In particular, financial trends start from rational causes but are often prolonged to a point where prices swing far into the irrational. In the context of a market, a feedback loop can be understood simply as the price-amplification that occurs whenever profits are reinvested. A trader profits, reinvests those profits and makes more profits. Others bring in new funding in hope of emulation.
That's on the upside. On the downside, the feedback loop consists of selling that triggers more selling. Quite apart from panic situations, this can occur due to program-trading. One trader's stop-loss is triggered and he sells. That pushes prices down to hit a second stop-loss where another trader sells. That triggers margin calls that cause a third wave of selling, etc.
In the past few years we have definitely seen reflexivity at work. By May 2006, when the Nifty was trading at 3,650, valuations no longer justified buying Indian equities. The PE ratio was 20 and T-Bill yields were over 7 per cent. At that point the market crashed there was a wave of panic selling.
But the correction was short-lived and prices surged again. The Nifty eventually rose 72 per cent to peak in January 2008 at 6,350 with an associated PE of 29. Throughout 2007 we saw reflexive buying as prices stayed consistently above fair-value.
Since then, index levels have fallen 35 per cent. We can't say that the correction has been reflexive because equity valuations are still rich at the current Nifty PE
of 17. Reflexivity would involve prices correcting to below fair value.
Incidentally the investors who exited in panic in May 2006 haven't necessarily done badly. In theory, those who stuck it out and liquidated in January 2008 earned 70 per cent return in 19 months.
In practice, relatively few people managed to liquidate anywhere close to the peak. Those who stayed continuously in equity have earned a total of about 12 per cent in two years any FD-holder has done better than that.
If the current bear market turns out to be reflexive, prices could be driven a long way below fair value. What is fair value now? Assume a risk-free rate of 10 per cent this is conservative with WPI at 11.4 per cent. Assume that Nifty earnings will grow at an average of about 15 per cent in 2008-09. A value investor would say a PE of 10 is justified on the basis of high interest rates while a PEG follower would be prepared to go up to PE 15.
If we allow for some degree of "fuzziness", the Nifty could therefore bottom in a band between PE 10-15. That works out to a price-band of Nifty 2,800-3,300. If the bear market turns reflexive, prices should drop till at least the lower end of that band. This is a frightening prospect but it is possible given history.
This would be the equivalent of about 55 per cent correction from the market peak and that has happened. A big bull run followed by a major correction followed by another big bull market is a familiar Indian pattern.
If you do decide to back the possibility of this bear market being reflexive, then you should be looking to exit now and try to re-enter below say, 3500. One danger is that if the market isn't actually reflexive you might wait forever!
If you aren't certain about the reflexivity or about your ability to time the exit-re-entry, there is not much point in selling at current prices. If you hold, you stand to lose another 20-25 per cent. If those are un-booked "paper losses", a 2-3 year perspective suggests eventual gains will be in the range of 50 per cent or more as the market gradually recovers back to 6,000 levels.
An average investor who has stayed committed to equity thus far should now be averaging down and buying on every dip rather than selling. A very good trader may be able to get out now and re-enter near the bear market bottom to maximise theoretical returns.
Most people simply are not that good.