Some Thoughts on Risk Parity

This is a very good piece by Cliff Asness on his new blog.  If you don’t follow it then add it.  Cliff is one of the smartest dudes around.  Anyhow, his firm AQR runs a strategy called “risk parity”.  I am not positive where this idea originated, but it was made relatively famous by Asness and Ray Dalio who has been running a form of risk parity for decades.  You can read Dalio’s description of the approach here.

I’m oversimplifying here, but the basic approach is to create a risk balanced approach.  So, if you have a 60/40 stock/bond portfolio most people don’t realize that over 80% of the portfolio’s performance is being driven by the 60% portion because stocks are so much more volatile than bonds.  So you’re not really in a 60/40 when you buy a 60/40.  You’re really in something more like a 80/20.  The idea of risk parity seeks to eliminate this imbalance by making the portfolio risk balanced.  This is generally done by overweighting the low volatility assets in the portfolio to make them equal contributors to the portfolio’s overall risk.  So, if you wanted to make a 60/40 more balanced you might buy a 40/60 stock/bond portfolio and leverage it up 1.5X. This gets you to the same standard deviation, but it beats the living daylights out of a 60/40 portfolio on a nominal and risk adjusted basis (over the last 20 years it generates a CAGR of ~11% vs the 8.4% CAGR of the 60/40 with better risk adjusted figures).

So, what’s good and bad about this approach?   First the good:

• It’s a smart form of strategic diversification.  In other words, this isn’t just some cookie cutter index fund approach that anyone and everyone can implement.  It’s a very sophisticated and value adding methodology.

 

• The focus on fixed income is a nice deviation from so much of Modern Portfolio Theory’s obsessive focus on equity returns and the nonsensical idea that “risk = returns”.  This changes the focus of the traditional debate by ensuring that shareholders are taking a more balanced approach rather than naively jumping into what they think is a “balanced index” like the Vanguard Balanced Index (which actually isn’t very “balanced” at all because the majority of the volatility is derived from the 60% equity portion of the portfolio).

 

• Historically, risk parity portfolios add diversification to a portfolio in a way that reduces overall volatility and increases nominal and risk adjusted returns.

It’s not all good though (sorry Cliff).  There are some cons to the risk parity story (as there are in any portfolio):

• The focus on “risk” as volatility leaves the door open for potential misalignment between the way client’s perceive risk and the way a portfolio’s risks are managed.  Shareholders don’t view risk merely as volatility.  This could result in periods of performance which don’t properly protect shareholders from the way they perceive risk.

 

• There is some significant forecasting error risk involved in risk parity.  As with any allocation approach there is some degree of forecasting, but in a portfolio that is fixed income heavy the portfolio relies, to a large degree, on the positive risk adjusted returns of the fixed income portion.  This means that a risk parity portfolio, is, to some degree, a bullish forecast on the future of bonds (at least more so than a traditional balanced index).  The underlying model also involves some forecasting of changing risk dynamics.  This is difficult, if not impossible….

 

• The portfolios are sophisticated.  The modeling is a little black boxy (is that a word?) because the idea of “risk” can be perceived differently at different points in the market cycle.  How risky are bonds in a world of ZIRP?  Who the eff knows?  Different models will come up with very different answers and so the degree of sophistication in the  black boxy model is a significant driver of future returns.  The investor doesn’t know how this is being done which creates some degree of added risk.

 

• The fees on these portfolios are usually high.  We’re not talking about cookie cutter index funds here.  You’re paying for the modelling and strategic diversification that these portfolios add.  Personally, I am not always comfortable with a 1%+ fee structure, but given the degree of strategic diversification these funds add it could be appropriate for slices of a portfolio for certain people.

 

• Tax efficiency is also a concern. Publicly traded Risk Parity funds tend to have relatively high tax cost ratios when compared to something like a simple 60/40 Balanced Index.

 

• The leverage issue doesn’t scare me as much as it scares some other people, but it can be deadly in the wrong hands.  Leverage is like steroids – in the right hands it can be used in a controlled and intelligent manner.  In the wrong hands it can be very dangerous.  The use of leverage by someone like Asness or Dalio doesn’t scare me.  But you pay for that management expertise.

Now, the average indexer might say that risk parity portfolios are just another form of “active” management or a “better mousetrap”.  Well, I hate to inform you, but all of those Vanguard funds these indexers own are also active deviations from the global cap weighting being paraded as “passive” in order to create brand differentiation.  They’re just different forms of an actively picked index with lower fees.  They’re no less a “mousetrap” than any other index of assets that deviates from global cap weighting.  It’s just that people who buy Vanguard funds don’t often realize they’re in a mousetrap that’s just a lot less expensive than other mousetraps.

On the whole, I think risk parity is a smart approach, but like a lot of Wall Street’s recent innovations it’s probably too expensive to own in any substantial quantity.  That doesn’t mean it’s inappropriate for all asset allocators, but in a world where future returns are likely to be lower than most people expect the fee story becomes a glaring part of the equation.  If I could buy a risk parity portfolio for the same cost as a balanced Vanguard Index then it would be a no-brainer.  But we’re not there yet….

Read some more on Risk Parity approaches:

• BridgeWater on Risk Parity
• Why we lever  -Asness
• The Hidden Risks of Risk Parity – Montier