Who let the dogs out?

As regular readers will know, we've been tracking a momentum portfolio based on the quarterly performance of the 10 best and worst-performing blue chips of the previous three months, starting off when the market peaked in June 2007, four years ago.

Once again the long stock picks, based on the best-performing shares of the previous three months, have put in a very impressive performance, up 13 per cent relative to the market. But our momentum strategy has again failed to identify losers. The short picks, based on the worst-performing stocks of the previous three months, have outperformed the market by 5.9 per cent. In fact, our momentum portfolio's shorting record is so woeful it provides compelling reasons for adopting a so-called dog strategy of buying underperforming shares.

All academic

Although academic research has demonstrated the merits of short selling the market's worst performing shares, over the four-year period we've monitored you would have done far better from buying the losers instead of short selling them. Indeed, the losers' portfolio has outperformed the FTSE 100 in seven of the 12 three-month periods we've looked at (see table). That includes the most recent three months to 15 June when the short portfolio dropped just 0.8 per cent compared with the FTSE 100’s 6.7 per cent fall. What's more, in six of the 12 quarters we’ve monitored you would have produced better returns by buying the 10 worst-performing shares of the previous quarter than buying the best performers.

That could be to do with shifting risk appetite. The VIX index, which measures nervousness in the market, has been falling fast since it peaking in late May. So it’s worth noting that the last period of substantial index-beating performance from the short positions happened while the VIX fell sharply over a prolonged period starting in March 2009, as investors regained their appetite for riskier shares that had probably been oversold.

So, looking at the short portfolio, which is mainly made up of consumer-related stocks and resources companies, it isn't hard to imagine how any further improvement in market sentiment could lead to share prices again heading higher fast.

Playing to win

That said, buying the winners has still proved a better bet over the last four year than buying the losers. In the three months to 15 June 2010 the winners portfolio produced a 6.3 per cent rise and two of the stock picks, Aggreko and ARM Holdings, were also among the 10 best-performing blue-chip shares in the period. However, as always the performance of the portfolio on a stock-by-stock basis was extremely mixed.

Over four years the winners portfolio has delivered nine quarters of outperformance (see table) and produced an impressive 33.4 per cent capital uplift compared with the FTSE 100's 22.5 per cent fall. This compares with the 14.9 per cent loss that investors would have made buying the blue-chip losers over the same period. And even though the short portfolio has produced a stunning 140 per cent return since the market bottomed out in March - even better than the 104 per cent from the longs - that's not all that surprising given the catastrophe being priced into many cyclical shares at the height of the downturn.

All in all, while the performance from our momentum strategy remains far from perfect, there's enough evidence to suggest it's working, with the performance of the long positions increasingly noteworthy. Both the long and short stock picks for the coming quarter are set out below.

Source: Datastream