Good in parts

There are two reasons for the additional scepticism. First - as I promised last week - I combined momentum with the Bearbull Income Portfolio and found that the effect was, as Sam Goldwyn might have said, definitely inconclusive. Second, I stumbled across a paper from 2010 by Andrew Haldane, nowadays the Bank of England's chief economist. Mr Haldane's paper - 'Patience and Finance' (http://www.bankofengland.co.uk/archive/Documents/historicpubs/speeches/2010/speech445.pdf) – tore into momentum as another symptom of our obsession with 'short termism' and, as such, inherently 'bad'. Inadvertently, it defended momentum as much as it attacked it, but at least that makes for an interesting discussion.

The paper starts with the assumption that momentum, short-termism and impatience, which are all part of the same sub-set, are bad because they lead to bad outcomes. They stand in contrast to patience, which leads to good outcomes. Patience places restraint over instant gratification; prefers the future to the present and favours saving over spending. As a result, it provides the finance with which capital is built and long-term growth assured.

Impatience has the opposite effects. It seeks the instant fix and prefers spending over saving. Then - when the savings are spent - it chooses borrowing in order to sustain spending instead of renewed saving. As such, it leads to a gradual deterioration of the capital base, which reduces the scope for long-term growth. Or, as Mr Haldane says, "Some jam today would come at the expense of a whole jam jar tomorrow."

Specifically within financial markets, the struggle between patience and impatience is reflected in the approach of 'investors' and 'speculators'. Investors think long term and seek 'fundamental value'; speculators think short term and look for 'momentum'. Or, in a scenario that Mr Haldane outlines, "momentum-based speculators cause deviations from fundamentals, while long-term investors drive prices back towards fundamentals".

This picture, which carries a moral undertone, is conventional enough and suffers from a familiar shortcoming - no one can put a figure on the fundamental value of any asset. Sure 'fundamental value' is easy enough to define - the sum of all future cash flows from the asset discounted to today's value - but it's impossible to quantify. That's because estimating future cash flows is guesswork. And the discount rate used to reduce future cash flows to today's values suffers a similar deficiency - each of us knows the discount rate that we would use, but we can't know the market's average discount rate.

This is more than semantics. If it is impossible to put a figure on fundamental value, then it is equally impossible to know whether long-term investing is driving prices to or from this value. So it's also impossible to know who is doing the better job - the value-seeking investor or the momentum-following speculator.

Besides which, let's be clear that momentum-following speculators commit capital just as surely as investors. The differences are their perspectives and their investment horizons. Effectively, a momentum follower is saying; "I'm putting capital into this asset because its price is doing well and I'm willing to bet that it will do better still." Meanwhile, the fundamentals-biased investor says: "I'm investing because the price has done badly and I believe that it will correct back to some ill-defined mean, though I have little idea when." But the important point is that both are willing to commit capital. It's not as if the momentum follower has chosen to spend rather than save.

The difference in their investment horizons is that the momentum follower only has a short-term commitment to each situation. He is more willing than the value seeker to take his profits or losses and move on. The investor is making a long-term commitment; he thinks he knows why, but he doesn't know when. Yet maybe these contrasting approaches simply illustrate an observation that Keynes (again) made to an investment committee of his Cambridge college: "A speculator is one who runs risks of which he is aware and an investor is one who runs risks of which he is unaware."

Morally efficient

A proof, of sorts, of Keynes' contention is that momentum followers seem to have produced better results than value seekers, and Mr Haldane's paper provides some evidence. The paper uses extensive data from the website of Nobel laureate economist Robert Shiller - www.irrationalexuberance.com. In particular, the Shiller website shows the relative returns of a simple momentum plan versus an equally simple value plan. Over the very long haul - 1880 to 2009 - the momentum plan would have turned $1 invested in US equities into $50,000-plus. Over the same period, the value plan would have shrunk $1 to 11¢. Mr Haldane does a similar exercise for UK equities and finds that £1 invested in 1920 would have become £1.56 using the momentum plan but just 32p with the value alternative. As Mr Haldane observes: "Momentum would have mopped the floor with fundamentals."

