverted yield curve.
Recession |
10year/2 year inversion |
Duration |
10year/3month inversion |
Duration |
Begin recession |
1980 recession |
August 1978 |
17 months |
No data |
No data |
January 1980 |
1981 recession |
September 1980 |
10 months |
No data |
No data |
July 1981 |
1990/91 recession |
December 1988 |
19 months |
March 1989 |
17 months |
July 1990 |
Dotcom Bubble |
February 2000 |
13 months |
July 2000 |
8 months |
March 2001 |
Great Financial Crisis |
January 2006 |
23 months |
January 2006 |
23 months |
December 2007 |
2020 recession |
August 2019 |
6 months |
March 2019 |
11 months |
February 2020 |
This time, the 10-year vs. 2-year treasury inverted in April 2022 and the 10-year vs. 3-month treasury inverted in October 2022. And this is leading to 26 respectively 20 months since the yield curve inverted. When looking at the data above it never took that long between the first inversion and the actual beginning of a recession. However, before the Great Financial Crisis both metrics inverted 23 months before the start of the recession which is still similar to the situation right now.
And of course, we also must point out that recessions require two consecutive quarters of GDP declining and therefore recessions can only be declared in retrospect. This means we could already be in a recession without knowing it.
When looking at past data, the inversion almost always ended before the recession began and right now both metrics is still below zero meaning the yield curve is still inverted. In my opinion we won’t see the recession before the yield curve is not back to “normal”. And at this point I also don’t have any explanation why the yield curve is inverted for such a long timeframe. One reason could be – but this is clearly speculation – that the steeply inverted yield curve is a hint for a severe recession (or depression).
But as outlined in several articles, I am expecting the United States being very close to the end of a long-term debt cycle making a recession or even a depression very likely. We have countries rather shifting from internal order to disorder, we have hot and cold conflicts between states increasing and we have the long-term debt cycle being close to the end which is illustrated by most assets trading for extremely high valuation multiples and the federal funds rate being close to zero (until recently).
For further information, I wrote about the long-term debt cycle in my last article about Moody’s Corporation (MCO) and I also wrote about the bigger picture in my last article why gold might be a better investment than equities.
But one indicator is certainly not enough. Even if we might be sure to have found the perfect indicator, we should not forget that markets are a complex system and especially a system with feedback. Market participants are influencing the stock market and the economy with their decisions.
Based on the sentiment of market participants they make buying and selling decisions – for products, services and assets. And these decisions impact the economy and stock market. And even if we would have found one indicator that is perfectly showing us when a recession will happen other market participants will see this indicator as well an act based on this indicator. And these actions will have consequences (intended as well as unintended) that might either reinforce consequences or maybe lead to some indicators stop working.
The best indicators would be those that are independent from any humans and human decisions, but in financial markets and the economy such indicators don’t exist (in my opinion). The economy (and stock market) is a social system and within such a social system individuals act and make decisions which determine in which direction the system is moving.
Therefore, we should look at other indicators, which we did in the past and will do again in the following section.
When looking at other metrics that could also tell us about the state of the economy and maybe an upcoming recession, we can make up several categories – including the labor market, major investments like houses, bankruptcies and the credit market. Let’s start by looking at the labor market.
First, we can look at the labor market. People losing jobs, being unemployment or having difficult to find jobs paying a solid salary is a problem for the economy. But the labor market seems to be in great and in many countries around the world – including the United States and Germany – companies rather seem to have problems to find qualified workers for many occupations.
We can start by looking at the unemployment rate, but it is not really a leading indicator as it takes time for the unemployment rate to rise. Nevertheless, from its low of 3.4% in early 2023 the unemployment rate rose to 4.0% in April 2024. But this is still a very low rate and no reason to be concerned.
A better early warning indicator would be the initial claims for unemployment. And in the last few months the number has been rising. But when looking at the numbers of the last few years (I am excluding 2020 as it would mess up the entire chart) the numbers have been rising in the recent past, but this is also no reason to be concerned and should still fall under the category of normal fluctuations.
We can also look at housing and construction as those numbers could be early warning indicators as well. Buying a house or a flat is a huge investment for most people and the most expensive asset they will buy during their lifetime. And these buying decisions are usually not made when times are tough, and people must rather cut their spendings – instead these are the first investments that will decline when times are tough. Housing permits and housing units under construction can be seen as two early warning indicators for the situation turning. We can also look at the housing market as the housing market is not only one of the major markets and has the potential to tank the economy (as we saw during the Great Financial Crisis).
