Momentum stocks soar

Thanks largely to this, momentum has done very well in the past 12 months. Its 19.5 per cent rise means it has beaten all but six funds in Trustnet's database of all companies' unit trusts. This suggests that, relative to a no-brain stockpicking strategy, most fund managers' efforts have actually subtracted value.

The reason for momentum's success is simple. Momentum investing in June means we avoided some terrible losses among commodity stocks, while getting into some nice risers such as Betfair (BET), Dart (DTG) and JD Sports (JD.).

It's not just positive momentum that's done well, though. The converse strategy - negative momentum - did atrociously. The 20 biggest fallers in the 12 months to June fell by a further 25 per cent in Q3. This is simply because our negative momentum portfolio was biased to commodity stocks, holding Tullow Oil (TLW), Lonmin (LMI) and Premier Oil (PMO) among others. The investor who went long of our positive momentum portfolio and short of our negative momentum portfolio in June would have made over 35 per cent in the third quarter, before dealing costs.

You might object - reasonably - that shorting stocks is tricky and expensive. However, I'm not advocating that you do so. Simply avoiding negative momentum would have at least saved you from some disasters.

All this suggests that momentum works well during bubbles. As they inflate, positive momentum gets us into them. And after they have started to deflate, it gets us out of them, while negative momentum alerts us to potential losers. It is only around the time that the bubble starts to deflate that momentum fails.

This, however, is not the whole story. Our momentum strategies - both positive and negative - have done well for years, a period that covers both the rise and fall of commodity stocks. This corroborates a wealth of evidence from around the world that momentum investing succeeds.

Not all strategies have done as well as momentum, though. Value investing has had a poor time, with our equal-weighted basket of the 20 highest yielders losing 12 per cent in the quarter. Again, this is because these were victims of the slump in emerging markets and commodity stocks: our portfolio included Standard Chartered (STAN), BHP Billiton (BLT) and Vedanta (VED).

This reminds us that value stocks carry two risks.

One is cyclical risk. A high yield can be a warning that a stock is vulnerable to a recession. Back in 2008 it was builders and mortgage lenders that were on high yields and they got clobbered. Similarly, a few months ago it was commodity stocks that were, and they have suffered due to China's downturn.

The second risk is momentum risk. Stocks are often on high yields simply because they have fallen in price. But if investors underreact to bad news - and the long-term evidence suggests they often do so - then even quite a large fall will still leave a share overpriced. If so, it will subsequently fall further. This has been the fate of many high-yielding commodity stocks.

There's a clear warning here for the many of you who invest for yield. Doing so isn't a bad idea; in normal times, high yielders do outperform to compensate for cyclical risk. However, a yield can be a sign that a stock has negative momentum. And such stocks are to be avoided.

Another victim of the commodity crash has been mega-caps. These included many mining stocks - including Glencore (GLEN) - and so have suffered. Aside from the short-term possibility that they will continue to suffer some negative momentum, I would be loath to regard this as part of a long-term trend. For one thing, mega-caps now include fewer miners and more defensives, including National Grid (NG.), GSK (GSK) and Diageo (DGE). And for another, I find it implausible that big stocks will underperform in the long run. If they did, we would eventually end up with a market in which all companies were the same size - which is surely unlikely.

Yet another loser has been high-beta stocks. These fell 12.5 per cent in Q3. You might find this unsurprising; high-beta stocks should underperform a falling market.

What is surprising, though, is that they have not outperformed a rising one. In fact, they have fallen in the past five years even though the market generally has risen.

Here's a theory as to why. In principle, a bullish investor who wants big exposure to equities should borrow and so take a leveraged position in the market generally. However, many investors cannot do this, either because of a lack of credit or institutional constraints. Such investors instead express their bullishness by buying high-beta shares - ones they expect to outperform a rising market. This, however, means that such stocks will be overpriced and so will actually underperform. For this reason, Andrea Frazzini and Lasse Heje Pedersen at AQR Capital Management have shown that policies of "betting against beta" have paid off not just in equities but in commodities and bonds, too.

