A look at the (yield) curve

A segment of the US yield curve inverted for a few hours in early April. Was it much ado about nothing, or was it a key economic and market signal? After all, the yield curve in the past has been a reliable harbinger of oncoming recessions. Taking hasty action seems inappropriate because this time really is quite different than before. That said, though, ignorance likewise is rarely beneficial on the financial markets.

This time is different – or is it?

Hardly any other indicator has as good a track record as the US yield curve. Depending on which segment of the curve or which yield spread is taken into consideration, it has accurately predicted seven out of eight or even all US recessions over the last good 50 years. Specifically, whenever the short-term yield rises above the long-term market interest rate, causing the yield curve to invert, a recession always  follows. The theory behind this is that the high short-term yield indicates an existing or forthcoming restrictive environment that slows economic activity while the comparatively lower long-term yield reflects the market’s expectation of a return to falling short-term interest rates in the medium to long run (due to the weaker economic climate and the usually inevitable recession).

But which yield curve spread is the relevant one? Precisely therein lies the crux of the matter. Because, after all, there are almost four dozen different ways to combine US Treasury bonds’ various (remaining) terms to maturity – ranging between one month and 30 years – with each other. The question above couldn’t be more topical because in early April, a segment of the US yield curve briefly inverted, causing an uproar in the financial media.

A brief inversion

Detectable only with a magnifying glass

US yields curves

Sources: Bloomberg, Kaiser Partner Privatbank

However, the inversion (which has since disappeared in the meantime) was microscopically small and was only observable in the two- to ten-year segment of the yield curve. It thus was nothing in comparison to the previous inversion in August 2019, when practically the entire yield curve exhibited an inverted structure. Is this time simply different, and should the yield curve signal thus be disregarded? Caution is called for here because the four words “this time is different” arguably constitute the costliest assertion on Wall Street (because it is often fraught with investment mistakes).

Which signal matters?

Discordance between curves

US yield curve spreads and recessions

Sources: Bloomberg, Kaiser Partner Privatbank

The “discordance” between the curves is in fact striking at present, however. And it is explainable. The 10-year/2-year yield curve spread amounts to just a few basis points while the 10-year/3-month yield curve spread and an alternative measure preferred by the US Federal Reserve (“Fed slope”) are both extremely steep at more than 200 basis points. The difference is located in the maturity spectrum between three months and two years. That segment of the yield curve is very steep, which illustrates that market participants expect the Fed to raise its policy interest rate forcefully in the quarters ahead. In other words, the monetary policy climate is not yet restrictive, but that will soon change. In this sense, the “forewarning” signaled by the 10-year/2-year yield curve spread perhaps is not necessarily false, but simply too early. This would also jibe with history because in the past, this specific yield curve spread has always been the first to send a signal before any of the other possible spread combinations. Its advance warning time before a recession has historically been between nine and 24 months.

This time really is different

Inversion already at the start of the rate-hiking cycle

10-year/2-year yield curve spread and Fed rate-hiking cycles

Sources: Bloomberg, Kaiser Partner Privatbank

The curve’s relevance for the markets

Spotting a recession and identifying its beginning and end may seem to be a trivial exercise devoid of any practical usefulness. Celebrated mutual fund manager Peter Lynch once said that the time an investor spends on analyzing the economy is wasted time. One can definitely have a different opinion in this particular instance, though, because bear markets reliably coincide with recessions. Since the late 1960s, there has only been one recession without a bear market (the first part of the Volcker double-dip recession from January to July 1980), and by the same token, there has only been one bear market without a recession (the 1987 stock-market crash). Foreknowledge of an oncoming recession therefore may eminently work to an investor’s advantage (theoretically). In any case, the relationship between equity returns and the economic activity cycle is not entirely a random one. Share-price fluctuations correlate closely with fluctuations in corporate earnings, and corporate earnings growth is ultimately a function of the economic activity cycle. Statistics corroborate these correlations: ever since consensus earnings estimates for the companies in the S&P 500 index began to be compiled, analysts’ estimates have invariably fallen over the course of a recession (aside from the one exception cited above). In this sense, poor performance in the form of a bear market was also explained by correspondingly weak earnings fundamentals.

It’s the economy, stupid

No recession without a bear market

US recessions and bear markets

Sources: Bloomberg, Kaiser Partner Privatbank

A question of timing

So far, so good – the relevance of the yield curve is beyond dispute. But investors are faced with (at least) two questions. First: Which of the many different yield curve spreads really matter? Our answer is that one should ideally analyze as many yield curve spread combinations as possible. In the past, a meaningful warning sign of an impending recession has consistently flashed whenever at least half of all 45 possible yield curve spreads were negative. During the latest inversion episode in early April, yield curve spreads were negative in only a good 10% of all conceivable combinations. As described at the outset of this article, the inversion was located specifically at the short end of the currently still very steep overall yield curve. A simple recession model devised by the Federal Reserve Bank of New York, which is based on the 10-year/3-month yield curve spread, therefore is not flashing a warning sign yet at the moment. A recession in the next twelve months thus seems very unlikely on the whole. The second question is: When should an investor take action? Our answer is that even though the risk of a recession is not acute yet, the recent “mini-inversion” should not be taken lightly. In fact, the countdown to the next economic downturn in the USA probably has already started. It is highly questionable whether the Fed will succeed in engineering a hoped-for soft landing in the face of an extremely overheated employment market and rampant inflation. The yield curve has already submitted its opinion on this matter and could ultimately prove once again to be the canary in the coal mine. Investors, however, shouldn’t necessarily reduce their equity positions yet. In the past, the US stock market has gained more than 10% on average in the months immediately following a yield-curve inversion. Nevertheless, it makes sense to think about implementing hedging strategies (as inexpensively as possible).

No warning signs (yet)

Fed recession indicator in expansion territory

Federal Reserve Bank of New York Recession Indicator

Sources: Bloomberg, Kaiser Partner Privatbank