The Great Steepening

A new bull market in bonds...

I’ve spilled significant ink on the slope of the yield curve in recent months including in the most recent quarterly update for Smart Bonds, “It’s All About the Cycles.” 

I track the yield on 10-year Treasuries less the yield on the 2-year; on that basis, the curve remained continuously inverted from July 5, 2022, until September 3, 2024, a record total of 565 trading days.

Since the early 1980s, a prolonged period of inversion followed by disinversion, where the curve regains a positive slope, has been a reliable warning signal of recession ahead. I prefer to wait for the curve to regain a slope of at least +25 to +50 basis points before considering it a valid disinversion signal; since last week, the maximum slope of the yield curve has been just 6.2 basis points (+0.062%).

To support a rapid steepening to that +25 to +50 bps target range, we’ll likely need to see further evidence of economic deterioration. We’ll also likely need confirmation from the Fed that market expectations for a whopping 100+ basis points of Fed easing by December 18^th this year are at least in the realm of possibility.

And, of course, there’s no such thing as an infallible economic indicator. This yield curve disinversion signal has been “wrong” before. In the late 1990s, the yield curve inverted and disinverted; yet, the central bank managed to engineer a soft landing, and the US economy didn’t see recession until March 2001. What’s more, the stock market enjoyed a blistering final leg higher from late 1998 through the ultimate bull market peak in March 2000.

The implications of the ongoing recession vs. soft landing debate are different for fixed income compared to equities and the S&P 500.

Regardless of whether the current disinversion ultimately ends in recession or the Fed manages to engineer a 90s-style soft landing, history suggests it’s likely to signal at least an intermediate-term bull market in bonds.

Let’s start with this:

The Port in the Storm

Most investors are aware of the bond market’s reputation as a haven during recessions.

Just consider the 2001 experience:

Source: Bloomberg

This is a weekly chart of the Bloomberg US Aggregate Total Return Bond Index from early 1998 through the end of 2003. This index tracks both Treasuries and investment grade corporates issued in US dollars. It’s a total return index, meaning it tracks both price appreciation and coupon payments (yield) on bonds.

I’ve labeled just a handful of the more important events over this period including those related to the economy, interest rates and the S&P 500 (equities). As you can see, in this cycle the bond market was essentially flat from late 1998 through early 2000, then began to rally around the time equities peaked. That rally gained steam a few months before the Fed’s first cut in this cycle, a 50-basis point surprise move in early January 2001.

Bonds continued to rally straight through the March – November 2001 recession and the entire bear market in the S&P 500, which didn’t end until late 2002. Bonds then continued their positive bias long after stocks began their recovery in 2002-03.

The 2007-09 cycle brought a similar pattern, albeit more extreme due to the severity of the Great Recession cycle of that era:

Source: Bloomberg

This is the same chart with a similar sequence of economic, stock market and Fed events labeled.

Much like in 2000-02, the bond market started a significant rally just ahead of the peak in the S&P 500 – the index traced a double top in July and October 2007 – which accelerated as the economy entered the Great Recession in December 2007.

Overall gains for bonds continued long after the lows in the S&P 500 in March 2009 and the official end of the Great Recession in June of the same year. In total, the Bloomberg Aggregate Bond Index returned more than 18% from June 30, 2007 through the end of 2009, a period when the S&P 500 fell well over 20%.

Even more impressive, long-term Treasury bonds jumped more than 50% from the end of June 2007 through the end of 2008 alone.  

And that brings me to this:

No Recession Necessary

With apologies to longstanding readers for repetition, an inverted yield curve alone is NOT a signal of recession ahead.

An inverted yield curve, where 2-year Treasury yields are higher than 10-Year yields, simply means the market regards current Fed policy as restrictive. 

That’s because the central bank exercises the most direct control over short-term interest rates, so the two-year yield generally rises when the Fed has been hiking rates and is expected to continue to do so. In contrast, the 10-year Treasury represents the market’s expectations for rates and, by extension, economic growth and inflation, over the longer haul.

By the same token, when the curve disinverts after a period of continuous inversion that simply signals the market sees Fed rate cuts ahead, and less restrictive policy. In many cases, a rapid disinversion signals growing market fear the central bank has committed a policy error – this “mistake” happens when, after a period of tight monetary policy, the economy weakens faster than expected and the central bank is forced to cut rates aggressively to avoid recession.

However, the US economy doesn’t need to enter recession, and the S&P 500 doesn’t have to enter a bear market for the bond market to see strong gains following curve dininversion. Indeed, historically, Fed “pivots” evident in the yield curve can act as a powerful tailwind for bonds even if the economy reaccelerates.

Take a look:

Source: Bloomberg

This weekly chart shows the slope of the yield curve on the same 10-year/2-Year basis as an orange line (see the right-hand scale) from the beginning of 1993 through the end of 1999. The dark blue line (see the left-hand scale) represents the same Bloomberg Aggregate Bond Index on a total return basis I’ve been showing you throughout this update. The green line represents a zero slope for the yield curve, so readings below this line represent an inverted curve.

