Momentum wins again

This portfolio outperformed all but 13 of the 257 unit trusts in Trustnet's database of UK all companies funds. The lesson here is important and overlooked: getting the right strategy is more useful than time-consuming stockpicking.

It's not just positive momentum that did well in 2015, however. Negative momentum did horribly. Our portfolio (comprising the 20 worst-performing stocks in the previous 12 months) lost 37.2 per cent. This was largely due to continued falls in mining stocks. It's no accident that the worst-performing all companies unit trust last year was Schroder's recovery fund: it invests in stocks 'that have suffered a setback' and so is exposed to negative momentum.

Closely related to the fall in negative momentum was the underperformance of value stocks last year: this is because mining stocks have been over-represented among high yielders.

There is, however, usually a big difference between negative momentum and value. Whereas negative momentum seems to do badly even over the long run, value does not: in fact, despite last year's fall, our value portfolio has beaten the market over the past three years.

There's a simple reason for this. Deep value stocks - the highest yielders which enter our value portfolio - tend to be cyclical. They do especially badly when investors worry about recessions. For example, builders and mortgage lenders got clobbered in 2008-09 just as miners got badly hit last year by fears about China's downturn. However, in normal times investors pick up a risk premium for taking on cyclical risk, and so value stocks do well.

There's one point about momentum that I must emphasise. Its behaviour in recent years is not an idiosyncratic artefact of our own portfolio. Quite the opposite. I would regard it as simply one more data-point which corroborates what is an almost ubiquitous fact - that momentum investing generally does well. This is true of US equities; commodities; currencies; international stock markets; and even sports betting. And it is momentum that explains why a simple rule - buy when prices are above their 10-month average and sell when they are below - works so well.

There might be a good reason why momentum works. It's because investors underreact to good or bad news. For example, when there were some early signs of a slowdown in demand for commodities, mining investors shrugged these off and stuck to their belief in the commodity 'super-cycle'. This meant that prices didn't immediately fall sufficiently to embody the bad news, causing mining stocks to be overpriced with the result that they fell in the following months. The same thing happens in reverse to stocks enjoying good news: investors cleave too much to their prior belief that the stocks are mediocre, with the result that the shares are initially underpriced and so drift upwards later.

In part, this underreaction arises from a wholly reasonable motive. Sticking to your prior beliefs is often the right thing to do. Those beliefs are (or should be) shaped by a lot of evidence, whereas a lot of news about a stock is often just temporary noise. However, investors are often overconfident about what they know. When they are, they tend to stick too much to those prior beliefs - which gives us momentum effects.

But why doesn't the smart money exploit this irrational behaviour?

In the case of overpriced negative momentum stocks, it is simply because short selling is risky and difficult: even if you are right that a share will fall eventually, short-term volatility can lose you money.

In the case of underpriced positive momentum, Victoria Dobrynskaya at Moscow's Higher School of Economics proposes an explanation. She points out that past winners have big downside betas but also low upside betas. This means they carry lots of benchmark risk; there's a big danger of them underperforming a rising market. This makes them unattractive to the many fund managers who are judged by their relative performance.

Benchmark risk explains another fact about our no-thought portfolios: the good performance of defensive stocks. Our low-risk portfolio (which comprises the 20 lowest-beta stocks, subject to no more than three from any one sector) rose by 10.8 per cent last year. This means that in the last five years it has risen by over 40 per cent, during which time the FTSE 350 has risen 11.7 per cent. This contradicts textbook economic theory, which says that defensive stocks should underperform a rising market, and fall less than a falling one.

Again, this is no idiosyncrasy of our portfolios. It fits a pattern: defensive stocks have tended to outperform around the world over the long run.

So, why is the theory wrong? One reason is that investors under-rate the virtues of dullness. They look for 'exciting' get-rich-quick growth stocks and in doing so leave get-rich-slow stocks underpriced.

Another reason lies in the counterpart to defensives' good performance - the underperformance of high-beta stocks; this has been a feature not just of 2015 but of the last five years generally.

One reason for this has been pointed out by Andrea Frazzini and Lasse Heje Pedersen at AQR Capital Management. Imagine, they say, that you are bullish of the market generally. What do you do? Theory tells us that you should simply alter your mix of cash and equities generally, perhaps even borrowing to buy more shares. Many funds, however, can't do this: their mandates limit the amount of cash they can hold - why pay a fund manager 1.5 per cent just to hold cash? - and the amount they can borrow. Such funds must therefore express their bullishness in another way - by twisting their portfolios to hold more high-beta stocks and fewer low-beta ones. Because most fund managers are bullish most of the time, this causes high-beta stocks to be relatively overpriced and low-beta ones to be underpriced. The upshot, say Ms Frazzini and Mr Pedersen, is that a strategy of betting against beta works - not just in equities but in other assets, too.

Yet again, our no-thought portfolios fit the general international pattern.

And, again, benchmark risk can explain why this anomaly persists. The fund manager who believes defensive stocks are underpriced must take a risk to buy them: if the stock market soars, he would almost certainly underperform - a fact that might cost him his job. This deters him from exploiting the mispricing.

We retail investors, however, don't need to worry so much about the risk of underperforming: the cost of doing so is an opportunity cost rather than the loss of a job. We can therefore afford to take on the benchmark risk contained in defensives and momentum stocks. We don't have many advantages over professional investors, but this is one.

That said, I would be wary of the negative momentum portfolio. There might well come a time when badly hit mining stocks snap back. This might be because we see signs of stabilisation in China's economy; or because bargain-hunters return; or because short-sellers close their positions. For these reasons, I expect a lot of volatility in our negative momentum portfolio.

However, I'm not sure that the point of all this is merely to recommend particular stocks or strategies. Instead, I think of it as a monitoring exercise. Evidence from around the world tells us that equity markets are not as efficient as economists (including me) once believed. Instead, there are some systematic deviations from efficiency, as manifested by the good performance of positive momentum and defensive stocks and the bad performance of high beta and negative momentum. My portfolios corroborate this research. Whether they will continue to do so in 2016 is something we will be tracking.