Risk parity

Asset allocation, risk, diversification and rebalancing. Pros/cons of hiring a financial advisor. Seeking advice on your portfolio?
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Risk parity

Post by Quebec »

"Risk parity" has been mentioned in passing in various threads but there isn't a dedicated one yet I could find. According to Wikipedia,
Risk parity (or risk premia parity) is an approach to investment portfolio management which focuses on allocation of risk, usually defined as volatility, rather than allocation of capital. The risk parity approach asserts that when asset allocations are adjusted (leveraged or deleveraged) to the same risk level, the risk parity portfolio can achieve a higher Sharpe ratio and can be more resistant to market downturns than the traditional portfolio.
Risk parity is a conceptual approach to investing which attempts to provide a lower risk and lower fee alternative to the traditional portfolio allocation of 60% stocks and 40% bonds which carries 90% of its risk in the stock portion of the portfolio (...) The risk parity approach attempts to equalize risk by allocating funds to a wider range of categories such as stocks, government bonds, credit-related securities and inflation hedges (including real assets, commodities, real estate and inflation-protected bonds), while maximizing gains through financial leveraging.
A good explanation is provided, and a number of research papers are listed here, and here is a quote from this website:
The Risk Parity framework seeks to keep portfolio volatility levels constant throughout all market conditions. Allocations are determined by the contribution of risk or volatility of an investment class to the overall portfolio, rather than its anticipated return. As a result, the risk contribution of more volatile asset classes such as equities and commodities is reduced, whilst the risk contribution of less volatile asset classes, such as bonds and interest rates is increased.
Some institutional investors in Canada are doing it, for example OMERS says it now manages it's "beta" portfolio this way instead of the old 60/40 approach. It all sounds good, but can it be done by Canadian DIY investors with an ETF portfolio rebalanced once a year?

It seems obvious that replicating something like the AQR Risk parity fund is not straightforward, as it invests primarily in futures.
User avatar
newguy
Veteran Contributor
Veteran Contributor
Posts: 8088
Joined: 10 May 2009 18:24
Location: Montreal

Re: Risk parity

Post by newguy »

Coincidentally, I was just reading some R programming topics and saw this page A Review of Risk Parity in the side bar.
If everyone could just allocate resources without a forecast then we would not need investment/resource managers. Even with simple econometric based forecasts such as those generated from a ‘simple’ AR(1) model we still managed to beat the RP and EW portfolios. Perhaps a little more attention should be given to the modelling of the underlying dynamics and a little less to ‘fashionable’ trends in asset allocation and catchy phrases meant to provide yet another sales drive, particularly given the resulting highly levered bets on a certain asset class whose rally depends heavily on a printing press which is fast running out of steam.
newguy
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

My (limited) understanding of risk parity is that it involves weighting assets in a portfolio -- whether these assets be bonds, equities, real estate, commodities, whatever -- by the inverse of the variance of their returns or the inverse of their volatilities (betas). As a result, risk parity tends to overweight bonds (low variance) relative to equities (high variance). Such a portfolio tends to do poorly in sharp upswings (like the 1990s) but well during crashes (such as 2007 on).

There is also the problem of identifying future volatilities, which is what matters. Simple extrapolation of past volatilities is possible but doubtful. A better way is to use GARCH models, but these are usually beyond the skill of individual DIY investors.

Of course, this is the most elementary interpretation of risk parity. There seem to be lots of more sophisticated superstructures built upon this idea.

George
The juice is worth the squeeze
ig17
Veteran Contributor
Veteran Contributor
Posts: 3418
Joined: 21 Feb 2005 20:54

Re: Risk parity

Post by ig17 »

Asset Allocation Strategies – Meb Faber

Time frame: 1973-2012

Image

Risk Parity benefits:
- lower Stdev
- lower Stddev (down)
- higher Sharpe ratio
- higher Sortino ratio
- lower max drawdown

None of those things pay the bills.

RispP had the same CAGR as 60/40. As a lowly retail investor, that's all I care about. I'm not going to tie myself into a knot trying to achieve a better Sharpe or Sortino. :D
User avatar
Shakespeare
Veteran Contributor
Veteran Contributor
Posts: 23396
Joined: 15 Feb 2005 23:25
Location: Calgary, AB

Re: Risk parity

Post by Shakespeare »

Given the instability of the numbers, I see these things as of little practical use.
Sic transit gloria mundi. Tuesday is usually worse. - Robert A. Heinlein, Starman Jones
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

Shakespeare wrote:Given the instability of the numbers, I see these things as of little practical use.
Yes.

