I just got this from Morningstar - they're tweaking the HBPP. Here's an excerpt:
Implementing Risk Parity How, then, to implement a risk-parity-type portfolio? One of the most popular portfolios of the type is Harry Browne's Permanent Portfolio. It's simple: 25% allocations each to gold, long Treasuries, equities, and cash, roughly balancing one's risk exposures across all four economic configurations. The strategy's performance has been admirably steady for the period 1973-2012. However, it lagged by leagues during the 20-year secular bull market that began in the early 1980s. Its Sharpe ratio, defined as return above cash divided by its standard deviation, was a respectable 0.48, better than the 60/40 portfolio's 0.40. The strategy's Sharpe ratio understates its advantage; it had smaller drawdowns, less fat-tail risk, and held up well when you needed it to.
We can improve upon Harry Browne's Permanent Portfolio. It doesn't hold broad commodities, ignores foreign bonds and stocks, and doesn't adjust weightings for volatility. I've created a model portfolio diversified across all four economic configurations that goes some way toward rectifying the imbalances. It's not designed to shoot the lights out, but rather ensures no one economic environment devastates your wealth.
To see our Risk-Balanced Portfolio, subscribe to Morningstar ETFInvestor.
For $189/yr I can find out how.
Interested in opinions/commentary on this.
Last edited by Ariadne22 on Thu Jul 19, 2012 5:00 pm, edited 1 time in total.
Risk parity is a backward looking measure in many cases I've recently read about. Designing a portfolio based on past volatility alone can have significant problems if the volatility of those assets in relation to each other going forward changes (and it will).
Also the idea of adding in foreign bonds and broad commodities is just diluting the currency protection of the gold allocation. There is no need for someone in the US to go loading up on bonds from Europe for instance (and plenty of currency risk in doing so). Likewise, someone living in Canada probably should avoid loading up on US Treasury bonds so they don't risk importing potential US dollar inflation into their lives either.
With the above said, I suppose the Permanent Portfolio was a risk parity portfolio before the term risk parity was even coined. But the validity of it rests in the asset allocation based on economic environments and not risk in relation to other assets. Doing that kind of thing weakens the model. IMO.
MediumTex wrote:
I will provide the same service for $179/yr, which is a $10 savings over the Morningstar price.
LOL - funny.
Since all of you are so much geekier on this stuff than I will ever be, guess I'm wondering to what degree these comments have validity and how much better would the HBPP performance be with these additions? Or, possibly are they more suited to the VP?
craigr wrote:
Risk parity is a backward looking measure in many cases I've recently read about. Designing a portfolio based on past volatility alone can have significant problems if the volatility of those assets in relation to each other going forward changes (and it will).
Also the idea of adding in foreign bonds and broad commodities is just diluting the currency protection of the gold allocation. There is no need for someone in the US to go loading up on bonds from Europe for instance (and plenty of currency risk in doing so). Likewise, someone living in Canada probably should avoid loading up on US Treasury bonds so they don't risk importing potential US dollar inflation into their lives either.
With the above said, I suppose the Permanent Portfolio was a risk parity portfolio before the term risk parity was even coined. But the validity of it rests in the asset allocation based on economic environments and not risk in relation to other assets. Doing that kind of thing weakens the model. IMO.
That's what I wanted to know. Thank you. Sort of what I suspected. Just like PRPFX has weakened the model.
MediumTex wrote:
I will provide the same service for $179/yr, which is a $10 savings over the Morningstar price.
LOL - funny.
Since all of you are so much geekier on this stuff than I will ever be, guess I'm wondering to what degree these comments have validity and how much better would the HBPP performance be with these additions? Or, possibly are they more suited to the VP?
I would say that they are only Variable Portfolio material. Going into volatility analysis is always going to be very murky stuff.
Ariadne22 wrote:That's what I wanted to know. Thank you. Sort of what I suspected. Just like PRPFX has weakened the model.
I don't think PRPFX weakened the model. It was the first model! I just prefer a more neutral portfolio that doesn't overweight any particular scenario. Problem with risk parity is it is taking past asset class volatility and assuming it will remain the same going forward and that just isn't true.
