Defensives wobble, momentum surges

This is consistent with the possibility that, a few weeks ago, investors wised up to the virtues of defensives and so bought them, and in doing so raised their prices to a level from which subsequent returns will be mediocre. If this is the case, then perhaps the great long-run performance of lower-risk stocks is now over.

It is, however, also consistent with another possibility - just dumb luck. Since it began in its current form in 2005, the volatility of relative returns on our low-risk portfolio has been 1.38 percentage points per week, or five percentage points per quarter. Its underperformance in the second quarter is therefore only a fraction of a standard deviation. It's far too little, too soon, to say with any confidence whether the anomaly really has vanished.

What we can say - with a little more confidence - is that the longer-term evidence supports the theory that defensives are underpriced on average and high-beta stocks are overpriced. For example, over the last seven years, our low-risk portfolio has risen by 34.7 per cent - beating the FTSE 350 by 23.7 percentage points while our high-beta portfolio has risen just 1.4 per cent. This is consistent with the possibility that the incentive structures faced by professional investors biases them against holding defensives, leaving them consistently underpriced, and so offering good returns, while at the same time encouraging the holding of high-beta stocks, causing these to be overpriced.

However, not every good long-term strategy had a poor second quarter. Our momentum portfolio rose by 3 per cent in the quarter, cementing its great long-run performance. What has been even more spectacular, though, is the performance of our negative momentum portfolio - the 20 worst performers in the previous quarter. Thanks to some big drops in resources stocks such as Evraz, Petropavlovsk and New World Resources, this fell by 21 per cent in the second quarter. This means that a long-short momentum portfolio (going long of positive momentum and short of negative) would have made 24 per cent in the quarter, and 39.8 per cent in the last 12 months.

This same momentum effect might help explain the poor performance of our loser portfolio - the biggest losers over the last three years. Note, though, that there's a wrinkle here. All my portfolios are rebalanced every quarter. It's possible that investors who hold losers for a longer period would see better returns eventually. However, such returns would be compensation for poor short-run performance and in this sense would be a reward for taking on near-term volatility.

With momentum doing so well and defensives slightly poorly in the second quarter, does this mean we should favour momentum over defensives as a long-term strategy?

I'm not sure it does. We have good reason to suspect that defensives' long-run performance is due not just to cognitive biases, but to the structure of the financial services industry. However, momentum's performance is more likely due merely to cognitive bias - to the tendency for investors to underreact to good and bad news. This difference matters. Cognitive biases are, in principle, corrigible; people should learn from their past errors. This means that the momentum anomaly is more likely to disappear than the more deeply-rooted defensive anomaly. Indeed, you could read momentum's recent good rise as a sign that investors have learnt the error of their ways and so bid up the prices of momentum stocks.

Our negative momentum portfolio underperformed a whopping minus 21 per cent in the second quarter of 2013, yet our positive momentum portfolio produced a return of 3 per cent

So far, the portfolios I've discussed seem to refute the standard theory of asset pricing, the capital asset pricing model (CAPM); over the last five years, the performance of momentum, high-beta and defensive portfolios have all been inconsistent with its predictions.

But this doesn't mean the model is a complete write-off. Our portfolio of stocks with the most idiosyncratic risk has done badly. This is consistent with the CAPMs' prediction that investors are not rewarded for taking on diversifiable risk, but only systematic risk - the sort that we cannot so easily avoid.

A good example of the importance of systematic risk is the behaviour of our value portfolio - the highest yielders in the FTSE 350. Over the last five or seven years, this has done badly. This is because value stocks, very often, carry cyclical risk and this risk materialized horribly in 2007-09; back then, remember, high yielders included housebuilders and mortgage lenders such as Bradford & Bingley and Northern Rock and these were massacred by the financial crisis.

By contrast, over the last 12 months expectations for the economy have improved, which means cyclical risk has paid off. So high yielders have outperformed the market.

There's another portfolio that carries cyclical risk - small caps. These too have benefited from the diminution in economic pessimism over the last 12 months.

However, this does not mean small caps and value stocks are similar. Small caps were much less devastated during the 2008 crisis than were value stocks, as they tended not to be mortgage lenders and housebuilders. Thanks to this, small caps have done better over the last five years than value stocks.

Whether this means small caps will remain a better bet in the long run than value stocks is, however, questionable. The fact that small caps did badly for a lot of the 1990s warns us that they are not a one-way bet.