And even if they were stable, it would be dangerous to extrapolate. As the mandatory warning on mutual funds says about past performance...

George
The juice is worth the squeeze
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Re: Risk parity

Post by Quebec »

The way I understand it, it's a more sophisticated version of the "permanent portfolio". They sometimes call it the "all weather portfolio". You prepare for rising inflation, decreasing inflation, fast growth, slow growth. In contrast, stock-heavy portfolios will not do well in all scenarios.
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

ig17 wrote:Risk Parity benefits:
- lower Stdev
- lower Stddev (down)
- higher Sharpe ratio
- higher Sortino ratio
- lower max drawdown

None of those things pay the bills.

RispP had the same CAGR as 60/40. As a lowly retail investor, that's all I care about. I'm not going to tie myself into a knot trying to achieve a better Sharpe or Sortino. :D
Cliff Asness advocates for risk parity.

As I read him, the idea is not the risk weights in solation. Rather, the higher Sharpe ratio allows you to leverage so as to achieve the same risk level with higher returns. Indeed, use of leverage is a critical part of risk parity strategies.

In summary, construct your ideal portfolio. According to Asness, it will include a lot more bonds (and other non-equities) than the traditional 60/40 portfolio. The result may not yield an acceptable return going forward. Instead of increasing your return by increasing your equity allocation, stick with your portfolio and leverage it until you get the expected return you want (presumably with an acceptable level of risk).

As to the criticism that "you can't eat a Sharpe ratio", Asness retorts:
•One of the arguments against risk parity and other quantitative techniques that attempt to balance risk and return is “you can’t eat Sharpe ratio.” This leads to the “stocks for the long run” idea that equity concentration, generating a high expected return but not the best Sharpe ratio, is not really a risk to the very long-term because the high expected return on stocks will eventually bail you out of any hole dug by the horrendous crashes or lost decades. Well, with moderate leverage, a better Sharpe ratio actually makes a fine meal. It gives you the high expected return you get from equities along with the benefits of a higher Sharpe ratio. Same bail out, smaller holes (or same holes, bigger bail out!). It monetizes the gain from lower and more predictable volatility so you can extract more cash flow from your portfolio, or see it grow faster. Risk parity uses leverage because the goal is real, tangible, spendable portfolio performance, not winning an award for the best Sharpe ratio (though if they have such an award it should be called “The Sharpies” and be sponsored by the marker people).
Asness claims that the reason risk parity continues to produce above-average results is that most investors are unable or unwilling to use leverage. For example, many institutional investors have constraints on shorting. As well, many retail investors are psychologically uncomfortable with leverage. (That's certainly true for me.)

I dunno. That's all predicated on bonds being less risky than equities. Right now, I would guess that bonds are actually more risky than equities, especially long bonds. If true, that flips the above argument onto its head.

60/40, with no leverage, looks pretty good to me. My estate can absorb tail risk.

George
The juice is worth the squeeze
User avatar
newguy
Veteran Contributor
Veteran Contributor
Posts: 8088
Joined: 10 May 2009 18:24
Location: Montreal

Re: Risk parity

Post by newguy »

What's a leveraged bond portfolio? Borrow short and lend long I guess. I agree that sharpe ratio is great for maximizing leverage, but it makes no sense for the buy and hold investor to borrow money to buy bonds. It's more for the trader who has a system with a long enough track record to make that determination.

newguy
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Re: Risk parity

Post by Quebec »

I guess what intreagues me is that the allocations that come out are so unlike what I own. Here is a US example, with rounding of percentages:
Stocks 19%
Emerging Mkt Debt 15%
Commodities GLD 15%
CorporateSpread / junk bonds 6%
Nominal Bonds 25%
Inflation Protected Bonds 21%
I have no emerging markets debt or junk bonds. I have 3% gold, not 15%; I have nearly 70% stocks, not 19%, etc. I think I can ignore junk bonds, but I'll have a look at EM debt: what's the stdev, what is it correlated to, etc. There is a CAD-hedged ETF for this asset class.
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

Quebec wrote:I guess what intreagues me is that the allocations that come out are so unlike what I own.
Yup, me too. That's the whole point.

But then volatility doesn't particularly bother me.

George
The juice is worth the squeeze
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

A two-part video by Mebane Faber, comparing risk parity to a 60 / 40 portfolio, and also to an Endowment Portfolio and a Permanent Portfolio. In the second part, he overlays a trend-following strategy and does even better. Finally, he eliminates bonds, and does even better.

At one point Faber says that risk parity has taken on the aspects of a religious or political cause. From what I can see, he's right: Either you believe or you don't.

