FTAlphavilleFT Alphaville is a 2008 Webby Award Winner

• News
• Markets Live

What will the quants do next? Judging by presentations at the CFA conference in Vancouver, many think they have worked out what happened to them last August, and believe they have learned the lesson.

The main problems - known well enough by now - as presented by various quants here in Vancouver over the last two days seem to be as follows:

First, the quants had too much leverage (and available data could not tell them that this was going on).

Second, they crowded into trades, although this worked differently from the normal perception. The overlap in their holdings of stocks, which would have shown up in quite a lot of publicly available data, was not that great. Rather, they were crowded into particular “factors” - such as staying long in stocks with a low price-to-book ratio. This created the downdraft for them when the market turned in early August.

Third, they overconfident and continued to bet that relatively cheap stocks would beat relatively expensive stocks, even though by 2004 the spread between the two, after half a decade of outperformance by value strategies, was historically low. As Vadim Zlotnikov of Alliance Bernstein put it: “Value spreads were very low but they continued to pile into value arbitrage trades. They were betting that LBOS that had never previously made sense would continue.”

In spite of the fascination with ‘Black Swan’ events that is going through the risk management community, many now admit that this was not a “one in 10,000 years event,” even though some risk managers tried to describe it that way at the time.

Mr Zlotnikov said: “I strongly disagree that subprime was either unanticipated or unknown. How many of you didn’t know that credit spreads started to widen in August? This was the most forecast, predicted and discussed event. And you’re telling me this was a black swan?”

Sergio Focardi, of the Intertek Group, said the subprime crisis was “relatively easy to predict” and that managers should be careful to distinguish “events that are black swans because you want them to be black swans”.

Quants say that they have now brought their leverage under control. The future now appears to involve looking for new factors to model, and looking in more detail at what drives performance in different industries. This will probably involve trying to model balance sheet factors, and moving closer to what fundamental managers do.

Mr Zlotnikov cautions that there is a risk of data-mining here, with a temptation to “overfit” models to limited data. But mathematical models could still make a difference to the way fundamental managers operate. And the risk of crowding should be reduced.

One other well-known quant, from a large firm, put it this way: “It’s just being systematic - there are lots and lots of factors that you can use going forward that are not particularly crowded. We are all scattering from the centre, but there’s no reason why we should all scatter in the same direction, and I don’t think we are.”

This quant admitted that “quants are now behaving a lot more like fundamental managers in that respect.

Short View columnist John Authers is blogging for FT Alphaville from Vancouver at the annual gathering of the CFA Institute