Join our community of smart investors

The QE threat

The poses the question: how can an event which everyone has seen coming for months possibly unsettle markets? Surely the prospect of QE ending should have been discounted by investors long ago.

Not necessarily, because of something first pointed out by Olivier Blanchard and Mark Watson back in 1982. They showed that it's possible to get bubbles in share prices even when investors are rational and far-sighted. Even if everyone believes there might be a crash in the future, they might continue to buy overpriced assets if they believe that prices will rise sufficiently in the short term to compensate them for that risk of a slump.

Two things make such rational bubbles especially likely.

One is an environment of low returns on assets generally. When interest rates are low even a modest expected return on a bubbly asset will be sufficient to compensate for crash risk. And in a world of secular stagnation, taking on such risk might be the only way to achieve decent returns. Low returns lead to a 'reach for yield'.

The second lies in institutional incentive structures. We retail investors, many of whom are concerned merely to protect capital, might think it too risky to bet on being able to sell before a crash. For many institutional investors, however, things are different. Bankers and fund managers are often judged on quarterly performance. The trader who sells too soon in anticipation of a crash risks underperforming his peers, thus losing his bonus or even his job. His incentives therefore tell him to ride the bubble, even if he knows there is trouble coming.

This was what Charles Prince, then chief executive of Citigroup meant when he said in 2007: "When the music stops, in terms of liquidity, things will be complicated. But as long as the music is playing, you've got to get up and dance." Yes, the music stopped soon after he said that and he lost his job. But the fact that he left with a multi-million dollar payout suggests he wasn't being stupid.

To change the metaphor, financial markets are like a game of chicken. The winner is not the man who does the safe thing, but the one who swerves only at the very last minute - even if doing so carries the danger of colliding fatally.

So, rational bubbles are possible. It is theoretically possible that markets have not discounted the fact that the Fed will soon withdraw monetary support for share prices.

But where's the evidence that this is actually happening? Valuations are ambiguous. Yes, the price-earnings ratio on the S&P 500 is above its 10-year average, at 18.9 against 16.7. But it is below its post-1988 average, of 20.9. And although the cyclically-adjusted PE ratio compiled by Yale University's Robert Shiller is well above its long-run average (that is, since 1881), it is close to its post-1990 average.

What we do know, though, is that there are precedents for bonds and equities to sell off because of a widely expected normalisation of monetary policy. In 1994 the Fed started to raise interest rates after cutting them in response to the Savings and Loans Crisis. Although everyone had expected the move for a long time, bond and equity prices fell. Nor did markets learn this lesson. When the Fed warned last spring that it would begin to taper QE, emerging markets shares fell more than 10 per cent even though nobody should have been surprised. (They have since recovered those losses - although it is moot whether this is any comfort or not.)

Personally, I don't know whether this history will repeat itself when the Fed stops QE. It's possible that stronger economic growth will replace easy money as a support for shares. What I do know, however, is that if a widely expected return to normal does cause serious trouble, then financial markets are grossly dysfunctional. A system can be stupid even if all its individual members are rational.