Friday, May 29, 2015
Jim Grant Takes On Risk Parity, Ray Dalio, And Cliff Asness
“Dalio and Asness are, of course, formidable Wall Street thinkers and doers. Formidable, too, are the critics, who include Paul Singer maître d’hotel of Elliott Management and Ben Inker, co-head of GMO’s asset allocation department. We stand with the critics,” Grant writes.
The basic idea is that equities provide the majority of returns for a typical 60/40 portfolio, but they also make up as much as 90% of the portfolio’s risk (in the sense of volatility). Risk parity, as the name implies, allocates risk equally between stocks and bonds (possibly adding other asset classes to the mix), using leverage to even things out. Once the base portfolio is put together, you can lever the entire thing to the level of risk you’re comfortable with and get a better Sharpe ratio than the 60/40 portfolio would have given you. As a practical matter, that means you’re going to end up buying a lot of bonds on leverage.
But leverage makes portfolios riskier in a way that isn’t well captured by volatility. When the market sours on some stock, or industry, or asset class (like nearly every credit instrument in 2008), an unlevered investor can choose to ride out the low and wait for his position to recover. It might not, and getting out is often the right decision, but a leveraged investor can be forced to sell and realize losses even when he expects the assets to rebound in a few years. Asness has said that investors should have a plan to shed leverage as losses pile up and then take leverage back on at the market bottom, but market timing is by no means easy to do and investors’ plans to protect themselves don’t always work out.
Negative skew to asset returns exacerbates the path dependence problem. If you lever up to buy corporate credit, for example, then your gains will come gradually in the form of interest payments, but your losses can come swiftly in the form of defaults. Even if the long-term average would put you ahead, the loss of principle caused by a levered position collapsing can prevent you from taking part in the long-term.
Grant objects that risk parity ignores the ‘loss of principal’ sense of risk that clients often care about, and that it doesn’t pay enough attention to underlying value. But he also just doesn’t see any asset classes that are so attractive you would want to lever up to buy them. Backtesting shows risk parity outperforming 60/40 portfolios, but the last 30+ years in particular have been great for anyone loading up on bonds. Even if you don’t want to try to time the market it’s clear that yields have a lot more room to move up than down, so even Treasuries (with no chance of default) look negatively skewed in the current environment. A sudden spike in rates may not seem likely, but it’s more likely than a sudden plummet from current levels.
Posted by Bud Fox at 5:11 AM