Join our community of smart investors
OPINION

When the Fed moves

When the Fed moves
September 17, 2014
When the Fed moves

Theory is ambiguous. On the one hand, because almost everyone expects a rise this prospect should by now be already embedded into share prices. But on the other hand, professional traders are incentivized to have short-term time horizons and these can generate “rational bubbles” in which asset prices don’t factor in future events.

With theory so ambiguous we need empirical evidence. And this suggests that a rise in US rates is slightly bad for UK equities. Since January 1985 the correlation between the fed funds rate and the dividend yield on the All-share index has been positive, at 0.24. This means higher US interest rates are, on balance, associated with higher dividend yields - which means lower share prices.

Whilst this correlation is statistically significant, it is not hugely economically significant. It implies that the rise in the fed funds rate which the market expects next year – to 0.76 per cent by December 2015 – will take only 1.4 per cent off the All-share index. This is only the difference between a slightly bad and slightly good day for the market, which is hardly enough to justify a significant change in one’s asset allocation.

However, this correlation is an average for the last 30 years. And averages can disguise big variations. Sometimes the correlation between the fed funds rate and All-share yield has been strongly positive, such as in 1994 when a universally expected rise in US interest rates after a financial crisis led to falling share prices. But at other times – for example in the 00s – the correlation has been negative.

There’s a reason for this. Very often, the circumstances in which interest rates rise are those in which the economy is doing well and so share prices are rising. For example in 2003-05 share prices and the economy recovered after the tech crash and the Fed responded by raising interest rates. And in 2008 the financial crisis caused shares to slump and the Fed to slash rates. It would be odd to claim that rising rates in 2003-05 and falling ones in 2008 caused shares to rise and then fall.

To get a clearer picture of the link between interest rates and UK share prices we should therefore control for the state of the US economy. The easiest way to do so is to control for changes in the unemployment rate. Doing this shows that the impact of US rate rises upon UK equities is greater than the raw correlation between the two would suggest. Post-1985 relationships show that, for a given unemployment rate, a percentage point rise in the fed funds rate is associated with a 0.17 percentage point rise in the All-share dividend yield. This implies that the rise in the fed funds rate which the traders expect next year would wipe 3.3 per cent off shares. Again, though, this isn’t very much.

Optimists might point out here that we shouldn’t take the unemployment rate as given. Forecasts of a rise in US rates are predicated on the belief that the US economy will continue to grow and so unemployment will fall. Post-1985 relationships tell us that a half point fall in the unemployment rate would raise the All-share index sufficiently to offset a percentage point rise in the funds rate. With the Fed expecting the jobless rate to fall by this amount next year, this should more than offset the adverse effect of the rise in the funds rate which the market expects.

Sadly, however, things aren’t quite so simple. If unemployment falls it will do so gradually throughout the year whereas even the most gradual tightening of monetary policy would see some jumps in the funds rate. This alone would generate short periods when equities worry about monetary policy.

Worse still, post-1985 relationships might break down. There’s been a positive correlation between US economic activity and share prices because of a simple sentiment effect; a stronger economy increases investors’ appetite for risk. But there’s no reason in theory why this must always be the case. For example, if growth expectations are mean-reverting, high actual growth should depress expected growth; a long or strong expansion merely brings us closer to the next recession. Before the mid-90s, this mean-reversion mechanism more than offset the sentiment mechanism, generating a negative correlation between economic activity and share prices. A repeat of this would mean that investors would actually be disquieted by a further fall in unemployment. And this would cause rising rates to do more damage to shares.

It is therefore possible to believe that higher US interest rates would seriously hurt UK equities. To do so, though, requires one to believe both that historic relationships will break down and that financial markets are so informationally inefficient that they cannot discount in advance even obvious and widely expected future events. Such a worldview calls into question both the desirability of free financial markets and the possibility of having any useful knowledge about economies – because if we can’t rely upon history or theory, what can we know? In this sense, such a view has more disturbing implications than a mere fall in share prices.