Risk Parity Still Stinks (Part X)

I wanted to continue following up on funds/strategies we've looked at before to see what more we can learn as more time has gone on. For ages, I've been saying I don't think risk parity works in retail accessible funds or ETFs. One of the easier funds to look at in this space is the RPAR Risk Parity ETF (RPAR) which Yahoo Finance shows as being down 12.7% for the year through Friday's close. The Wealthfront Risk Parity Fund (WFRPX) shows as being down 8% YTD through Friday.

A quick reminder that risk parity allocates to assets based on how risky they are although really I'd say volatility is more accurate. Generically, stocks are much more volatile than bonds so balancing out the volatility between the two in a portfolio would call for a much heavier weighting to bonds than equities, often to the point of leveraging up to own enough bonds. From that simplistic idea, other assets can be included  but that is the general idea. Risk parity has done poorly, at least the funds above anyway, because bonds have had a protracted, steep decline and any fund that actually did leverage up to own bonds or even just go heavy long duration has done poorly.

Rodrigo Gordillo made a comment on Twitter saying he thought risk parity was doing fine. He said that RPAR isn't a good proxy for true risk parity due to the constraints of the ETF wrapper. I don't know what he meant by that but he cited the AQR Multi-Asset Fund (AQRIX) as a better example of risk parity. If you go to the page at AQR for this fund you will see that it compares itself to the common 60/40 stock/bonds portfolio. The fund weights exposure to equities, fixed income, commodities, currencies and TIPS by volatility and can go short and use leverage. 

So how has it done versus 60/40? First, going back as far as Portfoliovisualizer can track compared to VBAIX.

 

The CAGR is much less than 60/40 and standard deviation is just about the same. Now just going back to the start of 2018.

The CAGR is a little closer and the dispersion in standard deviation does widen to further favor AQRIX but I am not sure it would make sense to give up over 200 basis points in annual return to reduce volatility by 173 basis points.

AQRIX did give some crisis alpha in 2022, outperforming VBAIX by 635 basis points but there were some years where AQRIX was far behind VBAIX. I don't see anything compelling to think AQRIX could be a core holding with a large portfolio weighting. If you play around with AQRIX in Portfoliovisualizer, maybe you can find a way to use it as an alternative in a much smaller weighting. 

Next, I wanted to circle back to the Absolute Convertible Arbitrage Fund (ARBIX). The strategy is to go long convertible bonds and then sell short the corresponding common stock. If done well, the result should look like a horizontal line that tilts upward as an market neutral type of strategy and that has been the case going back to the start of 2018.

 

I think that sort of return profile is what people would like to see from bonds; slow appreciation with very little volatility. The strategy is exposed to interest rate risk but last year the fund was only down 54 basis points. Bonds went down a lot of course but perhaps that was offset by the stocks sold short in the fund. This year, bonds are again down but outside of the largest tech stocks, most equities are not doing well and ARBIX is up 3.69% through Sept 29th. 

The fund literature bills the fund as an alternative or fixed income substitute so with that in mind, I built out the following. 

 

Results as follows.

 

IEI is considerably shorter in duration than the commonly used iShares Aggregate Bond Fund (AGG) but still is subject to some interest rate risk. You can see Portfolio 1 tracks closely to VBAIX, starting to deviate away when longer duration bonds start to do badly. In 2022, Portfolio 1 outperformed VBAIX by 395 basis points. That's not heroic but IEI wasn't spared that much from the bond bloodbath, dropping 9.51% last year. You can look through the archives to see other portfolios we created to really avoid last year's carnage and funds I actually used (and still do) but the point is that something with ARBIX' attributes can provide returns and volatility that people associate with bonds (repeated for emphasis) with little to no interest rate risk. 

A quick update on the Alpha Architect Tail Risk ETF (CAOS) that we looked at the other day. Most of the fund is allocated box spreads to create a cash like return. Google how they work, done correctly, box spreads are cash proxies. Part of the strategy is to offset the slow bleed that would happen by continually rolling out of the money puts by smaller allocations to other types of option combos. This is an active strategy and while I think the managers have a handle on how to do this, it is not infallible. Despite being a tail risk strategy, it didn't "work" in 2022 via the predecessor AVOLX.

 

As we looked at the other day, AVOLX (AVOLX converted into CAOS in March of this year) spiked up in 2020 which was a legit crash versus last year which was more of a continuous grind lower. I'm not entirely sure why AVOLX/CAOS went down more like a longer duration fund last year

While the Alpha Architect Box Spread ETF (BOXX) wasn't around last year, we can see that it has avoided the bond market volatility of 2023. I'm less likely to consider CAOS than BOXX. If the current manifestation of the bond bear market continues, then we can see whether BOXX continues to stay above that fray. 

The information, analysis and opinions expressed herein reflect our judgment and opinions as of the date of writing and are subject to change at any time without notice. They are not intended to constitute legal, tax, securities or investment advice or a recommended course of action in any given situation.