Monday 31 October 2016

Hillary, Donald and Investment Management

Based on current betting odds and Nate Silver's fivethirtyeight polls analysis (link here), the probability of Trump winning the US Presidential election is only around 25%. With such favourable odds, the easy conclusion would be: let's go long on US (and global) equities and secure the 1-week 3-5% celebration rally that will tend to follow Hillary Clinton's victory.

But it would be too easy. And bad decision making.

Investing is not really about the probability of success vs. failure. It is not even about expected values and expected returns. It is all about assessing the consequences of failure. Binary outcomes and risk-return asymmetry.

In this case, we have a binary outcome: 0 - Hillary Clinton wins; 1 - Donald Trump wins. If Hillary wins the US equity market's upside is arguably 3-5% over a 1-week period; if Donald wins the downside is possibly 10-15% over the same period. Do you really want to be exposed to a binary outcome were you can win 3-5% if things go your way at the risk of losing 10-15% if they go against you? This is an highly unfavourable asymmetrical risk-return profile. The negative consequences of failure way outweighing the positive consequences of success. Thus, the sensible answer can only be no.

The more academic types among us could always argue that if the potential upside is 5% and the downside is 10% (after all you can exit before the losses reach 15%), with the currently implied 75% probability of Hillary winning the election, the expected 1-week return of building a long exposure to US equities the day before the results are announced is 1.25%. All nice and dandy. However, we should never forget that a 75% ex-ante probability of success can easily become a 100% ex-post probability of failure; a 1.25% ex-ante expected return become a 10% ex-post loss. And that in the age of "the rage against the establishment", the polls' - and implied probabilities - accuracy is not what it used to be.

Wouldn't actually a short position in US equities have a much more attractive risk-return profile ahead of the election than a long one? Yes, it would. Then again, if Trump wins there is enough time to make money with equities. 

Don't be too greedy. Think binary. Think negative asymmetrical risk-return profile. And stay away from the US (and global) equity markets ahead of the US election day.