There was no recession in the 1993-1999 period, and the S&P 500 did not experience a peak-to-trough closing decline of 20% or more.

However, the Greenspan Fed of this era was known for subtle adjustments and “tweaks” to monetary policy aimed at prolonging the economic expansion cycle while forestalling any resurgence in inflation. Some would argue, rather convincingly in my view, the Greenspan Fed also attempted to use monetary policy to support the stock market during periods of turmoil such as the summer and autumn of 1998.

Regardless, you can see the Greenspan Fed’s machinations in the yield curve.

For example, the yield curve flattened aggressively in 1994 as the Fed hiked its target Fed Funds rate from 3% to 6%. The market clearly viewed rates in late 1994 and early 1995 as restrictive as the 10-year/2-Year curve nearly inverted; the low for the slope on a weekly closing basis was around 11 basis points in late December 1994.

However, the yield curve re-steepened dramatically from late 1994 through early 1996 as the market began to anticipate Fed cuts. The central bank followed through and delivered three 25 basis point cuts, the first in July 1995, followed by cuts in December 1995 and February 1996.

Now, look at the Bloomberg Aggregate bond index over this time, which I’ve boxed in my chart above. The index flatlined amid rate hikes in 1994, then bottomed out around the nadir for the yield curve slope in late 1994. Bonds then enjoyed a solid rally into early 1996.

Indeed, after a brief dip in early 1996, bonds continued to move higher up until around the middle of 1999.

In the summer of 1998, the yield curve inverted, albeit slightly and for only a short period of time. Then, following the Russian debt default in August 1998, and the collapse and subsequent bailout of the Long-Term Capital Management (LTCM) hedge fund in September, the Fed cut rates from 5.50% to 4.75% between September and November 1998.

As the market began to sniff rate cuts just ahead, the yield curve re-steepened dramatically into late 1998 and retained its positive slope through all of 1999.

Granted, this inversion cycle wasn’t as bullish for bonds as the 1994-95 episode; however, the bond market did experience a meaningful rally through the final half of 1998 and the first half of 1999.

Let’s examine a more recent cycle:

Source: Bloomberg

I’ll allow the trends in 2017-19 aren’t quite as clean; however, the pattern rhymes.

The 10-year/2-Year yield curve steadily flattened through 2017 and 2018. The main reason was a shift in the cadence of central bank tightening.

The Fed hiked 25 basis points once in December 2015 followed by a single 25 basis point hike a year later in December 2016. In 2017, however, the pace of Fed hikes began to accelerate and, by early 2018, the market began to fret that the central bank was tightening too quickly, risking a policy mistake.

By late 2018, the stock market collapsed, and the yield curve flattened to a weekly closing low of under +15 basis points in early December 2018. Even as the Fed went ahead with a 25-basis point hike in December 2018, and confidently projected further Fed hikes into 2019, the market began to price in a growing probability the Fed’s next move would be a cut.

In early 2019, the central bank began to tone down the hawkish rhetoric and in late July, the Fed followed through with a 25-basis point cut while pivoting to focus more on the growth side of its dual mandate. The yield curve responded by stabilizing through the first half of 2019.

The curve did not steepen rapidly in the first half of 2019; indeed, the 10-year/-2-year spread inverted outright by August 2019 amid growing fears the Fed needed to ease more aggressively to avoid recession. However, the simple shift from aggressive flattening to a more stable slope in late 2018 and early 2019 was enough to catalyze a rally in the Bloomberg Aggregate Bond Index, which jumped over 6% in the first half of 2019 alone.

My point is simple: In the last few weeks, the Federal Reserve has clearly shifted its bias in favor of rate cuts, all-but-promising at least a 25-basis point at its next meeting on September 18^th. The 10-year/2-Year curve has disinverted after a prolonged period of inversion, further confirmation of the shift in the rate backdrop from restrictive to, at a minimum, a more neutral setting.

As I’ve explained in recent updates and alerts, I’m still getting mixed signals regarding the economy and most of the indicators I follow have yet to deteriorate enough to suggest imminent recession. However, whether the economy does enter a downturn, or the Fed sticks a soft landing, one key point is clear:

The gale-force headwind of a hawkish Fed is morphing into the tailwind of a central bank more focused on supporting growth and the labor market. The latter is a more benign scenario for the bond market.

Selectivity Remains Crucial

Throughout this update, I’ve been using the Bloomberg Aggregate Bond Index on a total return basis to illustrate historical trends in bonds around important shifts in the yield curve and Fed policy. This is a broad-based benchmark comprised of US dollar-denominated fixed-rate, taxable bonds including Treasuries, investment grade corporate, mortgage-backed securities and commercial mortgage-backed securities. The modified duration here – a measure of interest rate sensitivity – is about 6.4.

This index is intended to act as a benchmark for virtually all corners of the US bond market.

In contrast, one of the core tenets of Smart Bonds is there’s no indivisible, monolithic “bond market.” That’s because specific niches of the bond, credit and preferred markets, including several not included in the Bloomberg Aggregate Index, can produce widely divergent returns at various points in the economic and rate cycle.