Indeed, that episode warns us of a risk to both our momentum and defensives portfolios. The thing is that markets are neither efficient nor inefficient, but adaptive. Think of them as being like ecosystems in which species (strategies) compete for food (profits). If a food source becomes available, the species will feed on it and multiply. But as it multiplies, it depletes the food source with the result that the species goes into decline.

This is the story of small-cap investing. In the 1980s, researchers pointed out that there were big profits to be made from small caps. Small-cap strategies then proliferated. But their activity raised the prices of small caps too far, so returns in the 90s were poor. As small caps underperformed, investors lost interest in them. But this led to small caps being underpriced in the late 90s, and so they took off again. This outperformance was revered in 2007-08, but later re-emerged.

In other words, there are longish cycles for stock market strategies. In recent years, momentum and defensives have enjoyed the upswing of these cycles. Will they continue to do so? Knowledge of the future is, of course, a contradiction in terms. My hunch is that defensives are more likely to do so than momentum. The point of my no-thought portfolios is to check whether this will be correct.

Our new no-thought portfolios

Momentum (the FTSE 350's biggest risers in Q2): Capital & Counties, Countrywide, Dixons, easyJet, Homeserve, Keller, Kingfisher, Lloyds Banking, Ocado, Paypoint, Perform, Persimmon, Sports Direct, Supergroup, Ted Baker, Telecom Plus, Thomas Cook, Wetherspoon, Whitbread, Xaar.

Negative momentum (the biggest fallers in the FTSE 350 in Q2): African Barrick, Anglo American, Antofagasta, AZ Electronic, Bwin, Centamin, Evraz, Ferrexpo, FirstGroup, Fresnillo, Glencore, Hochschild, Imagination Technologies, Kazakhmys, Ophir Energy, Oxford Instruments, Polymetal, Randgold, RPS, Salamander.

Idiosyncratic risk (the 20 stocks in the FTSE 350 with highest stock-specific volatility in Q2): African Barrick, AZ Electronic, Centamin, ENRC, Evraz, Ferrexpo, FirstGroup, Fresnillo, Halfords, Hochschild, Imagination Technologies, IPF, Kazakhmys, Lonmin, Man, Ocado, Polymetal, Salamander, Thos Cook, Xaar.

Losers (the biggest fallers in the 350 in the three years to June 30): African Barrick, Anglo American, Antofagasta, AZ Electronic, Bwin, Centamin, Evraz, Ferrexpo, FirstGroup, Fresnillo, Glencore, Hochschild, Imagination Technologies, Kazakhmys, Ophir Energy, Oxford Instruments, Polymetal, Randgold Resources, RPS, Salamander.

High beta (the 20 highest beta stocks in the FTSE 350, based on monthly returns in the last five years): 888 Holdings, Afren, African Barrick, Centamin, De La Rue, Dixons, Enterprise Inns, Ferrexpo, Hochschild, IP Group, Kcom, Kenmare, Ophir Energy, RBS, St Modwen, Supergroup, Taylor Wimpey, Telecom Plus, Thomas Cook, Unite.

Value (the highest yielders in the FTSE 350): African Barrick, Amlin, AstraZeneca, Aviva, Balfour Beatty, C&W Communications, Carillion, Catlin, Evraz, Fresnillo, Go Ahead, Icap, Kier, Laird, Man, National Grid, Phoenix Group, Resolution, RSA, SSE.

Low risk (the least volatile stocks in the last five years, subject to no more than three from one FTSE sector): ABF, AstraZeneca, British American Tobacco, Bunzl, Capital & Counties, Countrywide, Diageo, Dignity, GlaxoSmithKline, Greggs, Jardine Lloyd Thompson, Londonmetric, Morrison (Wm), National Grid, Reckitt Benckiser, Reed Elsevier, SSE, Tesco, Unilever, United Utilities.

Megacaps (the 20 biggest stocks): AstraZeneca, Barclays, British American Tobacco, BG, BHP Billiton, BP, Diageo, GlaxoSmithKline, Glencore, HSBC, Lloyds Banking, National Grid, Prudential, Reckitt Benckiser, Rio Tinto, Royal Dutch Shell, SABMiller, Standard Chartered, Unilever, Vodafone.

Small caps (the 20 smallest in the FTSE 350): 888 Holdings, African Barrick, Anite, Big Yellow, Carpetright, Centamin, Devro, Dialight, Fisher (James), Greggs, Hansteen, IP Group, Kcom, Kentz, Kier, Laird, Menzies (J), NMC Health, RPS, Salamander.