When momentum fails

In the first quarter, our momentum portfolio (which comprised the 20 best-performing stocks in 2015) fell by 3.3 per cent. And our low-risk portfolio (the 20 lowest beta stocks) fell by 1.6 per cent. Two strategies that have beaten the market for years, therefore, have stumbled.

Although the losses are small, they could be worrying. They are consistent with the possibility that investors last year spotted that momentum and defensive stocks tended to be underpriced and so bought them - and such buying might have bid their prices up to a high level from which subsequent returns might be modest. As the FT’s John Authers has said, "once you know that cheap stocks outperform, the logical response is to buy cheap stocks. If many do this, cheap stocks' price will rise until they no longer outperform".

Although this phenomenon undoubtedly occurs in many cases, I’m not sure it applies to momentum and defensives. We’ve known about their tendency to outperform for years. The momentum effect was first documented – to my knowledge – back in 2001. And the defensive anomaly was pointed out way back in 1972. I'm no admirer of fund managers, but even I don't think they are so stupid as to have failed to spot these anomalies for decades.

Instead, I suspect there’s a different reason why fund managers have traditionally avoided momentum and defensives. It’s because, from their point of view, such stocks are risky. Fund managers are judged by relative performance: if one makes 10 per cent while his peers make 20 per cent, he is considered a failure. And defensive stocks carry this risk of underperforming: if the market rises sharply, we’d expect them to rise less. This makes the fund manager look bad, so he avoids them. Victoria Dobrynskaya at Moscow’s Higher School of Economics shows that a similar thing is true for momentum stocks.

Perhaps, therefore, momentum and defensives have traditionally done well not because they have been irrationally overlooked, but because fund managers have rationally under-weighted them because they carry an extra risk.

If this is the case, it’s fantastic news for retail investors. We don’t – or shouldn’t – need to worry about keeping up with the herd. We can therefore invest in the stocks that fund managers are avoiding. In effect, we can pick up a risk premium (on average) for taking a risk that needn’t bother us.

This poses the question: why then did momentum and defensives have a bad first quarter?

It might be just bad luck. Even the best strategies have spells of underperformance: returns are noisy. We can quantify this by looking at tracking error – how much returns deviate from the market’s returns. Over the last 10 years, our defensive portfolio has had a monthly tracking error of 2.7 percentage points, with average outperformance of 0.2 percentage points per month. This implies that the 0.2 percentage points of underperformance we saw in Q1 is a 0.17 standard deviation event – the sort of thing that should happen around 43 per cent of the time. Momentum’s underperformance has been bigger and therefore rarer. But even so, it’s been the sort of thing we’d expect to see one quarter in every five. That’s not so unusual.

There is, though, a more proximate reason why momentum and defensives underperformed in Q1. It’s because some battered mining stocks bounced back: Anglo American and Glencore, for example, rose more than 70 per cent in the quarter. This was part of a rotation out of defensives and positive momentum and into value and negative momentum.

In the case of negative momentum, even the 19.1 per cent rise in Q1 doesn’t overturn the strategy’s long-term underperformance. It has lost almost 45 per cent in the last three years, which is the counterpart of positive momentum’s good long-run performance.

Value stocks, however, are more interesting, Their rise in Q1 means they have now risen 33.6 per cent (excluding dividends) in the last five years.

For me, this confirms the fact that value stocks are cyclical. Their rise in Q1 was due in part to fears of a global recession receding, just as their horrible underperformance in 2007-09 was due to those fears intensifying.

I suspect this fact also explains the poorish returns of mega caps. Smaller stocks tend to be more cyclical than big ones and so do well when economic sentiment improves – but the counterpart to this is that mega caps then underperform.

Many readers, I know, like value stocks. Be aware, though, that their returns are a reward for taking on a risk.

