Tuesday, March 4th, 2014

Crimea-Free Zone

Last week US equities reached an all-time high, and our model decided to reduce its exposure to the asset class by a modest amount. If our process were based on maximum limits or mean reversion, there would be nothing unusual about this. But it isn’t. Our probability-based approach is much closer to a momentum style than mean reversion or long-run optimisation. We assume that the near future will look pretty much like the recent past. We spend our time looking for continuing trends, not infection points.

The big difference between probability and momentum is that we are more sensitive to changes in risk conditions and the performance of alternative asset classes. We will only invest in equities if it is risk-efficient to do so, and we measure risk efficiency in relation to the return from US treasuries, not cash. The simplest way to explain this is to say that equities need to beat a hurdle rate in order for us to stay invested. If the hurdle rate becomes higher than the return generated by equities, we will reduce our equity exposure, even if the index has just hit an all-time high. Most momentum models would not do this.

So what is the hurdle rate and what’s happening to it? We believe that investors should be fully compensated for the extra risk of investing in equities. In arithmetic terms, this means the return from equities should be greater than the Treasury return plus the excess volatility of equities (equity vol minus bond vol). We stress that we use bond returns, not bond yields, and that we look at these numbers over multiple periods, not a simple rolling 26-week period or similar.

For most of last year, the return from US Treasuries ran at a negative rate, but since the FOMC meeting in December this has changed, and it is now running an average annualised rate of 3.2%. This may not sound like much, but it is a big swing from the -5% to -10% numbers we saw in Q2 and Q3 of last year. From April 2013 to January 2014 excess volatility was stable around 4%, but following the sell-off in January this has risen to 5.2%, and could move higher still. Over the last 20 years, the median figure for excess volatility is 7.5%. All of which means that the hurdle rate is now 8.4%.

It is quite possible that US equities will produce a higher return over the next 12 months – or a higher annualised return until the next review date – whichever you prefer. But if investors were asked to forecast a range of returns, we suspect that most would put the bottom of this range below our hurdle rate. In other words, there is a chance (a probability) that Treasuries will do better than equities on a risk-adjusted basis. Back in Q3, when the hurdle rate was negative, very few investors, if any, would have made that call. If we have another sell-off in equities, the one thing that we can be sure of is that the hurdle rate for risk-efficient investment will rise.

Leave a Reply

Full Blog Archive:


Other Recent Posts