George
The juice is worth the squeeze
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Re: Risk parity

Post by Quebec »

Quebec wrote: I'll have a look at EM debt: what's the stdev, what is it correlated to, etc.
Well it seems that EM bonds are correlated more to rich-country stocks than they are to rich-country bonds. So they likely would not help diversify my stock-heavy portfolio that much. Vanguard writes:
Unlike a typical efficient frontier—which shows a curvature due to correlation effects—emerging market bonds show a linear impact on both volatility and return (...) Historically speaking, the greater the portfolio allocation to emerging market bonds, the greater the portfolio’s return and volatility. This is because (...) emerging market bonds have experienced higher correlations with equity sub-asset classes as well as higher standard deviations than other fixed income categories.
emerging-market-bonds.jpg
In addition the current yields of EM bonds are low, implying low future returns (e.g. the BMO etf has a weighted average yield to maturity of 5%, compared to the 10% annualized return in the table above). Another concern is that in a panic/flight to quality, rich-country investors would (temporarily) ditch EM bonds in favour of "safer" ones, so EM bonds would not help a stock-heavy portfolio in such an environment.

So EM bonds look like interesting diversifiers for bond-heavy portfolios (such as the one reported upthread, which seems representative of "risk parity" portfolios), but not so much for stock-heavy ones. But then I shouldn't be investing in EM equities either and I do, so more thinking is needed about the "all weather" concept with the four economic regimes. I'm well equiped for "growth rising" (stocks, corp bonds included in the broad indices), and for "inflation falling" (govt bonds, stocks). I'm OK for "growth falling" (govt bonds, some inflation-liked bonds). I'm not ready for "inflation rising" (a tiny bit of gold, no EM debt, not enough inflation-liked bonds).

Edited: replaced "no commodities" by "a tiny bit of gold" in the last sentence.
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Re: Risk parity

Post by Quebec »

Well I have not quite given up on EM debt.
Data sources: Libra compilation for everything (all returns in Canadian dolalrs) except EM debt in US dollars (Barclays US EM Bond) from http://www.bogleheads.org/wiki/Emerging_market_bonds, and the final year is a different
Data sources: Libra compilation for everything (all returns in Canadian dolalrs) except EM debt in US dollars (Barclays US EM Bond) from http://www.bogleheads.org/wiki/Emerging_market_bonds, and the final year is a different
The annualized performance of EM bonds over the last two decades was excellent (in US dollars), with relatively low standard deviations. Probably not to be repeated.

EM equities, on the other hand, were very volatile with no adequate compensation in the form of extra returns.

Added: US dollar returns are adequate for EM debt because the canadian investor would likely hedge his USD exposure for this asset class.
Last edited by Quebec on 17 Dec 2014 08:02, edited 1 time in total.
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

Quebec wrote:EM equities, on the other hand, were very volatile with no adequate compensation in the form of extra returns.
Tell me about it. I invested in VWO because, well, I believe in wide diversification. It makes you buy what you hate. Stick with the plan. And so on, and so on.

Down a cumulative 24% today from what I paid.

Isn't theory wonderful?

George
The juice is worth the squeeze
User avatar
Peculiar_Investor
Administrator
Administrator
Posts: 13271
Joined: 01 Mar 2005 14:52
Location: Calgary
Contact:

Re: Risk parity

Post by Peculiar_Investor »

ghariton wrote:Isn't theory wonderful?
Have you read the linked article in http://www.financialwisdomforum.org/for ... 88#p540388? Have you contemplated the timeframe behind your question?
Imagefiniki, the Canadian financial wiki New editors wanted and welcomed, please help collaborate and improve the wiki.

Normal people… believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet. – Scott Adams
gsp_
Veteran Contributor
Veteran Contributor
Posts: 2318
Joined: 12 Apr 2011 17:48
Location: Montreal

Re: Risk parity

Post by gsp_ »

ghariton wrote:Down a cumulative 24% today from what I paid.
Could be worse. You could have bought a subset of VWO (RSX), watched it drop 25% in a few weeks, bought some more at the close last Friday, only for it to drop 12% on Monday. :oops: All in an RRSP to boot.
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

Peculiar_Investor wrote:Have you read the linked article in http://www.financialwisdomforum.org/for ... 88#p540388?
Yes, thanks to your link. I've been reading Cliff Asness intermittently for some ten years now.

