Have investors finally wised up to longstanding stock market anomalies, and so caused the market to become harder to beat? This is the question posed by the recent performance of our no-thought portfolios.
In the last three months all but one of these have under-performed the FTSE 350. This is odd, because you’d expect at least one to do well; for example, if our low-risk portfolio under-performs, our high beta one should out-perform, and vice versa.
The fact that this hasn’t happened raises the possibility that investors have finally learned that some simple strategies have beaten the market in the past – and in so learning, they might have bid away these sources of profit. For example, if investors had realized a few months ago that momentum investing is a good idea, they would have bought momentum shares, thus raising their prices to levels from which subsequent returns have been poor. The same fate might have befallen value and defensive investing.
| Performance of no-thought portfolios | |||||
| in Q3 | last 12M | last 3Y | last 5Y | Last 7Y | |
| Momentum | -3.2 | 1.6 | 33.0 | 45.0 | 62.3 |
| Negative momentum | -8.9 | -14.3 | -8.1 | n/a | n/a |
| Value | -5.4 | 3.3 | 44.5 | 15.2 | -51.4 |
| High beta | -9.4 | -15.1 | 42.0 | 1.3 | -17.3 |
| Low risk | -1.7 | -0.5 | 25.2 | 45.1 | 26.6 |
| Mega caps | 0.7 | 2.9 | 23.2 | 17.5 | 2.2 |
| FTSE 350 | -1.5 | 2.9 | 33.1 | 34.2 | 6.9 |
| Price performance only: excludes dividends and delaing costs | |||||
If this is the case, then it's going to be much harder to beat the market in future.
Whilst this is possible, three things make me doubt that it is the explanation for the under-performance of our portfolios.
First, in one respect momentum investing still works. Our negative momentum portfolio (the 20 worst performers in the previous 12 months) fell by 8.9 per cent in Q3, continuing three years of under-performance. This poses the question: if investors learned of the virtues of positive momentum investing and so bid away its profits a few months ago, why did they not also learn to avoid stocks with negative momentum? Is this really simply because there’s a constraint upon investors’ ability to short-sell shares?
Secondly, it’s possible that this under-performance is due to mere bad luck. All stock-picking strategies will sometimes do badly simply because there’s lots of noise in equity returns. We can measure this by the standard deviation of relative returns. Since we began it ten years ago, our positive momentum portfolio has had an annualized standard deviation of relative returns of 13.2 percentage points. If we assume that momentum will on average out-perform the FTSE 350 by 6.7 percentage points per year – which it has over the past 10 years - this implies that there’s roughly a one in six chance of it under-performing by 6.5 percentage points over a 12 month period. And it tells us that there’s around a 30 percent chance of the sort of under-performance we’ve seen in Q3.
There is, though, a third reason to suspect that the underperformance of most of our benchmark portfolios isn’t because investors have wised up. They might instead have done badly because a risk to all of them has materialized – namely, benchmark risk.
In the last three months, big stocks have done well; our megacap portfolio (an unweighted average of the 20 biggest stocks) has risen thanks to good performances by the banks. Because indices are weighted by market capitalization, this drags up the performance of indices relative to the performance of most stocks. The three biggest stocks on the market – HSBC. BP and Royal Dutch – have a combined market capitalization greater than that of the entire FTSE 250 and so their performance matters more for the FTSE 350 index than does the fate of the 250. This means that when megacaps do well, most stocks will under-perform the market. It follows that stock-picking strategies which pay no attention to market capitalization are then likely to under-perform too.
This problem doesn’t, of course, merely afflict our no-thought portfolios. It’s a difficulty for all truly active stock-picking strategies.
This poses the question: is the under-performance of our momentum and defensive portfolios due to investors wising up, or due to bad luck and benchmark risk materializing?
Ordinarily, the answer would depend in part upon our Bayesian priors; the stronger your prior belief in the merits of defensive and momentum investing, the more likely you are to write off their recent under-performance as mere bad luck.
Herein, though, lies a paradox. The evidence that momentum and defensive investing has paid off historically is indeed overwhelming. But the very power of this evidence is a reason to discount it. Surely, if it’s so obvious that momentum and defensive investing have done well, investors should have wised up by now and bid away their excess returns.
For investors’ purposes, therefore, the weight of evidence for momentum and defensive investing is ambiguous as a guide to future stock-picking.
This, though, is where benchmark risk enters. The fact that investors know that a strategy is profitable does not ensure that they will pursue it, because it might be too risky. And benchmark risk makes momentum and defensive investing risky for many professional investors. Defensive portfolios carry the risk of under-performing if the market rises very sharply. Momentum portfolios also carry this risk, as Victoria Dobrynskaya at the LSE points out. This makes the strategies unattractive to fund managers who are judged by relative performance, because under-performing a rising market could get them the sack.
Such managers will therefore steer clear of momentum and defensive stocks. This will cause them to be under-priced, and so offer good long-term returns on average for the investor willing or able to take on such risk. Retail investors should be in this position - either because they don't care about relative performance or, to the extent they do, they can reduce it by simply holding tracker funds.
But of course, the key phrase here is “long-term”. Sometimes, benchmark risk does materialize. It did do in Q3 – albeit by megacaps out-performing rather than by the market soaring ahead. When this happens, defensives and momentum do poorly. But this is entirely consistent with them doing well over the long run.
In this sense, the strongest evidence we have for retail investors to use momentum and defensive investment is not just the fact that these strategies have out-performed around the world over the long-run, but also the very fact that they’ve under-performed recently.
Now, you might object here that I’m trying to have it both ways; out-performance and under-performance are both evidence in favour of momentum and defensive investment!
Not so. The argument for such strategies is falsifiable. If they were to under-perform significantly in circumstances where benchmark risk doesn’t materialise – say because the market moves sideways or if megacaps under-perform the market – then this would be evidence against those strategies.
So far, though, such evidence has not emerged.