The Bloomberg Aggregate Index gives us an idea of the mean performance of fixed income securities; averages are frequently misleading, hiding both dramatic underperformance in certain segments, and strong pockets of strength in other corners of the market.

Selectivity in these markets is never more important than at crucial turning points, such as we’re seeing right now.

Let’s look at a a recent example of how this plays out:

Source: Bloomberg

I know this chart is a little busy, but please bear with me.

The green line at the bottom of the chart shows the slope of the 10-year/2-Year yield curve over the period from January 2016 through the end of 2021.

The five other lines are total return indices I’ve created for 5 bond and preferred stock exchange-traded funds (ETFs): A long-term Treasury ETF (dark blue line), intermediate-term Treasuries (orange line), intermediate-term corporates (dark green line), high-yield “junk” bonds (light blue), and preferreds (purple).

I discussed how the yield curve behaved through this period in a bit more depth above. However, in broad strokes, the yield curve flattened from late 2016 through most of 2018 as the Fed accelerated the cadence of rate hikes. This was not a great period for the bond market, at least as measured by the Bloomberg Aggregate Bond Index – the index fell 1.14% from the end of September 2016 through the end of September 2018.

However, around half of the Bloomberg Aggregate Bond Index is Treasuries including long-term Treasuries, which carry a high level of interest rate sensitivity (when rates rise, these Treasuries fall in price). So, the long-term Treasury Bond ETF on my chart dragged down the performance of the overall Aggregate Bond Index, falling 10.3% over a similar holding period.

That does NOT mean all corners of this market lost money.

Indeed, over the same period, the high yield “junk” bond ETF on my chart, not tracked by the Aggregate Index, rallied almost 10%. Preferred stocks also saw their day in the Sun, particularly through the middle of 2018.

Then, as I explained earlier, the yield curve stopped flattening significantly by early 2019 as the Fed shifted its bias in favor of cuts. By early 2020, the Fed accelerated the pace of cuts amid the coronavirus outbreak, lockdowns and travel restrictions which tipped the US economy into recession.

As you can see, when the yield curve stopped flattening in the first half of 2019, Treasuries “woke up” and began to rally. Intermediate-term investment grade corporates, high yield bonds and preferreds also saw strong gains through the first half of 2019 powered by the tailwind of a Fed pivot and market expectations the economy might well recover and avoid a hard landing.

Finally, once recession was imminent, all corners of the bond market stopped working except for Treasuries, which took off in the spring of 2020.

This, in a nutshell, is what we seek to track in the Smart Bonds model portfolios – we’re looking to identify the corners of the bond market best-positioned to benefit at any given stage of the economic and market cycle. As in late 2018 and early 2019, we’re seeing a crucial shift in the cycle right now, and that demands a shift in our exposure in the model portfolios as well.  

The good news is that a “bull” steepening in the yield curve such as we’re seeing right now generally represents a tailwind for most ETFs we track in the service though, certainly, there are some danger zones as well as niches that stand to perform far better than average.

The second bit of good news is that we don’t have to be particularly early in “calling” a recession to profit through the cycle. Nor do we need to make dramatic shifts in the model portfolio suddenly to respond to market conditions.  

The bond market can be more forgiving than the stock market.

After all, look at the first 8 months of 2008, the start of the most vicious economic downturn since the 1930s, and you’ll find investment grade corporates still managed to produce a small positive return. In fact, preferred stocks – the majority issued by the financials, the industry at the epicenter of the credit crisis of that era – were down less than 5% through the first 8 months of 2008. That was through a period that included the collapse of Bear Stearns, and Bank of America’s bailout of Countrywide Financial.  

Our strategy has been, and remains, to gradually adjust our Smart Bonds portfolio exposure over time to maximize returns and minimize risk and volatility.

Here’s an update on total returns since inception for the three model portfolios, incorporating the handful of portfolio adjustments we made at the end of August:  

Source: Bloomberg

As you can see, since inception on May 30^th, the Smart Bonds Baseline model portfolio has returned 3.83% including both price appreciation and (monthly) distributions paid by the recommended ETFs. The Defensive Allocation is up 3.29% and the Dynamic 60/40 Portfolio, which allocates across a portfolio of bond, credit and preferred ETFs as well as the S&P 500 is up 4.83%.

Returns represent the total gain over a period of just over 3 months since portfolio inception in late May, NOT an annualized return.

I calculate returns as of the market close every Friday; this week, I’m also presenting returns in the chart as of the close on Wednesday September 11th, 2024. I’ve decided to present historical performance in chart form, so you can get an idea of the volatility for these three portfolios through the summer, a period of significant turmoil in stocks, currencies and commodity markets. As you can see, amid all those wild swings, bond ETFs have offered an island of relative calm and steady performance.

With those points in mind, I’m not recommending any additional adjustments to the three model portfolios in this issue.

Here’s an updated look at the current portfolios and recommended allocations to each ETF recommendation: