Wednesday, May 29th, 2013

Narrative neutral

It is widely accepted by financial journalists that last Thursday’s 7% decline in Japanese equities was caused by Mr Bernanke’s testimony to the Joint Economic Committee of Congress, even though there is disagreement about exactly what he meant, and whether his comments in the Q&A conflicted with the written testimony. Japanese equities fell after his comments, so that must have been the cause.

We have no wish to dispute the sequence of events, but we are uncomfortable with the idea that everything which happens in financial markets must have an identifiable cause. Nassim Taleb has written extensively about the narrative fallacy, the tendency to construct stories around a series of facts in order to give them coherence. This is not an argument that says nothing ever has a cause, rather that causation is often more complex than the simple narratives which get reproduced in the financial press.

Causation is embedded in the efficient markets hypothesis (EMH), the idea that all publicly available information is discounted in the market price and that only new information can cause the price to change. New information is released; the price changes; therefore the new information caused the price to change. This is similar to the idea that every asset has a fair or equilibrium value, to which it will correct in the absence of any stimulus to the contrary.

We are sceptical about EMH and the concept of fair value. The valuation tools are so imprecise and the inputs so difficult to forecast that the range of answers is often too wide to have any predictive power. The results are practically useless except at extremes, which themselves are easier to recognise with hindsight rather than in real time. The good news is that if you don’t believe in fair value, you don’t need to spend so much time thinking about fundamental causes.

We describe our process is narrative neutral. We don’t need stories to make sense of the relationship between risk and return, because we observe it directly, and we know how to convert those observations into an investment portfolio. We don’t deny that QE and its potential withdrawal may have an impact on asset prices, but forecasting quantum, timing, and impact is impossible. So why bother? We think effects are more important than causes. That’s why we spend so much time analyzing them. Effects, not causes, are also the benchmark against which investment managers are measured.

Leave a Reply

Full Blog Archive:

Other Recent Posts