What Happened To The Harvard Endowment?

Just some quick hits in this post. 

The Tradr 2X SPY Quarterly ETF (SPYQ) will start trading tomorrow (per a Tweet) as originally expected. They need some amount of time to show they can deliver on a quarterly result but I think these would be the best tool for building any sort of capitally efficient portfolio, with idea being to avoid the complexity of multi-asset funds. Instead investors could build it themselves. Again, they need to prove they work but a way to implement in the manner I mean would not be to target a 60% weighting to equities with a 30% weight to SPYQ but put most of that 60% into a plain vanilla fund, maybe 50%, then 5% into SPYQ to get to 60% exposure leaving 5% to put into an alternative. If something hideous happened to SPYQ, the 5% exposure would merely be frustrating, not catastrophic. 

My first job after college was at Lehman Brothers starting in the summer of 1989. One useful piece of advice I got was to stay away from the IPOs and other syndicated deals. The idea behind that idea was that I wouldn't actually want shares in the companies that would ever be made available to me, a guy who cold called for 10 hours a day. All that would be available to me would be shitty companies. Good enough for me, I'm out. 

I think the IPO market now might be different but I still don't get involved in that manner. But that brings us to this potential private equity ETF from State Street in partnership with Apollo. If I'm wrong, I'm wrong but the odds that this thing becomes a repository for shitty deals seem pretty high. We're not a multi billion dollar endowment or CALPERS or one of the teachers unions or whatever. If you absolutely, positively have to have some exposure, as a retail sized investors with a brokerage account, I think the private equity operating companies like Blackstone or KKR would be better bets. Those are not private equity funds, they are operators of private equity funds. I think it would be better to get paid for selling the shitty deal via the operating company than it would be to buy the shitty deal via the new ETF. 

Bloomberg wrote about the struggles of the Harvard Management Company, speaking of endowments. This is the most recent asset allocation I could find from Harvard.

There's no great way to replicate 39% into private equity in the public markets without concentrating into some crazy risk.

This replication considerably rearranges the deck chairs for the equity sleeve. The 5% in BX is a nod to private equity operators without leveraging up like crazy. The "hedge fund" exposure with the three mutual funds are just names we talk about here along with quite a few other similar funds that could be substituted in. VNQ and IEF are pretty plain vanilla proxies for their respective asset classes. 

The long term result doesn't differentiate by that much. 2020 was the only year it lagged VBAIX by more than 5% and in 2022 my Harvard version outperformed VBAIX by 11.86% only dropping 5.01%. I think we've gotten better results from other blog portfolios that were much simpler than this one. 

Newfound/ReturnStacked updated their model portfolios so I took a look of course. I pick on them a lot but they are very smart guys and the work they do is interesting. In one of their many models was a fund I don't recall seeing before, I probably just missed it, the Newfound/Resolve Robust Momentum ETF (ROMO). The fund is sort of a risk on/risk off swinging from equities to treasuries based on momentum. Right now it has 84% in iShares S&P 500, 9% in an EFA ETF and then some random bits in treasury ETFs including a small weight to iShares 1-3 Year Treasury ETF (SHY). So it can hold T-bill ETFs which is important to note.

The Cambria fund has symbol GAA and I think is comparable to what ROMO is trying to achieve. In 2022 it was down 19.5%. I don't understand how a momentum/trend sort of strategy as it proports to be didn't end up in T-bills, or end up in T-bills sooner, which it can own. This reminds me of a mutual fund from Newfound that had symbol NFDIX. It leveraged up 75% each into stocks and bonds and it did badly before finally closing.

The chart is from a blog post I wrote in April. Looking at ROMO, I don't know man, it doesn't seem like they can make their ideas work in the mutual fund or ETF wrappers. This is part of why I have been so skeptical about the ReturnStacked ETFs. Something just seems off.

RSBT is the oldest fund in the ReturnStacked suite, it is 100% bonds with AGG-like exposure and 100% managed futures and is the yellow line on the chart. The blue line replicates RSBT. The red line is an unlevered version of the same allocation and the green line is just the AGG. 

The same exercise with RSST which is 100% S&P 500 and 100% managed futures is a little better but....

It doesn't appear to be solving any problem. I really just wouldn't with any of there funds and I will be the first one to own it if I turn out to be wrong. 

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.

Twenty Years Of Blogging! Part 1

A few days ago I hit the 20 year anniversary of when I first started blogging. I found blogger when I heard about the Mozilla browser, I downloaded it and Blogger was prepopulated in the bookmarks. I wrote a couple of posts that weren't work related, pretty much ran out of things to talk about, so I started writing about markets and investing. 

