Momentum “happens”. It is a well documented fact although argument persists about its causes and whether it is possible to profit from it. Scepticism also surrounds its reputation for being a “market timing” mechanism and quite rightly. Trend following scepticism is also, again quite rightly, rooted in the typical application of trend following by CTAs with their extravagantly geared product which inevitably entails huge volatility and colossal drawdown.
The track record of CTAs is littered with disasters, most of which get covered up when yet another program bites the dust and disappears from easily accessible records. Hence you have idiots proclaiming that such and such trend follower has a 40 year track record and is a billionaire. Yes, fine but they fail to mention the trail of disasters he has left behind. Nor the fact that many investor will have lost money in even the successful programs.
Most are fed up with the black box trend followers and their hidden models. And probably in most cases their secrets are not worth keeping hidden anyway. Its a very obvious trade.
There is a version of trend following which may be of greater use and longevity however.
You may be able to minimise the risks and get at least some of the benefit of trend following by picking a wide variety of assets within the two classes which, historically at least, have shown a positive upwards drift. And if you are worried about increasing correlation between bonds and equities then stick to the short end of the yield curve. Long term figures I have looked at seem to give lie to the theory of an upward sloping yield curve over the very long term….but I must re-visit the question and triple check my figures.
Momentum is “good”. It’s “real” and it can increase your returns. Provided you don’t concentrate in too small a group of assets which is bound to lead to higher volatility, drawdown and instability (while often giving correspondingly higher returns over time).
Forget about on/off market timing but achieve something similar whereby you never have to take on off bets but simply let your allocations drift between bonds and equities as relative performance dictates.
Here is a backtest with 30 assets, half bond funds, half stock funds. The portfolio is rebalanced on a monthly basis and rotates into the top 20 performing funds over the lookback period. It is never wholly in bonds. It is never wholly in stocks. It does not rely on “market timing” in most accepted senses – just gentle wavering from one asset class to the other.
Importantly, weighting is inverse volatility rather than equal.
Stat Momentum Inverse Volatility S&P 500 ------------------- ----------------------------- ----------- Start 1997-06-20 1997-06-20 End 2016-12-20 2016-12-20 Risk-free rate 0.00% 0.00% Total Return 279.52% 263.72% Daily Sharpe 1.25 0.43 CAGR 7.08% 6.85% Max Drawdown -13.39% -55.24% Monthly Vol (ann.) 5.60% 15.28%