Weekly Newsletter | July 10, 2023
In this report, we will discuss yield curve inversions and emphasize the importance of considering the broader context before interpreting them as a recession signal. Our aim is to present the information in a clear and accessible manner and avoid using elaborate fixed income math. We do not advocate fully dismissing inverted yield curves but instead highlight how bad the inversion signal is at timing the market and provide two key arguments for why this time may be different. First, we argue that the inversion is mainly driven by expected normalization of monetary policy and a return to low long-run real yields. And secondly, we argue that the persistent demand for duration by pension fund-like investors makes for structurally flatter yield curves.
Contents
Yield curve inversions
Interest rate curves are multidimensional matrices
Key fixed income concepts
Matrix of forward rates
Trust the model – at your own peril
Key drivers behind the inverted yield curve
Conclusion
Parting (market) thoughts
Yield Curve Inversions
At its most basic level, the yield curve represents the difference between the yields of longer-term (e.g., 10-year) and shorter-term (e.g., overnight or two-year) US Treasuries. Normally, long-term yields are higher because investors require greater compensation for the uncertainty of the distant future. However, starting in 2022, the yield curve has inverted, and it now exhibits the opposite pattern. This inversion has traditionally been regarded as a reliable leading indicator of an upcoming economic downturn. And the US Treasury market, as the most liquid and deep fixed income market worldwide, attracts significant attention, making these signals particularly important.
Figure 1: Yield curve inversions and recessions
There have been ten recessions in the US since 1955, and the 2/10 year yield curve inverted 6 to 24 months before each, with only one false positive (in the mid-1960s). Note however the 2019 inversion was small and brief, and its preceding the 2020 pandemic recession should be seen as a coincidence.
Inversions have a negative effect on the economy, via two channels: credit markets and behavioral economics:
On the credit side, consider banks typically borrow money short-term and lend it long-term, earning a profit on the spread. When the yield curve inverts, those spreads narrow and may even become negative, compressing banks' profit margins. As a result, banks become more cautious about extending credit, leading to reduced lending and slower economic growth.
On the behavioral side, inversions can be self-fulfilling prophecies. Given that it is widely known that inversions precede recessions, economic actors such as investors and CEOs start anticipating an economic downturn the moment the curve inverts, and thus become cautious, which then triggers a downturn. Hence, a self-fulfilling prophecy.
Interest Rate Curves Are Multidimensional Matrices
Now getting slightly technical, interest rate curves are multidimensional matrices. Focusing solely on two points in the curve (i.e., the 2 year and the 10 year) provides a one-dimensional snapshot of the relationship among interest rates. Traders familiar with swaps know that a curve typically consists of more than 20 tenor points, with numerous interconnections among these points and neighboring curves. The phenomenon resembles a complex fabric, the Fabric of Interest Rates, where pulling one point affects the entire web of interconnected nodes (see Figure 1). Therefore, making sweeping conclusions about the future of the economy on just two points is overly simplistic.
Rates traders plot the Fabric of Interest Rates using the so-called yield curve surface, which plots on a three-dimensional chart the term structure of a yield curve over time.
Figure 2: The fabric of interest rates
Key Fixed Income Concepts
Before we continue, let’s ensure we are all on the same page and revisit some fixed income concepts that will make the argument easier to understand.
Concept 1: The yield in fixed income reflects the average yield for that period
What does this mean? Suppose there is a 5-year bond trading at a 4% yield. This indicates the market expects average yields to be around 4% for each of the next 5 years. There are various combinations of yearly yields that can lead to this average.
Concept 2: Forward rates
Think of forward rates as interest rates for a fixed time period in the future. If the 1y1y forward rate is currently 4.5%, then a 1y rate starting in July 2024 and ending in July 2025 is priced at 4.5%.
Concept 3: Bootstrapping the yield curve
Bootstrapping the yield curve is a technique to construct the term structure of interest rates. It involves using market prices of fixed-income securities to infer the yields at different maturities.
Suppose the 1y rate is 5.5% and the 2y rate is 5%. Given that the 2y rate represents the average of the next two 1-year periods, we can easily solve for the 1y1y forward rate, which is equal to 4.5% (5% * 2 - 5.5% = 4.5%).
Table 1: Calculating forward yields
Matrix of Forward Rates
It is therefore good practice to not obsess about the spread between two tenors of the yield curve. Instead, it is best to focus on the matrix of forward rates and how it behaves (see Table 2). In this matrix we can see where the market thinks 2y treasury yields will be in 3m, 6m, 1y and so on (row highlighted in red).
Table 2: Matrix of forward rates
Trust the Model – At Your Own Peril
The market's fixation with yield curve inversions started with economist Campbell Harvey, who in a 1986 dissertation presented a linear regression model showing that the yield curve spread had ability to predict future consumption growth. This model has had an impressive track record at predicting recessions, although a bad one at timing them, as the span of time between inversion and recession is long and highly variable (i.e., 6 to 24 months).
However, this model appears outdated, considering the wealth of data available, real-time forecasting, and the advanced tools at our disposal.
Let’s use the Taylor Rule as a potential analogy. The Taylor Rule is an economic model created to help central bankers set interest rates based on inputs of inflation and growth potential. It worked very well for years and was widely adopted by policy makers and investors. However, the model ceased to work in 2009, when central banks had to loosen monetary policy aggressively and keep rates at zero for longer than expected. Many Fed critics were stubborn and lost fortunes in the decade following the Great Financial Crisis (GFC), as they believed overnight rates should be much higher because the Taylor rule said so.
