Don’t interpret rising US Treasury yields as a sign that a crash is coming

Richard Harris says the psychologically important 3 per cent rate mark for the 10-year US Treasury note may not signal doom and gloom if we look at the bigger historical picture

PUBLISHED : Thursday, 03 May, 2018, 10:37am
UPDATED : Thursday, 03 May, 2018, 10:35pm

“Why did nobody notice it?” said Queen Elizabeth. Like the academics at the London School of Economics she addressed, I never thought that a little local lending difficulty in the US would cause the 2008 crash.  

But, after the event, it was clear to every market observer in the world that it was inevitable. Some claimed to have predicted it – the number being so small that statistically they could have done so by luck. This is a classic symptom of hindsight bias, or the ability of experts to fully explain why an event happened – after the result.  

The benefit of a long weekend in China, without a fully functioning internet, meant I was able to pore over Michael Lewis’s The Undoing Project, the story of behavioural psychologists, Daniel Kahneman and Amos Tversky, written for non-psychologist dummies. Kahneman became the first psychologist to be awarded a Nobel Prize in Economics (Tversky missing out because of his early death).  

Market fixation with US Treasury bond yields is misguided

China’s debt emerges as surprise refuge in global bond retreat

Kahneman and Tversky were puzzled by the fact that, in many professional fields, the experts were just as prone to the “gambler’s fallacy” as anyone else. Specialists made their errors in a more complex, sophisticated and convincing way, using jargon and the mantle of experience. 

Essentially, the professionals told the two researchers that the more heads appear as a coin is tossed, the more likely it becomes that the next spin will be a tail. Yet the chance of a tail coming up is always 50 per cent. If you were able to do a million tosses, tails would indeed come up half the time – but not with a small sample. And we investors always work with small samples.  

Kahneman and Tversky pointed out that the judgment of experts from medicine, the military and the markets is often coloured by a particular emphasis that has only been derived from a few examples – which might not be representative themselves. Our human minds are very bad at judging the correct probabilities for a predicted event because we get distracted and emphasise our subjectivity.  

Bond market bulls and bears are prowling in the dark

Treasury notes gain in reaction to cuts in rates

This means, for instance, that if we are following a stock on a chart and it reaches an inflection point – it is likely to “do something”. We may divine that it will move in a particular direction – but is that based on a big enough sample? If we had prior knowledge (which is illegal), or were able to take a multitude of factors and weigh them perfectly, we might be able to predict “up, down, or sideways”. We can’t, so we put evidence into boxes based on events that made a pattern a few times before.  

Sometimes, our investment analysis may not even be based on objective observation but on the oft-spoken prevailing wisdom that substitutes for market knowledge at that moment. This perceived wisdom might become rules, mantras or fake news (the January effect, the Nifty Fifty, Crooked Hillary), which in other market contexts makes no sense. Kahneman and Tversky called this approach to problem-solving “representativeness” – leading to stereotyping bias.  

Perhaps the most closely watched economic benchmark today is the US 10-year Treasury note hovering around 3 per cent. Interest rates haven’t been this high in years. Doom and gloom is said to befall should it pass over this magic marker.  

China’s falling bond yields steal the spotlight from stock-market bulls

Yet, in December 2013 when the rate broke 3 per cent, there was less concern than in today’s high-growth, low-inflation, low-unemployment Goldilocks economy. Rates just before the global financial crisis were around 5 per cent and they had fallen steadily from almost 16 per cent in July 1981. In the 1960s, rates were rarely below 4.5 per cent and peaked at 7.5 per cent just before the 1970s oil shock. Today’s oil and gold prices would have given a 1960s economist apoplexy. Three generations of market participants have never seen sustained rising long rates.  

Kahneman’s and Tversky’s work encourages us to look at the whole data-based, evidence-led picture rather than just a few figures. Interest rates of 3, 4 or 5 per cent are merely a small addition to the cost of doing business in the current economic environment – but we worry because we don’t know at what point the pain will hit. 

We have no comparisons for modern interest rates. Today’s economy is wildly different from 1962. So we invent stereotypes based on small samples. The impact of rising interest rates to a modern economy is uncertain but you can be certain, after rates have gone up, that hindsight bias will give us 20:20 vision. 

Richard Harris is a veteran investment manager, banker, writer and broadcaster and financial expert witness.