Recent history suggests it would be good news. Since 2000 there's been a positive correlation between changes in 10-year gilt yields and changes in the All-Share index. Both tend to rise in good economic times and fall in bad.
My table shows what changes in 10-year yields have meant for FTSE sectors. It shows the annual change in some interesting sectors associated with a one percentage point rise in gilt yields, controlling for changes in the All-Share index. So, for example, a percentage point rise in 10-year yields (given a level of the All-Share index) has been accompanied by an 8.9 per cent rise in construction stocks but an 11.5 per cent fall in tech stocks.
| Change in FTSE sectors associated with a one percentage point rise in gilt yields | |
|---|---|
| Oil & gas | -3.1 |
| Construction | 8.9 |
| Transport | 8.8 |
| Beverages | -3.9 |
| Pharmaceuticals | -5.6 |
| General retailers | 10.1 |
| Travel/leisure | 7.9 |
| Banks | 6.4 |
| Non-life insurers | -5.6 |
| General financials | 7.0 |
| IT | -11.5 |
| Based on annual changes since Jan 2000. Controls for changes in the All-share index | |
To see what's going on here, consider four mechanisms whereby rising bond yields might affect shares.
The first is that rising yields are often a sign of expectations of stronger economic activity. Such expectations typically benefit cyclical stocks the most, such as retailers and construction.
Secondly, higher long-term interest rates mean that a higher discount rate is attached to expected future cash flows. This hurts growth stocks because these (by definition) offer more expected future cash flows than others. It's for this reason that tech stocks should do badly (controlling for moves in the general market) when yields rise.
Thirdly, rising bond yields will hurt companies that hold lots of bonds. This might be why insurers have done badly when yields have risen. You might think that banks would also suffer on this account. But history shows that this adverse effect has been more than offset by our first mechanism; better cyclical conditions encourage investors to take on riskier stocks such as banks.
Fourthly, some stocks are a little like bonds in that they offer stable-ish income. You'd expect these to do badly when bond prices fall simply because prices of things that are like each other should rise and fall together. This might explain why oil majors, pharmaceuticals and beverages have tended to do badly when bond yields rise.
For many stocks, these four mechanisms offset each other, with the result that they aren't significantly correlated with bond yields.
The message here seems clear. Insofar as history is a guide, a sell-off of bonds would be bad for tech stocks but good for many cyclicals.
But is history a guide? There's a saying in economics: never reason from a price change. The effects of a price change vary according to why the price changed. History tells us that bond yields have often changed because of expectations of stronger growth. However, bond yields might rise for other reasons. If they do, my table might not be a reliable guide to equity strategy.
For example, yields might rise because investors expect central banks to raise interest rates by more than they previously anticipated, or because they fear a rise in inflation without stronger growth - say because commodity prices rise or because they believe the global output gap has shrunk. In such cases, we might not see a positive response of cyclical stocks to higher yields.
For this reason, I suspect that the best equity defences against rising yields might not be cyclicals but rather defensive stocks - such as utilities, tobacco and food manufacturers - which have no strong correlation with bonds. Given that even these take on market risk, however, it might be that even better protection would be plain old cash. There are reasons why so many reasonable people are happy to hold this despite negative real interest rates.