Given that momentum following has performed so much better than value seeking for such a long period, it is hard to condemn it as somehow morally deficient, or at least as operating within a spectrum where hyper activity spills over into moral deficiency.

On the contrary, the rewards of momentum investing have been sufficient to suggest it has maximised returns. If that is so, then it must also have kept the cost of capital as low as possible. That should be seen as a benefit to society since the lowest possible cost of capital would permit the greatest amount of capital investment. True, that conclusion must be tentative because we can't know what would have been the cost of capital had value seeking clearly been the winning strategy. Even so, the general point seems to be that momentum following has wider merits.

Yet not necessarily so wide that it should be a part of every investment portfolio. When we dig deeper, the arguments in favour of momentum-following seem less convincing. Partly, that may be due to the absence of an accepted definition of what constitutes 'momentum' investing.

The 'irrationalexuberance' web site defines its momentum plan as buying $1-worth of the S&P 500 index when the index's returns the previous month were positive and short selling the index if the previous month was a loser. Thus it stands to make money whether the market rises or falls just so long as the market's trends are reasonably smooth.

As noted a couple of paragraphs ago, the tactic produced brilliant gains over the long haul. Yet it would be right to be sceptical about those findings. At least, apply the Shiller methodology to the monthly returns of the FTSE All-Share index starting in May 2001 and the result is lousy (see table). What is labelled the 'Shiller momentum' plan generated returns that fell far short of the other four plans. True, recent investment history may prove an unreliable guide, but the performance of Shiller momentum was killed by the market's volatility; in other words, by the lack of momentum. For example, in 2014, when the All-Share index recorded alternating up and down months with regularity, Shiller momentum suffered badly. Its plan lost 18 per cent that year while the All-Share fell just 2 per cent.

Meanwhile, index provider MSCI defines momentum quite differently. Its momentum indices are only interested in the upside and its components are those stocks comprising about 30 per cent of an index that have been producing the best volatility-adjusted performance over the past six and 12 months. The effect is not necessarily smoother returns than those produced by Shiller momentum. Using the dollar-denominated MSCI World Momentum index, the standard deviation of monthly returns produced by the two methods is almost the same (see table). But the MSCI's returns are certainly better – an average monthly gain of almost 0.6 per cent for the 164 months May 2001 to December 2014 compared with less than 0.2 per cent for Shiller momentum.

For that particular period, MSCI World Momentum's performance is even a touch better than that of the Bearbull Income Portfolio. True, we are not comparing like with like. The MSCI index re-invests the dividends it receives, whereas the Bearbull fund distributes all its income; factoring in distributed income would add about 0.35 per cent a month to its performance. It's also true that, if we took the start date for the comparison back to 2000, the Bearbull fund's performance would have skinned MSCI World Momentum. Most likely, that only goes to show what should be obvious – that the upside-only momentum index suffers in a bear market when even its best performers are doing badly. In contrast, the high-yield value-bias of an income fund should be resilient.

Even so, the performance of the World Momentum index is sufficiently good to give serious consideration to the question with which last week's column finished: should a momentum component be added to most diversified equity portfolios?

The answer has to depend on the portfolio's investment horizon and its likely performance. The longer the horizon, the less the need to smooth returns from month to month, which might be what the momentum component achieves. Similarly, the better the likely performance of the diversified fund, the less the need for a momentum component whose effect would be to slow overall performance.

So, an aggressive investor going for growth should have little need for a momentum component. True, it might help during the worst of times, but adding a minimum-volatility component might be a better course (see the table for details of the performance of MSCI World Minimum Volatility index). Meanwhile, a momentum input may be more useful for a portfolio with a value bias; at least, it should pep up its performance when the market is roaring and, implicitly, 'value' is out of favour.

As to the thought with which we began, that investors who use momentum in a portfolio are borderline morally compromised because they obsess over short-term fixes at the expense of long-run benefits, then perhaps the last word should also go to Keynes, who reminded us (admittedly in a different context) that "this 'long run' is a misleading guide to current affairs. In the long run we are all dead."

How do you want your momentum? Percentage change in monthly returns, May 2001 to December 2014