Referring to that article, if it takes four decades for things to revert to normal (or trend), does that imply that the optimal rebalancing interval is forty years? :wink:
Have you contemplated the timeframe behind your question?
:D

Yes. Of course I'm going to keep holding (I'm buy-and-hold forever, remember?) But I can be grumpy about what's happening. Remember, this forum is a substitute for whining to a financial adviser (who probably wouldn't return my calls anyway.)

George's three laws of investing still hold:

(1) Diversification

(2) Low cost

(3) Patience
The juice is worth the squeeze
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Re: Risk parity

Post by Quebec »

I've been playing with a mean variance optimizer using nine asset classes (the maximum allowed). For the initial portfolio I used 0% T-bills, 20% Can bonds, 10% RRB, 20% each Canadian, US and EAFE stocks, 5% EM stocks, 5% gold, and 0% EM bonds (which is not far from what I actually have). This has an expected return of 7.5% with a standard deviation of 10.1%, based on my calculations of annualized returns, standard deviations and correlation coefficients for the years 1993-2013. If I play with the coefficient of risk aversion until the standard deviation of the optimized portfolio is the same (10.1%), I get a 9.3% expected return from the following allocation: 10% can bonds, 20% US stocks, 10% gold, 60% EM bonds, and 0% everything else. Lowering risk aversion enough leads to 100% EM bonds (obviously as EM bonds had the highest return over the period), but with a higher stdev.

Taking the EM bonds out of the equation, we can get the same stdev with 41% can bonds, 52% can stocks, and 7% US stocks, and 0% everthing else, but the expected return of the optimized portfolio is "only" 8.3%. Still, it's higher than the initial return, with more bonds in the portfolio, because the bonds returns were high (7%) in the considered period.

Of course this uses the past to predict the future, is highly sensitive to the input data, etc. So pretty useless in real life.
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Re: Risk parity

Post by Quebec »

Well not that useless. Here are the correlations:
Source of yearly returns: http://libra-investments.com/Total-returns.xls and Bogleheads wiki
Source of yearly returns: http://libra-investments.com/Total-returns.xls and Bogleheads wiki
Studying the returns, stdevs, and correlations on their own (without playing with MVOs), plus thinking about the "all-weather" concept, leads me to conclude that:

- EM bonds have been wonderful, but may not be in the future. I'm staying out.

- EM stocks have been very volatile (30% stdev!), and have returned less then developed market stocks as a group in the period 1993-2013. Even if EM stocks return more over the next decades, they are likely to remain very volatile, and not great portfolio diversifiers for canadians (e.g., correlation of .78 with the TSX Comp). Three types of stocks (Can, US, EAFE) should be enough. I only have 3% EM equities, and I'm selling them to buy nominal Canadian bonds, which are much better diversifiers for stock-heavy portfolios.

- I like the idea of increasing my target allocation to RRBs but the yields are very low... and inflation seems controlled... So maybe later.

In conclusion, this was a lot of reading and thinking for a very modest change in allocations (sell 3% EM stocks to buy more Canadian bonds). But there is now one line less in my AA spreadsheet!

I don't think "risk parity", with leverage, will become popular among individual investors. It involves buying leveraged ETFs (which reset daily). Or taking a HELOC at 3-3.5% to buy bonds which will return maybe 2-2.5% in the current environment. This makes no sense to me.
George$
Veteran Contributor
Veteran Contributor
Posts: 2612
Joined: 18 Feb 2005 20:46
Location: Toronto

Re: Risk parity

Post by George$ »

Quebec wrote:"Risk parity" has been mentioned in passing in various threads but there isn't a dedicated one yet I could find. According to Wikipedia,
Risk parity (or risk premia parity) is an approach to investment portfolio management which focuses on allocation of risk, usually defined as volatility, rather than allocation of capital. The risk parity approach asserts that when asset allocations are adjusted (leveraged or deleveraged) to the same risk level, the risk parity portfolio can achieve a higher Sharpe ratio and can be more resistant to market downturns than the traditional portfolio.
Risk parity is a conceptual approach to investing which attempts to provide a lower risk and lower fee alternative to the traditional portfolio allocation of 60% stocks and 40% bonds which carries 90% of its risk in the stock portion of the portfolio (...) The risk parity approach attempts to equalize risk by allocating funds to a wider range of categories such as stocks, government bonds, credit-related securities and inflation hedges (including real assets, commodities, real estate and inflation-protected bonds), while maximizing gains through financial leveraging.
A good explanation is provided, and a number of research papers are listed here, and here is a quote from this website:
The Risk Parity framework seeks to keep portfolio volatility levels constant throughout all market conditions. Allocations are determined by the contribution of risk or volatility of an investment class to the overall portfolio, rather than its anticipated return. As a result, the risk contribution of more volatile asset classes such as equities and commodities is reduced, whilst the risk contribution of less volatile asset classes, such as bonds and interest rates is increased.
Some institutional investors in Canada are doing it, for example OMERS says it now manages it's "beta" portfolio this way instead of the old 60/40 approach. It all sounds good, but can it be done by Canadian DIY investors with an ETF portfolio rebalanced once a year?

