What should you measure your portfolio against?
If your strategy is trading value shares, which is what I write about here, how do you know if it is successful? What is success anyway?
Making a profit of any sort from shares over a long period might be considered a success but if the gain achieved is little different from that available from an investment like bank deposits, then you haven't been rewarded for the additional risks involved in your strategy of trading value shares.
Any measurement really needs a risk-adjusted approach but I'm not talking complex math here, just looking for a decent premium over the safe route.
Beating the bank
So, if say bank deposit interest over a suitably lengthy period of, for example, ten years was 3% per year on average (for a total return of 34%), then if you have traded value shares and averaged 8% a year including dividends received for a total return of 116%, you have done a lot better than keeping your money in the bank.
I don't propose discussing whether this is a sufficient premium to compensate for the risks, that would in any case be a matter of opinion, but clearly it is a degree of success on this comparison. And because of the way compound interest works, the longer the period for which you can do this, the greater will be the premium attained.
Beating the market
The other benchmark worth considering, and probably the more likely one for value players to use, is the market as measured by one of its leading indices such as the FTSE100 or the All Share. The only practical way to invest in the market is through one of the many commercial tracker funds available but these should follow their index pretty closely over time.
Note that unlike bank deposits, using a tracker as a benchmark cannot be regarded as the relatively risk-free option with which to compare the results of value investing. It's more like the action-free option but it is obviously not risk free. Other than making a decision to invest in a tracker in the first place, you don't have to do anything to maintain it whereas value trading requires repeated investor input on buy and sell decisions.
Consequently, it is logical for the premium a value player might expect over a tracker to be considerably less than that expected over bank deposits over a long period. And similarly, tracker investors would themselves expect some sort of premium over the bank in that time. But beating a tracker with a reasonable margin by using a value strategy would indicate good success in my view.
It is perhaps a little odd that I am writing about benchmarks because I have never been one for much measuring or comparing, even odder when you consider that I am an accountant. Then again I am about as similar to a typical member of my profession as England is to a leading international football team. So I do advocate that value investors measure their returns but only rarely, not too frequently or obsessively. That can lead to short termism and consequent disillusionment if your investments don't return what you think is desirable because the period being examined is too short.
Look at the long term
Value is a fairly long-term strategy even though individual trades may be very short term. There is no time scale for a value play, you buy at value and you sell when the value has evaporated sufficiently in your view. But you cannot know how long that will take. Maybe days if you are extremely lucky, but mostly very much longer, often years.
Thus the profits and losses are highly uneven and can never accrue smoothly. There will be whole years when you lose or make nothing because you haven't made any sales and there will be other years when you make what in isolation are relatively large gains.
It's only after a long time, maybe ten years at least, that you can look back with any sense of perspective to see how you have done. Individual short-term snapshots of a value strategy during that time will yield unrepresentative figures.
The dangers of extrapolation
The same could be said of the tracker benchmark. One year's tracker fund return is not going to tell you anything meaningful about what the ten-year return is likely to be. The question is though whether the one-year market return offers a worthwhile comparison against which to test the same year's return from a value approach.
The answer is no.
Whilst short-term returns from both the market and value should not used for extrapolation in an attempt to forecast their respective potential long-term trends, their failure as a useful basis for doing so will be dissimilar. In other words, not only are the short-term performances of each strategy unrepresentative of their longer term trends, the relative differences in that same short term are also completely unreliable.
What I mean is, assume in a year your benchmark tracker falls 10% in total return. That tells you nothing about what it might do over ten years and most certainly not that you can expect a 10% fall every year. Now consider that, in the same year, your value strategy has made a total return of 15%. Again you are aware that this tells you nothing about what to expect long term.
The outperformance in that year against the market is substantial, you've beaten it by 27.8%, but it does not follow at all that this will be the likely relative performance over many years. The returns from value and the market will fluctuate, but not in the same way at all.
Is beating the market by value trading a desirable objective at all? I'd say yes otherwise you are not being rewarded for the additional risks and effort you are incurring.
Value is more risky in the sense that you will hold far fewer shares than a tracker fund. But you need to give it considerable time, many years, to know really whether you are winning by this measure or not.
More from Stephen Bland:
