Defensives and momentum still win

One of these was defensive investing. My portfolio of these – an equal-weighted basket of the 20 stocks with the lowest beta over the past five years – rose by 8 per cent in the year.

Of course, we’d expect defensive portfolios to out-perform a falling market. But there are two oddities here. One is that my portfolio rose as the market fell, whereas conventional theory tells us that in bad times defensive stocks should only fall less than the market. The other is that defensive stocks have hugely beaten the market over the long term. In the past 10 years, my portfolio has risen almost 90 per cent, outperforming the FTSE 350 by more than 70 percentage points. This is not an artefact of how I’ve constructed the portfolio, and nor is it confined to the UK; over the long run, defensive stocks have outperformed around the world – a finding that is robust to different definitions of defensiveness.

While the outperformance of defensive stocks is well established, what’s not so certain is why they do so well. One possibility is that they carry benchmark risk. If the market rises strongly, defensives would lag behind and so fund managers who hold them would underperform. Because they are judged on relative performance, they are therefore loath to hold defensives, which causes them to be underpriced.

This risk, however, did not materialise much this year. When the market jumped in late 2020, defensives also did well, lagging the FTSE 350 by only 2 percentage points during the fourth quarter. What’s more, defensives have delivered huge outperformance – of 4.9 percentage points a year in the past 10 years. This seems too big a risk premium for benchmark risk alone.

Which suggests another possible explanation, proposed by economists at AQR Capital Management. They point out that many investors cannot borrow as much as they want. When they are bullish, therefore, they express this optimism not by borrowing to buy shares generally as textbook theory predicts but rather by overweighting high-beta stocks and underweighting low-beta ones. This causes the latter to be underpriced and so to outperform over the long run. Which is what we’ve seen.

This theory also predicts that high-beta stocks should underperform. Which they have over the past 10 years. This underperformance was, however, broken at the end of last year. My portfolio of high-beta stocks jumped by almost 40 per cent in the fourth quarter, with shares such as Cineworld (CINE), Hammerson (HMSO), EasyJet (EZJ), Micro Focus (MCRO) and Virgin Money (VMUK) all gaining more than 50 per cent. So great was this bounce that high-beta stocks actually rose during the year as a whole, bucking the long-term trend.

It’s not just defensives that had a good 2020, however. Momentum put in a spectacular performance, rising almost 40 per cent in the year. Much of this came in the fourth quarter thanks to huge gains on AO World (AO.) and Ceres Power (CWR) among others.

As with defensives, this continues a long-run trend. My momentum portfolio has trebled in the past 10 years while the FTSE 350 has gained less than 20 per cent. And its 100 per cent gain in the past five years is beaten by only one fund in Trustnet’s database of all companies unit trusts: Chelverton’s UK Equity Growth.

As with defensives, this great performance is not confined to the UK and it is robust to different definitions of momentum. Narasimhan Jegadeesh and Sheridan Titman, the economists who first noted the momentum effect, showed that stocks that have done well in the past three, six, nine or 12 months all continue to do well over various subsequent periods.

Exactly why this should be is unclear. The sheer scale of the outperformance means it is difficult to attribute it to compensation for taking on risk. My portfolio has delivered a risk premium of almost 10 per cent per year in the past 10 years. I find it hard to imagine risks great enough to justify that outperformance. This makes me suspect that at least some of momentum’s outperformance is due to a behavioural bias. Investors cleave too strongly to their prior beliefs about stocks and fail to update these ideas sufficiently when good news arises. This causes momentum stocks to underreact to such news and so be under-priced, with the result that they rise in the following weeks.

The flipside of past winners doing well is that past losers should do badly. For much of the past few years this has indeed been the case. But this changed in late 2020. In Q4 my negative momentum portfolio jumped by 55 per cent, having lost a third in the previous nine months. This was helped by 70 per cent-plus gains in Aston Martin Lagonda (AML), SSP (SSPG), Mitchells and Butler (MAB), Meggitt (MGGT) and Micro Focus.

We have Pfizer and BioNTech to thank for this. Much of the rise was a reaction to their development of a vaccine against Covid-19. In fact, the portfolio surged by almost 44 per cent in the first three weeks of November alone. Which contains an important lesson. If you are buying bombed-out stocks, timing matters. A lot of the gains on them – if they occur at all – can come very quickly. If you miss these jumps, you’ll be holding dead money or worse. And the long-run performance of my negative momentum portfolio suggests that investors are not well rewarded on average for holding them.

Something else happened in 2020 that was quite normal – the performance of value stocks. My basket of the 20 highest yielders had a good fourth quarter, gaining more than 20 per cent, thanks in part to big rises in Easyjet, Tui (TUI) and Workspace (WKP). But this followed a tough period: at its low point in late March, my value portfolio had lost 40 per cent since the start of the year.

Such swings, however, conform to the historic pattern. Deep value stocks are cyclical. Very high yields are often compensation for taking on recession risk. When this risk materialises, as it did early in 2020, value stocks get clobbered: Crest Nicholson (CRST), Hammerson, Tullow Oil (TLW) and Micro Focus were all on good yields at the start of the year, but lost more than 60 per cent by March. By the same token, though, value stocks bounce back well as recession risk recedes. Hence their good performance in November.

All these portfolios have one big thing in common. I form them purely by using the IC’s equities screener, and use no judgement at all about the merits or not of individual stocks. The fact that defensives and momentum have done so well over the long run therefore tells us something important. What matters in stockpicking is not so much which specific stocks you buy as which types of share you buy. If you’d bought defensive or momentum shares without any thought, you’d have done well on average in recent years. In this sense, stockpicking is overrated. What matters is strategy.

There is, however, a caveat to the good relative performance this year of these portfolios. It’s that this is flattered by the fact that the market has been dragged down by big falls on a few huge stocks such as HSBC (HSBA), BP (BP.) and Royal Dutch Shell (RDSA). If you had picked stocks at random and had avoided these, the chances are that you would have beaten the market. In this sense, 2020 was an easy year for stock-pickers.

.