Join our community of smart investors

The low-rate danger

Investors should not be tempted to buy shares merely because interest rates are low
February 27, 2020

The recent decision by National Savings and Investments to cut interest rates on its products reminds us of a depressing fact – that savers face the prospect of negligible returns on safe assets, and losses once we take account of inflation. This does not, however, mean we should ditch cash and buy equities in the hope of higher returns.

I don’t say this because there’s a chance that interest rates will rise later this year if we see a post-Brexit pick-up in economic activity; if the world economy recovers once the coronavirus is under control; or if we see a significant easing of fiscal policy. All these are possible – no more than that – but we should not base our asset allocation on futurology alone.

Instead, three things tell us to be wary of buying equities merely because interest rates are low: theory, experiments and history.

First, theory. Why are real interest rates negative? It’s because the Bank of England, and other central banks and bond investors, think the outlook for economic growth is weak and risky. Why is this a reason to buy equities?

Let’s put this another way. What should determine your split between cash and equities are the risk premium on equities relative to cash and the riskiness of equities. There’s no reason to suppose that low interest rates increase the risk premium, and even less to suppose that they make equities safer. So there’s no reason why they should change your asset allocation.

In fact, experimental evidence warns us not to do so. Harvard University’s Carmen Wang and colleagues asked laboratory subjects to divide money between two hypothetical assets – a safe one yielding 5 per cent and a risky one with an expected average return of 10 per cent and standard deviation of 18 percentage points. They then offered a choice between a safe asset paying 1 per cent and a risky one paying 6 per cent, again with a standard deviation of 18 percentage points. They found that people put significantly more into the risky asset when returns were low. This is clear evidence of a 'reach for yield': investors seek higher risk when returns on safe assets are low.

This is dangerous. If investors pile into risky assets merely because of revulsion at safe ones, the price of those assets will rise too much, which increases the chances of a sharp fall. And, indeed, this is just what other experiments tell us. Tobias Rotheli at the University of Erfurt has found that when he removed a safe option from subjects they took on even more risk than necessary and so suffered greater losses.

History tells us a similar story. The Bank of England has data on interest rates, share prices and inflation going back to 1700. These tell us something important.

For one thing, they show that negative real interest rates are not as unusual as you might think. If we define these as Bank rate in December minus CPI inflation in the same year there have been 85 years of negative rates. That’s more than a quarter of the time. Negative real rates seem unusual to us not so much because they really are, but because, for many of us, our impressions are distorted by our formative years. Real interest rates were abnormally high in the 1980s, and this causes us to regard them as more normal than they really are.

And for another thing, they show that real rates have historically been astonishingly volatile. Which warns us not to take stability in them for granted. 

But what happens to share prices in years after negative rates?

On average across these 85 years they have actually fallen slightly, by 0.7 per cent after inflation. That compares with an average rise of 3.3 per cent in years after real rates have been positive (in fact, the numbers are pretty similar if we measure real rates by the nominal Bank rate minus inflation in the following year).

Yes, equities on average outperform cash when real rates are negative. But their absolute returns have been poor: in fact, they fell in 45 of those 85 years and by more than 10 per cent in 20 of them.

The evidence, therefore, is clear – or at least as clear as it can be in noisy data. Low real interest rates don’t give us much reason to buy shares. It might be reasonabe to buy because valuations are reasonable – the dividend yield is well above average – or because sentiment is depressed. Just do not buy them because interest rates are low.