Another fact corroborates this. It’s that if we look at relative returns, value and defensives are slightly negatively correlated: the correlation coefficient for monthly returns has been minus 0.14 in the last 10 years. Bad months for value are slightly more often than not good months for defensives. This is despite the fact that some defensive stocks, such as utilities and tobacco, have been high yielders. This is consistent with defensives being counter-cyclical.

Sadly, this doesn’t mean defensives are a great hedge against value. Absolute returns are positively correlated, simply because both strategies tend to do badly when the market generally falls.

Although cyclical risk has paid off well on average in recent years, another risk has not – market risk. Our high-beta portfolio has done horribly over the last five years. Buying stocks that are sensitive to rises and falls in the general market has been a bad idea.

There’s a reason for this, pointed out by economists at AQR Capital Management. Many investors, they say, cannot or will not borrow. When they are bullish, therefore, they cannot express that optimism in the way that economic theory predicts – by taking a geared position in shares generally. Instead, they do so by buying shares that they expect to beat the market – that is, high-beta ones. This, though, means that such stocks are overpriced and so will offer poor returns.

Herein lies one point of this exercise. What I’ve tried to do with these no-thought portfolios is to provide a sort of map of equity returns. This map tells us that the standard capital asset pricing model has been wrong in recent years. For one thing, market risk has not paid off. In fact, the opposite has happened: defensives have outperformed high-beta stocks. And for another, market risk is not the only risk factor that generates good or bad returns. Cyclical risk also matters – hence the gyrations of value stocks. And so too does benchmark risk: hence the good performance of momentum and defensives.

There’s another thing. Because my portfolios exclude judgment, they should represent the lowest returns to these strategies. If they exercise stock-picking skills, value investors should outperform my value portfolio, defensive investors my low-risk portfolio, and so on. Whether this actually happens is, however, another question.

Our no-thought portfolios for the second quarter

Megacaps: AstraZeneca, British American Tobacco, BHP Billiton, BP, BT, Carnival, Diageo, GlaxoSmithKline, HSBC, Imperial Brands, Lloyds Banking, National Grid, Prudential, Reckitt Benckiser, Relx, Rio Tinto, Royal Dutch Shell, SABMiller, Unilever, Vodafone.

High beta: 3i, Aberdeen Asset Management, Barclays, Bodycote, Centamin, Glencore, Henderson, Hutchison China Med, Inchcape, Intermediate Capital, International Personal Finance, Jupiter, Michael Page, Mondi, Ocado, Regus, Schroders, Thomas Cook, Travis Perkins, Vedanta.

Low risk: Acacia Mining, Breedon Aggregates, Cineworld, Consort Medical, Cranswick, Dechra, Dignity, Emis, Entertainment One, Euromoney, GVC, Imperial Innovations, James Fisher, Jardine Lloyd Thompson, Kcom, National Grid, Novae, Tate & Lyle, Telecom Plus, Wm Morrison.

Value: Aberdeen Asset Management, Amec Foster Wheeler, Anglo American, Ashmore, BBA Aviation, Berkeley, BHP Billiton, BP, Carillion, easyJet, Glencore, HSBC, Legal & General, Pearson, Redefine, Rio Tinto, Soco, SSE, Standard Chartered, TalkTalk.

Momentum: 888, Auto Trader, Boohoo, Breedon Aggregates, Centamin, Cranswick, Dairy Crest, Dart, Darty, DCC, FDM, Fevertree, Hutchison China Med, JD Sports, NMC Health, Regus, Rightmove, RPC, SuperGroup, WH Smith.

Negative momentum: Aberdeen Asset Management, Amec Foster Wheeler, Anglo American, Antofagasta, Barclays, BHP Billiton, Entertainment One, Evraz, G4S,

Glencore, International Personal Finance, Melrose, Ophir Energy, Pearson, Petra Diamonds, Restaurant Group, Sports Direct, Standard Chartered, Thomas Cook, Weir.

All portfolios are equal-weighted baskets of 20 stocks with a market capitalisation of over £500m, taken from the IC’s stock screen.