If the Fed had continued to blindly follow the Taylor Rule, the economy would have most likely gone into a recession. The ECB followed a more orthodox stance following the GFC and hiked rates prematurely, which led to a series of economic crises. More recently, particularly during Covid, the Taylor rule was thrown out the window and you rarely hear about it anymore (outside academia).
Furthermore, consider the many financial anomalies we have witnessed in the past 3 years. These include a staggering amount of negative yielding debt (reaching $15 trillion at its peak), oil futures trading at deeply negative prices, and meme stocks of unprofitable bankrupted companies trading at absurd valuations. For example, GameStop and Hertz defied gravity during 2021 and traded at a combined valuation of $40 billion, something that blew up some of the most sophisticated investors.
This is the exact kind of world where old economic models are likely to start breaking.
Figure 3: The Taylor Rule (the series in blue represents where the overnight rate should be based on the Taylor Rule, while the series in red is the effective fed funds rate)
Key Drivers Behind the Inverted Yield Curve
We are not suggesting inversions should be disregarded. However, we do argue that the curve's signal is less straightforward than in the past. We are convinced a significant portion of the inversion observed in the US yield curve is not primarily driven by high recession probabilities. Instead, it can be attributed to persistently low long-run real rate levels (i.e., equilibrium or natural interest rates in an economy over an extended period), and investors pricing a rate normalization cycle without necessarily triggering a recession.
Supply & demand dynamics for long duration assets is another important reason behind the inverted curve. As appealing as 3-month bills may be at 5.5%, it is impossible for large institutional investors to shift their entire portfolios towards these short-dated assets. Large pension funds and insurance companies operate under strict mandates that only allow for minor adjustments within a predefined set of parameters, and they are constantly looking to find long duration assets to match their long duration liabilities, which means they are structurally biased to favor longer duration over yield.
The Federal Reserve can raise interest rates up to 5.5% to slow down demand and encourage investors to hold cash. But there's strong demand for long-term investments regardless, from global pension funds and other allocators who need to match their long-dated liabilities. They find 30-year treasuries with a 4% yield incredibly attractive. Pension funds demand has been trending higher, growing twice as fast as worldwide GDP since 2009. This creates a structural change in the equilibrium of yield curves, which are thus flatter than in prior decades.
Figure 4: Growth of global pension fund assets
That is why, regardless of what is happening with the economy, the yield curve may have inverted purely on supply & demand dynamics. On one hand, the Fed raised overnight rates by 500 basis points, sending the front-end of the curve flying higher, while on the other demand for long-duration assets remains strong, keeping the long end pinned down.
Other Considerations
Furthermore, the fact that the Federal Reserve has been actively reducing its holdings of long-duration assets through quantitative tightening (QT) serves as evidence of the resilient demand for these assets. Even with the largest holder of treasuries (the Fed) aggressively supplying duration to the market, longer yields still fail to rise above shorter ones.
Finally, it is worth mentioning that yield curve inversions in both advanced and emerging economies do not always result in recessions. In fact, they are often influenced by the supply and demand dynamics we mentioned earlier. It is particularly interesting to highlight the cases of Switzerland, Japan, and the UK, where the link between 2s10s yield curve inversions is not as clear as the one observed in the US. This emphasizes that the relationship between yield curve inversions and economic outcomes can vary across different countries and should be analyzed in a broader context.
Conclusion
Historically yield curve inversions have been accurate at predicting recessions, even if poor in timing the recessions. Furthermore, the 2/10 year yield curve is extremely inverted at present, and has been inverted for about a year. But contrary to popular belief, that does not mean a massive recession is around the corner.
We argue that the current inversion is mainly driven by expected normalization of monetary policy and a return to low long-run real yields, as well as persistent demand for long-duration assets by pension fund-like investors. Therefore, the current inversion may be creating the wrong signal, which explains why economists keep on pushing the impending recession forward.
This time may be different after all, something even the father of the inverted yield curve model has said so.
As covered in our prior report, The Big Picture, at Asgard we believe that even if a small and brief recession is the most likely scenario, this recession would be small enough to not have a major impact in risk assets. In fact, given how negative market participants are, the major risk for H2’2023 is for the economy to get too hot, rather than for a major recession to unfold.
Parting (Market) Thoughts
We want to stress that while the curve is extremely inverted, it will start steepening before the hiking cycle is done. So, if you are in the camp that believes the Fed is about 90% done, as we are, then this is the sweet spot to put on steepening trades (bet on the curve to slope upward). Moreover, expect a quick steepening move if there is a downward CPI surprise in the coming months, as lower rates should then compress further than longer rates, and start pricing a rate normalization cycle.
Don't miss out on our next report. Sign up here to stay tuned.
Disclaimer:
Please note that the information contained in this market presentation is for informational purposes only and should not be construed as investment advice or a recommendation to buy, sell, or hold any securities or financial instruments. Asgard and its affiliates do not provide investment, legal, or tax advice.
Investing in financial markets involves risks, and individuals should carefully consider their own financial situation, risk tolerance, and investment objectives before making any investment decisions. Past performance is not indicative of future results, and the value of investments may fluctuate.
The content presented in this presentation is based on publicly available information and sources believed to be reliable. However, Asgard does not guarantee the accuracy, completeness, or reliability of any information included herein. Any reliance on the information presented is at the individual's own risk.
Asgard and its representatives shall not be liable for any losses, damages, or expenses arising out of or in connection with the use of this market presentation or the information contained herein. It is recommended that individuals consult with their own financial advisors or professionals before making any investment decisions. This market presentation is the property of Asgard and may not be reproduced, distributed, or transmitted without prior written consent.