Advisers deal with people who have capital. For the vast majority the prime concern is not losing their capital. As people get to grips with the consequences of increased longevity and lower-for-longer interest rates, this is becoming more important. In decumulation, a serious market downturn can wipe you out in a way that it simply will not during the accumulation phase.
So the right starting point is containment of wipeout risk, and the conventional view is this is best achieved by asset allocation. That risk has to be assessed not in terms of volatility but of drawdown – the maximum historical peak-to-trough loss. For example, it is quite unrealistic to take the low volatility of physical property funds as a proxy for risk. As we saw in 2008, a 30 per cent decline in values can come out of a clear blue sky. Knowing that is possible, it would be irresponsible to allocate a large proportion of capital to physical property on the basis of its low volatility.
A typical balanced portfolio will show drawdowns of 30 per cent or more in 2008/09. Many clients will not sign up to that. But if you do not tell them about the historic drawdown and they do end up losing that much, you could lose a complaint case at the Financial Ombudsman Service and in my view you would deserve to lose it.
It is natural managers and advisers want to promise a greater degree of capital safety. The obvious answer is to promise active asset allocation that will minimise drawdowns. So we have a new fashion among investment managers, which is to “manage to volatility”.
You set volatility parameters for a portfolio and then switch capital between asset classes to prevent the volatility of the portfolio rising or falling too far. Importantly, this means you can ignore historical drawdowns in telling clients what to expect from their portfolios because you assume active asset allocation will enable you to step out of the way of approaching trains.
There are two problems with this. One is that “manage to volatility” is simply a variety of market timing and at best it will take decades more evidence to establish that it can work. Most academic work suggests it will not. A second is that you cannot compare portfolios managed using conventional asset allocation methods with those managed to volatility. You can trust the historic drawdowns, but not the volatility, of the first and you can trust the historic volatility, but not the drawdowns, of the second (remember those property crashes – there is no direct link between volatility and drawdown). Put both types of portfolio together in one table and you are comparing apples with bananas.
How do you assess a manager that suddenly tacks to aggressive cash allocation in all its portfolios? Perhaps you ought to ignore all their historic statistics, since they give no meaningful indication of likely future returns or drawdowns.
There is a bigger danger. The 1987 crash was partly caused by “portfolio insurance” triggering stop-loss selling, and manage to volatility incurs the same risk. If more managers target volatility, and there is serious money to be made out of spooking them, you can be sure a fraternity of evil hedgies will be ready and willing to try to “flash-crash” them out.
Chris Gilchrist is director of Fiveways Financial Planning