Many of you are worried by the possibility of higher inflation. What should worry you more, though, is something else – the danger of higher interest rates.
I say this because it is in principle easy to protect ourselves against inflation – albeit at a cost – simply by buying index-linked gilts. The problem is that doing this would not work if the Bank of England were to raise interest rates significantly, or if markets were to expect them to do so. If this were to happen, index-linked gilt yields would rise, imposing capital losses upon anybody not planning on holding them to maturity.
Note that the issue here is real interest rates. Because it is these that tighten monetary policy, the Bank would have to raise rates by more than the rise in inflation if it wanted to reduce inflation.
Worse still, if bond yields (real or nominal) rise in the US another supposed protection against inflation would also let us down – gold. The mechanism here is simple, even though it is often forgotten. Gold pays no interest. Which means that when returns on bonds and cash rise, we forego more income when we hold it. The metal therefore becomes less attractive, and so its price should fall. And this is just what happens. Since January 1988 there has been a correlation of minus 0.8 between the five-year US Treasury yield and the sterling price of gold. That means that changes in US yields alone explain almost two-thirds of the variation in the gold price in the last third of a century.
While it is obvious that higher interest rates (or expected interest rates) would be terrible for gold and bonds – and house prices too - their impact upon equities is much tougher to call.
On the one hand, the circumstances in which rates rise would be good for them: the same stronger economic activity that raises inflation expectations also boosts earnings expectations and appetite for risk.
On the other hand, though, rising rates could well trigger fears of a reversal of the “reach for yield” – the tendency for people to have bought equities simply because returns on cash are lousy. And they could also increase fears of slower economic growth.
Anyone who takes a strong position on which of these two mechanisms will win out is showing more courage than knowledge.
Which brings us to an even bigger problem. For most of this century it has been easy to spread equity risk because equities and gilts have been negatively correlated and so losses on equities have been offset by profits on gilts. The 10 worst months since 2000 for equities, for example, all saw gilts rise. With gold and often foreign currency also doing well in times of distress for equities, this means that running well-diversified portfolios has been a doddle: they’ve given us good returns with little risk for even less effort.
Rising interest rates, however, could change this.
My chart shows why. It shows the correlation between returns on gilts and the All-Share index, measured by monthly returns over the previous five years. This correlation has been negative for most of the past 20 years. But it hasn’t always been so. Until the late 90s the correlation was positive, often strongly so, with losses on equities often accompanied by losses on gilts too. Back then, diversification didn’t work so well.
So, what changed? My chart suggests one big reason – interest rate volatility, as measured by the volatility of monthly changes in five-year gilt yields. As this volatility fell in the late 1990s and early 2000s, so too did the equity-gilt correlation. The rise in yield volatility in the second half of the 2000s saw the equity-gilt correlation rise. And in recent years yield volatility and the equity-gilt correlation have both fallen.
What’s going on here is actually quite simple. Volatile interest rate expectations cause equities and gilts to move together: if investors expect significant rate rises they sell both, and if they expect significant falls they buy both. Cheaper or dearer money drives both assets. But when investors expect stable interest rates, this mechanism doesn’t operate and so other forces determine the equity-gilt correlation. Variations in appetite for risk, for example, cause the two to move in opposite directions.
This brings us to the danger. If interest rate expectations do become more volatile, we should expect equities and gilts to become positively correlated again. Which means we cannot rely upon gilts to protect us from losses on shares (or vice versa).
This threatens the return of a danger we’ve rarely seen this century – that assets could fall at the same time, meaning that diversification does not work so well.
Luckily, there’s a simple solution to this risk – cash. Not only does it gain from rising rates but also it protects us from the risk of simultaneous losses on other assets. Yes, cash has been dead money for a long time. But it might not remain so.
