In theory, it's a fool's errand. Markets are efficient, we're told. That means there should be no "secret". It should not be possible for a given strategy to beat the market consistently over the long run. As soon as any secret is uncovered, it should instantly be arbitraged away by a swarm of profit-seeking investors.
This is one of the key rationales behind index investing. You have to get very luck indeed to beat the market, so why not just settle for the security of knowing you are at least tracking it? And to be clear, that makes a lot of sense. Indexing is a low-maintenance, low-cost way to allow your retirement savings to grow over the long run without having to think too hard about it. It works well compared to many of the alternatives.
And yet - there are methods of investing that do manage to beat the market consistently over time. They are called "factors". In effect, these are styles of investing (we'll have more on the specific styles below) that take advantage of apparent anomalies in order to deliver above-average returns over time. This is one of the driving forces behind the "smart beta" industry - indices or ETFs that attempt to add value over and above your standard market cap index.
Now, factors have come in for a bit of stick in recent years. Academics, often funded by institutions hoping to launch new smart beta products, have been falling over themselves to uncover shiny "new" factors that can trounce markets. The problem is, while these often look good in backtests, when the products actually launch, the factor either disappears or doesn't work half as well in "real life" as it did on paper.
Subsequent studies have gone on to show that many of the 300 or more "new" factors are in fact simply the product of "data mining" (or "p-hacking" in the jargon). In other words, the researchers have tweaked the variables to the point where the backtest looks good, but all they've really done is to find a combination of features that might have randomly worked in the past but has virtually no chance of continuing to do so in the future.
However, that doesn't mean you should throw out factor investing. Indeed, Dutch investment group Robeco has recently conducted a major piece of research (Global Factor Premiums, by Guido Baltussen, Laurens Swinkels and Pim Van Vliet), with tight controls for "p-hacking", aimed at discovering just how effective the "classic" factors are. Taking data over an impressive 217 years (many studies only go back to 1980), the group looked at six factors across equity, government bond, currency and commodity markets, and found that in the vast majority of cases, they worked extremely well.
In other words - there are "secrets" to beating the market. And they don't get arbitraged away, presumably because they are the result of behavioural quirks that are difficult to resist or take advantage of.
So what are they? Here's the top six, with short explanations of what they mean.
Trend-following: in short, buy what's going up and sell what's going down. This was consistently the very best-performing factor, researchers found.
Momentum: this is similar to trend-following, except that it is about finding assets that perform well relative to others, rather than in absolute terms.
Seasonality: believe it or not, asset classes perform differently at different times of year in a consistent manner. So the old saying "sell in May and go away" contains some truth to it.
Carry: investing in high-yield assets or currencies delivers better returns than investing in low-yield over time.
Betting against beta: put (over-)simplistically, investing in the least risky assets delivers higher returns over time, than investing in the most risky ones, which is the opposite of what theory suggests should happen (this is sometimes known as the low-volatility factor, although some would argue it's not exactly the same). Put even more simply, defensive stocks beat lottery ticket stocks.
Value: if you buy cheap stocks, they will outperform over time.
Clearly, not all factors work all the time. These are for long-term investors, and sometimes the long-term is really very long indeed. Value, for example, has infamously struggled during the post-financial crash period, and every time we think it's making a comeback, it wilts again. And it's one thing to know that factors work - it's quite another to find a fund that is competent at implementing them. However, I do think the study gives investors two valuable points to chew over.
Firstly, as well as diversifying your portfolio across geographies and asset classes, you might want to consider diversifying across factors too. In particular, I'd suggest getting some exposure to trend-following. You may think that charting and technical analysis (which is the only real way to spot trends) is a lot of nonsense - but as it's one of the most persistently outperforming factors, it seems short-sighted to dismiss the technique.
Secondly, if you are considering smart beta ETFs, don't go for anything newfangled and overly complicated. If it's not based on one of these six proven factors - which can be tricky enough to wrap your head around as it is - then give it a miss and find something more straightforward. Remember, factors might help you to outperform the market, but if you give it all back in high fees, you're not doing yourself any favours.