Yield curve inversions are the worry du jour in the financial media with many proclaiming an oncoming recession because a certain part of the yield curve flipped upside down in recent days. While the shape of the yield curve has historically been an incredibly reliable recessionary indicator, there’s a lot more to the story than the headlines would suggest.
The yield curve itself is simply a graphical representation of the interest rates paid on different maturities of Treasury bonds. Normally the yield curve is upward sloping, like a ramp a skier would launch off of. This upward sloping curve, meaning you get more interest for buying longer term bonds, is typical in normal economic environments. Here’s a link to an excellent short video on the topic from The Wall Street Journal: https://www.youtube.com/watch?v=bItazfbSptI&t=96s.
This “normal” upward sloping yield curve is consistent with our natural feelings towards money lending. If you’re lending money for 30 years, you’d usually demand a higher interest rate than if you lent that money for 2 years. A lot can happen in the additional 28 years of the longer loan, so investors typically demand a higher interest rate to compensate them for their foregone future opportunities and for the risk associated with such a long term arrangement.
When buying bonds (lending money), apart from the creditworthiness of the borrower, what other question is there apart from “How much will you pay?”. With the Treasury market, nobody questions repayment, which is why the fascination with the yield curve is unique to that market; the future changes in those interest rates are the ONLY thing that will impact your return over time.
The image below displays a “normal” yield curve in the summer of 2004 in the middle of an economic expansion, which is the most common phase of the economic cycle. Notice the “upward ramp” at the end of the curve which reflects investors' increased interest demands as the duration lengthens. (Interesting sidenote: This curve doesn’t contain the 30 year Treasury which the government ceased issuing for a four year period starting in February 2002.)
So what does an inverted, “abnormal” yield curve look like? The image below displays the yield curve at various points in the last year. The curve on April 1st, 2019 is pink and the curve on April 1st, 2018 is shown in yellow. Various maturities in the middle of today’s curve yield less than shorter term maturities, therefore that part of the curve is inverted.
Notice how the yellow line (the curve a year ago) really only changed relative to today’s curve (the pink line) on the short end (less than 2 years)? This is telling us that the “long bond” hasn’t changed its opinion of the long run economy very much in the last year. The long bond has always believed long term growth would be low and slow. If 30 year Treasury holders didn’t think that, they wouldn’t own those bonds! After all, there’s not a lot of room for error in inflation and growth expectations factored into a 30 year Treasury yielding less than 3% per annum.
Ok, so we’ve covered what inversions look like.
But what causes these inversions? Why would someone accept a lower annualized yield on a 3 year Treasury (currently paying 2.21%) than they would on a 1 year Treasury (currently paying 2.4%)?
The answer is that the 3 year Treasury holder anticipates that at some point in the next three years, a 1 year Treasury will yield far less than 2.4%. In other words, 3 year Treasury holders expect the Federal Reserve to be cutting interest rates in the near future, presumably to fight an economic slowdown.
Another way of looking at it is that 3 year Treasury owners assume a high likelihood that today’s 1 year Treasury holders will be disappointed to find that short term rates (and therefore their reinvestment opportunities) have fallen substantially upon maturity of their bond. The 3 year Treasury holder wants to “lock in” a rate for longer, even if it is less than what a 1 year Treasury pays, because they fear short term rates falling in the near future. So while a 1 year Treasury owner will certainly collect more interest this year, years two and three are uncertain. If 2 year Treasuries yield 1% one year from now, today’s 1 year owner will only have earned a total return of 4.4% over three years (1 year earning 2.4% then reinvesting for 2 years earning 1%), whereas today’s 3 year Treasury owner will make a total return of 6.63% (2.21% for three years).
This is the most important takeaway from the yield curve inversion discussion; it isn’t a CAUSE of future recessions but rather an early SYMPTOM of economic slowdowns. It’s just reflecting investors’ collective perceptions of the current economic environment and their expectations for the future. The yield curve flattens and then inverts when market participants as a whole downgrade their expectations for the economy.
Now that we covered what the yield curve is and why inversions happen, let’s briefly look at various yield curve inversions being discussed today, their predictive power and their impact on markets going forward.
Here’s a chart we created on the Federal Reserve website illustrating the difference between the 10 year and 2 year Treasury (“2s10s spread”) with the last three recessions highlighted in the shaded areas. When the line crosses below zero 10 year bonds pay less than 2 year bonds and this part of the curve is inverted. Many economists believe this is the most important and relevant yield curve in relation to economic health.