It seems obvious that replicating something like the AQR Risk parity fund is not straightforward, as it invests primarily in futures.


Hi - I did start trying to follow - but I am getting tired of complexity - that leads to nothing or very little. And so thought that it probably is not worth my limited effort.

Yet - "risk" has and will always interest me. Have you read Howard Marks recent -Risk Revisited

I don't think Marks refers to "risk parity". Anything by Marks is worth reading - in particular about "risk".
“The search for truth is more precious than its possession.” Albert Einstein
User avatar
ghariton
Veteran Contributor
Veteran Contributor
Posts: 15954
Joined: 18 Feb 2005 18:59
Location: Ottawa

Re: Risk parity

Post by ghariton »

George$ wrote:Hi - I did start trying to follow - but I am getting tired of complexity - that leads to nothing or very little. And so thought that it probably is not worth my limited effort.
Just as well :wink:

Risk parity, as discussed in the industry, has very little to do with the sort of discussions we have about risk on this forum. Rather, it is a newfangled way of defining asset allocation weights for one's portfolio.

Traditionally, most of us use weights for different assets that reflect our tolerance for risk, however defined. Generally, bonds are considered less risky than equities. So if you are risk averse, you increase the allocation to bonds. If you are risk tolerant, you increase your allocation to equities. Of course, your "ideal" allocation might not produce a return that is sufficient for your needs. The traditional remedy is to increase your allocation to equities, which over long periods have produced much higher returns than bonds. This increases your risk, but that is the trade-off you make.

Risk-parity says: No. You should allocate assets to different asset classes according to the inverse of their riskiness. Risky asset classes, such as equities, get very low allocations. Safer asset classes, such as bonds, get much higher allocations. [Technically you are supposed to weight asset classes using weights that are the inverse of their risks, however defined -- perhaps subjectively. Contrast that with market capitalization weights, or more recently equal weights, or fundamental weights.]

But with such an allocation, the expected returns may not be adequate for your needs. Aren't you back to the old trade-off, i.e. increase your allocation to equities? Risk-parity says: No. Instead, leverage your whole portfolio until you get the desired expected return (or the expected riskiness) for the entire leveraged portfolio.

(As newguy points out upthread, the logic of leveraging a portfolio of mostly bonds is a little weird. Presumably you are borrowing short-term and using the proceeds to mostly buy long-term bonds. Isn't that the classic recipe for blowing yourself up?)

The other selling point of risk parity is that you are supposed to consider an expanded collection of asset classes, e.g. real estate and commodities and gold, in addition n to traditional bonds and equities. But many investors already do that as an extension of the traditional asset allocation approach, so it's not characteristic of risk parity.

And now you know everything I know about the subject. :wink:

(the other) George
The juice is worth the squeeze
George$
Veteran Contributor
Veteran Contributor
Posts: 2612
Joined: 18 Feb 2005 20:46
Location: Toronto

Re: Risk parity

Post by George$ »

ghariton wrote: And now you know everything I know about the subject. :wink:

(the other) George
:thumbsup: nuch thanks to George from George$

I vaguely remember that years ago you came in first as "George" -- and so I added $ :)
“The search for truth is more precious than its possession.” Albert Einstein
BRIAN5000
Veteran Contributor
Veteran Contributor
Posts: 9064
Joined: 08 Jun 2007 23:27

Re: Risk parity

Post by BRIAN5000 »

I'm not ready for "inflation rising" (a tiny bit of gold, no EM debt, not enough inflation-liked bonds).
With High growth & High Inflation - I thought Oil, Commodities and Real estate was supposed to be an Inflation hedge you don't have those in your heavy weighted equity portfolio?

With Low growth & High Inflation - RRB's and/or Cash
This information is believed to be from reliable sources but may include rumor and speculation. Accuracy is not guaranteed
User avatar
Quebec
Veteran Contributor
Veteran Contributor
Posts: 1645
Joined: 24 Oct 2009 16:49
Location: Quebec City

Re: Risk parity

Post by Quebec »

ghariton wrote:[Technically you are supposed to weight asset classes using weights that are the inverse of their risks
Correlations are also taken into account in some formulations.
Post Reply