The counterpart of high-beta stocks doing badly is of course that low-beta ones should do relatively well. Again, this is what we saw in Q3. And again, while Q3's performance is only to be expected in a falling market, what's less expected is that our low-risk portfolio has actually outperformed a rising market over the past five years. As with momentum, this too corroborates the long-term international pattern: like momentum, defensive stocks do better than they should on average.

One contributor to this, I suspect, is benchmark risk. For professional fund managers, defensive stocks are ill-named. They are likely to underperform if the general market does very well - and such underperformance could cost the fund manager his bonus or even job. This threat causes him to underweight such shares with the result that they are underpriced on average. This makes them especially attractive to retail investors. We don't have to worry much about underperforming a bull market and so can afford to take on benchmark risk. In effect, we can get compensation for taking on a risk that doesn't matter to us.

Herein, though, lies an important point. You might think that the aggregate market's performance in Q3 has been unusual. And it has to some extent: the FTSE 350's 6.7 per cent fall has been a slightly greater than one standard deviation event, which is the sort of thing we'd expect to see around one quarter in every 10.

What's not been so unusual, though, is the pattern of returns. The good performance of defensives and momentum, and poor performance of high beta, are all consistent with the long-term historical pattern. And, given heightened (global) cyclical risk and strong negative momentum effects, there is nothing odd about value's underperformance.

This tells us something. We cannot foresee the future: the fall in the aggregate market might have been a surprise - although not a costly one if you followed the 'sell in May' rule! But we do know the past. And this tells us that some strategies do work on average and over the long run. In this light, there's nothing surprising about the pattern of returns in the past quarter.

Our latest portfolios:

Momentum (the 20 biggest risers in the past 12 months): Barratt Devs, Betfair, Card Factory, Cineworld, Crest Nicholson, Dart, Greggs, JD Sports, Marshalls, Moneysupermarket, NMC Health, Novae, Onesavings Bank, Rank, Redrow, Regus, Rightmove, Synthomer, Taylor Wimpey, Ted Baker.

Negative momentum (the 20 biggest fallers in the past 12 months): Aggreko, Anglo American, Antofagasta, BHP Billiton, Drax, Eurasia Drilling, Evraz, Genel Energy, Glencore, Hunting, Ophir Energy, Pure Circle, Rio Tinto, Rolls-Royce, Rotork, Serco, Std Chartered, Tullow, Vedanta, Weir.

Value (the 20 highest yielders): Aberdeen AM, Anglo American, Ashmore, BHP Billiton, BP, Carillion, Electrocomponents, Esure, Eurasia Drilling, GSK, Glencore, HSBC, Morrison, Phoenix Gp, Redefine, Rio Tinto, Rotork, Soco, Std Chartered, Vedanta.

High beta (the 20 highest beta stocks): 3i, Aberdeen AM, ASOS, Barclays, Bodycote, Eurasia Drilling, Glencore, Henderson, Inchcape, IPF, Keller, Michael Page, Mondi, Northgate, Schroders, SIG, St Modwen, Thos Cook, Travis Perkins, Vedanta.

Low risk (the 20 lowest beta stocks, subject to no more than three from any one sector): Abcam, Betfair, Breedon Aggs, Cineworld, Dechra, Dignity, Diploma, Domino's Pizza, Euromoney, First Gp, Imperial Innovations, James Halstead, Jardine Lloyd Thompson, National Grid, Nichols, Synergy Health, Tate & Lyle, Telecity, Telecom Plus, Vectura.

Mega caps (the 20 largest stocks): AstraZeneca, Barclays, BAT, BHP Billiton, BP, BT, Diageo, GSK, HSBC, Imperial Tobacco, Lloyds Banking, National Grid, Prudential, RBS, Reckitt Benckiser, Rio Tinto, Royal Dutch Shell, SABMiller, Unilever, Vodafone.

All portfolios are equal-weighted and drawn from UK shares excluding investment trusts with a market capitalisation of over £500m, taken from the IC's stock screen.