When looking at the new privately-owned housing units authorized in permit issuing places, we see the numbers continuing to decline in the last few months after the numbers stagnated since late 2022. Of course, we are still above the lows from December 2022, but I expect new lows in the next few months.
We can also look at the new privately-owned housing units under construction and see the numbers now continuing to decline. But we are far away from a steep decline – compared to a high of 1.711 million in October 2022 the number is now 1.616 million.
In my article in November 2023, I mentioned bankruptcies as another important indicator. Especially in November 2023 and January 2024 the number of bankruptcies was rather low and in January 2024 “only” 35 corporations failed for bankruptcies. But since then, the number increased again, and we are still at a high level. Year-to-date (until April 2024) 210 companies filed for bankruptcy compared to 224 in the same timeframe in 2023 (but last year was the highest number since 2010).
A special category of bankruptcies are failing banks. And in the meantime, another small, regional bank in the United States failed. In April 2024 the Republic First Bank failed but had only $6 billion dollars in assets, which is barely worth mentioning in the banking world.
And while banks completely failing are the worst potential scenario, we should also look at loans and delinquency rates as they have the potential to tank the economy as well. Here we are often looking at the residential mortgages as they make up about 21% of total loans (according to data from the FDIC).
And in general, we can differentiate between consumer loans and commercial loans.
Consumer Loans
Among the consumer loans, the most important category of loans, which make up 70% of total consumer loans and have a volume of $12.4 trillion, are residential mortgages. And due to this massive volume, it is not surprising that residential mortgages are closely watched. But according to data from the New York FED, mortgages have a delinquency rate (90+ days delinquent) of only 0.6% right now. Compared to a peak of 9% during the Great Financial Crisis there seems to be no reason to worry.
And when looking at the mortgage originations by credit score, we see that the subprime category (which contributed heavily to the Great Financial Crisis) is hardly playing a role today, which is a good sign. During the last few years, we saw a high origination volume of mortgages, but most of the lenders had a credit score of 760 or higher.
The picture gets a little different when looking at car loans, a second major category among consumer loans. Here we have a delinquency rate of 4.4% which is still below the high of 5.3% set during the Great Financial Crisis. But here the delinquency rate is rising and when looking at the originations by credit score, we see much more loans with a low credit score (compared to mortgages) which is increasing the risk for default.
But we can also argue that the credit quality of auto loans got higher in the last few years. While the total amount of auto loans originated increased, the amount with a credit score below 620 stayed more or less the same. The following chart is also underscoring the fact that auto loans credit quality got higher.
Another important category of loans are credit card loans, which make up about 9% of total loans. And here we see delinquency rates rise since mid-2022. According to the New York FED, we had 10.7% of loans being 90+ days delinquent (at the GFC peak it was 13.7%).
Although mortgages are the most important category and these loans are holding up quite well, we see the picture getting worse and worse from quarter to quarter. Granted, it is still a process that is rather slow but when headed for a cliff I will fall even if I drive slowly over the cliff. And delinquency rates rising and rising is not a good sign but rather the driver moving towards the symbolic cliff.
Commercial
Aside from consumer loans there is a second major category we should not ignore – commercial loans. And some are arguing that the major problems this time are not consumer loans (or mortgages) but commercial loans – especially commercial real estate loans. And when looking at headlines in the last few months commercial real estate loans seem to be a massive problem – with warnings even from the FED.
And when looking at data from S&P Global for the first quarter of fiscal 2024 we see delinquency rates for commercial real estate still rising to 1.25%. While this represents a new cycle high, S&P Global is pointing out that the increase is decelerating with an increase of “only” 11bps compared to the previous quarter (in Q4/23 the increase was 21bps compared to Q3/23). And the number of banks exceeding regulatory guidance for CRE concentration also decline for the fourth consecutive quarter, which can be seen as a good sign. Some banks also have reduced concentration due to the sale of CRE loans.
Despite some signs of improvement, I still see delinquency rates rising and little reason to be optimistic at this point – neither about commercial loans nor about consumer loans.
I still think the inverted yield curve is a reliable indicator and just because it now takes several months between the first inversion of the yield curve until the beginning of the recession, I would not argue that the indicator is false this time. And I am still arguing for investors to be extremely cautious about a potential recession or a depression on the horizon. This assessment is not just based on the inverted yield curve but also on other indicators like bankruptcies, delinquency rates or unemployment which are slowly getting worse.