Look how long it took for a recession to follow the initial yield curve inversion? In the last three cases, around two years passed from the initial 2s10s spread inversion until the actual onset of a recession. Keep in mind that this type of inversion hasn’t happened yet. It may or may not.
And what about the false start in late 1994/early 1995 when this part of the curve flattened dramatically and was within a few basis points of inverting but never crossed and no recession came for the better part of a decade? That was the one and only modern-day “soft landing” orchestrated by the Fed where an interest rate tightening cycle commenced and ended without causing a recession. Investors who jumped the gun and sold, thinking the curve was inverting, missed another few years of positive returns.
What did recently invert was the 3 month Treasury spread relative to the 10 year Treasury (“3M’s10s”). Here’s the same chart as above with the different spread, the one that recently inverted and made headline news.
Here again, we only have three examples in 30 years and again the mid-to-late 1990s featured a few false starts.
There’s an important commonality between the two different yield curve spreads, and it’s the best “add-on” indicator that’s traditionally increased the accuracy and timing of inversion-based recession calls. Look at what’s happening to the curve when the economy actually enters a recession in the gray area? The curve is sharply steepening, not flattening or inverting further. That’s the sign you’d be looking for if you were trying to time a recession on a dime.
What would be causing the curve to steepen leading into the actual recession? Wouldn’t investors flee for safety in a downturn and snap up long-dated Treasuries? Yes, long Treasury yields generally do fall in recessions, but shorter term rates tend to fall faster and therefore re-steepen the curve. Leading into the last few recessions, the Fed has been cutting short term interest rates, sensing economic weakness before the recessions actually began. And barring a massive external event which forces the Fed to act, they’re probably 9-12 months away from even being able to consider rate cuts given their recent commentary.
So the “endgame” of inversions, the sharp re-steepening of the yield curve coinciding with the onset of a recession, appears to be at least a year away.
This probably all sounds very doom and gloom. If you’ve read this far you deserve some good news.
History suggests stellar stock returns in the near future.
Check out the far-above-average stock returns after the last four instances of 3M’s10s Treasury curve inversions.
The last four inversions were followed by positive stock returns which were, on average, in the 80th percentile of all returns. This doesn’t mean that the stock market went down after 30 months on average, but rather the returns were above average for over 30 months following the initial inversions over the last four occurrences. The 3M’s10s spread just inverted a few weeks ago, implying strong equity returns over the near term.
But averages can be deceiving. Did one or two outliers skew the results? Blowing up the small chart on the image above, we see that there wasn’t a single negative equity outcome for over two years following the past four initial 3M’s10s inversions!
That’s obviously encouraging news to equity owners.
While no one metric should be trusted wholeheartedly, this data along with other leading indicators suggest it is wise for stockholders to stay invested while remaining on the lookout for a rapid steepening of the yield curve.
The yield curve is one of the best recessionary indicators available to us as investors. But whatever predictive power it holds is only based on the economy itself, not markets. Even if we should expect a recession in the next year or two, we believe stocks have largely already priced in that outcome. Four out of ten S&P 500 stocks are trading below 15x earnings, while only 25% of the S&P 500 is trading above 25x earnings, the best “cheap stock-to-expensive stock” ratio since 2012! Consider, for example, three sectors which are core engines of the economy, homebuilders, automakers and big money center banks. They trade at high single digit/low double-digit P/E’s, valuation multiples consistent with actually being in a recession.
So while market averages overall are doing fine, economically sensitive sectors are already trading as if a recession has started. If that turns out to be the case, they are fairly valued. If not, those multiples should improve.
One last thought that gives us comfort as allocators is that if the Fed were to hike interest rates any further, they would essentially be doing so knowing that they would be causing a recession. It is rare for the Fed to hike rates when the yield curve is extremely flat or inverted. There have been economic environments where the Fed deliberately slowed down economic growth, but never with the economy barely growing above 1-2%. That type of intentional slowdown seems highly unlikely today.
With the recent yield curve inversions and slowing economic growth, we believe the Fed is likely done hiking interest rates for the cycle, meaning less pressure on stocks from interest rates. Additionally, an inverted yield curve is typically followed by strong stock returns before the onset of the forecasted recession.
Should you have any questions regarding this or any other topic, please reach out to us and we’ll be happy to assist you.