BLOG - Markets are efficient but sometimes irrational as well
Robert Davies, Fund Manager VT Maven Smart Dividend UK Fund
Nothing quite captures the challenges of investing so well as the Nobel prizes awarded for Economics in October. One prize was given for saying markets are efficient, one for saying they were irrational and one for saying it was hard to measure what was actually going on.
But the biggest oddity is this. Few working physicists would disagree with the theory behind, say, The Higgs particle, work on which led to the award of a Nobel prize to Professor Higgs. Yet the majority of professional fund managers, consultants and intermediaries involved in asset management fundamentally disagree with the theories that lie at the heart of the work by these three Nobel Prize winners for economics.
If they did agree they would not bother to try and guess the short-term evolution of security prices. Professor Fama demonstrated that changes in asset prices over recent weeks and months are no help in determining future prices.
It might be thought then that detailed financial modelling of future cash flows would provide a better way to discover true asset prices. Then Professor Shiller proved that feedback mechanisms and human emotions move security prices around much more than might be expected from the stable flow of dividends.
Finally, Professor Hansen demonstrated that the Capital Asset Pricing Model failed to reflect asset prices after adjusting for variations in the way risk was priced.
Despite these studies most advisers to pension funds and individual investors still advocate using funds where the manger attempts to second guess a share price at some point in the future and the general level of the market at the same time. This might seem a rather futile exercise but as Upton Sinclair put it so succinctly in the thirties it is very difficult to believe one thing if your pay cheque depends on believing another.
Shiller demonstrated that share prices bounce around a lot more than dividend streams. This means that dividends give a very useful and steady reference point on which to base a portfolio. The unique power of this approach enables a portfolio manager to harvest volatility. This is because portfolio weights are not determined by share prices but by dividend contribution. Consequently, when share prices move sharply a suitably constructed fund is able to exploit that and pick up shares that are depressed. Of course no one knows if that will turn out to be correct but consistently picking up oversold shares increases the potential to add value.
A recent example will help to explain this:
During the Labour Party conference in September 2013 Ed Milliband, the leader, announced that, if elected, he would freeze energy prices. Shares in the utility companies promptly sold off by about 8%. This shrank the holdings of these stocks from about 1.1% to just over 1% in the fund and left Centrica and SSE underweight in the portfolio relative to the model. Since no one knows if Labour will win the next election, or if it will actually execute this policy, or what the companies’ responses will be, then the future remains just as unknown as before.
All a process can do is harvest the volatility by adding to holdings at a price 8% lower than it was the previous day. So far the dividend forecasts for these two companies are essentially unchanged thus demonstrating Shiller’s thesis that share prices are more volatile than dividends and that behavioural economics has a large impact on share prices.
No one is claiming this process is perfect but at least fundamental tracking, or smart beta, can claim to be working with the grain of leading academic research rather than against it.