That original blog was called Random Roger's Big Picture. It has disappeared when I started my side-gig at AdvisorShares, I migrated the site to a different server and then when I left AdvisorShares, that was it. I can see all those old posts in the blogger template but cannot see the site. Much of that content is still on the internet at Seeking Alpha though. 

Back in 2004, there were very few bloggers, it was a new thing. Barry Ritholtz started before me, that much I know and we became blogging buddies talking every so often. Bill Cara was another blogger who started before me that I had a little contact with. Eddy Elfenbein and Cullen Roche came a long shortly after me, sometime in 2005 I believe and Tadas Viskanta from Abnormal Returns also started in 2005. Josh Brown was still a few years away from getting started. Because of how new and how few blogs there were, my blog was featured or mentioned in just about every financial publication, same as the other handful of blogs back then. My blog was the Forbes blog of the year for 2004 in less than three full months of blogging which should tell you how thin it was back then. 

I mentioned Seeking Alpha above. They were brand new at about the same time. I started to contribute posts to them in 2005, the first article there was about TIPS ETFs. I'm not sure, but I have may have been the first outside contributor to Seeking Alpha. There were articles that may have been written just by Seeking Alpha staff or friends of founder David Jackson. Either way I was very early there and for quite a few years had the most followers at about 85,000. The way that number went up though, I don't know if it was a real number or not. After about 2200 articles for them, I had a falling out when I tried to promote a short book I wrote in a post of mine and they deleted that part from a post of mine. After all I'd been through with them for so many years, the fact that they did me dirty on the book, I was done. 

The site is unrecognizable to me now. I'm sure they are making a mint and I know that some of the writers make a fortune for their content. I think what they want (because that's what their readers want I guess?) is articles that just hand out stock picks which is not what I want to write about. A few months ago I tried to submit something to them to see what the process would be like and the feed back was essentially to make it more actionable which I took as what stocks should people buy. I did not try to resubmit it after that.

Blogging led to several opportunities. The first one was writing for TheStreet.com from 2005 to early 2014. That was a big monetization of my writing. The other meaningful monetization was my side gig AdvisorShares which lasted for about four years until 2018. In between those two, I had a great opportunity managing an ETF through AdvisorShares. Amusingly, the fund had symbol RRGR. Phil Bak helped secure the ticker symbol for us at the NYSE. The performance did well but we failed miserably at raising assets. It was a great opportunity that didn't work out. There were two or three other small writing gigs where I contributed articles in the promise of getting paid when, more like if, the site ever had any revenue which they didn't. 

TheStreet.com also led to my appearing on stock market television more times than I can remember. I'd been on Fox News (before there was Fox Business) two or three times and was a regular on CNBC Asia via telephone. Then an article I wrote for TheStreet about solar ETFs got me onto CNBC with Melissa Francis and then I was C-team regular for a couple of years or so. They started calling every week wanting me to come on to talk about things I didn't know about (not their fault, they couldn't have known) and I saw no upside plus back then, going on the air meant driving down to Phoenix which I didn't want to do. My highlight appearance was one of my early ones in the spring of 2008, I was very bearish on domestic financials. Another highlight was in 2015 (I think), right after the Fitbit IPO. I was the bear and crapped all over the stock for being a gadget destined to be replaced by an app. 

Taking the gig at AdvisorShares pretty much was the end of my having a meaningful blog audience. Posts went from having hundreds or thousands of views down to dozens. Where bloggers try to monetize their writing, it's hard to beat a four year gig that paid a full time salary for part time work. I was very fortunate the way that worked out. Obviously, I have great fun with blogging and where I had several opportunities from it in the past, maybe I will again in the future. 

One theme to some of the things I've done over the years is to be willing to do things for free and let the opportunity find me. That has been the case professionally as well as with volunteer firefighting. 

In Part 2, we'll take a look at personal development or lack thereof as the time has gone by.

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. 

Trying To Learn From Risk Parity

This is going to be fun thanks to a research paper by Long Tail Alpha that looks at several permutations of risk parity by reducing or replacing bonds with trend following/managed futures. 

Risk parity equal weights assets by their risk (more like their volatility). Where stocks are far more volatile than bonds (usually), a risk parity program would have to own far more in bonds to equal out the volatility between the two assets. It takes so much more exposure to bonds that it needs to use leverage to get the weighting right. The paper tested several variations of risk parity and using just stocks and bonds, the weightings they used were 78% in stocks and 333% in bonds to give some idea of the leverage. The backtest with stocks, bonds and commodities weighed out at 59%, 282% and 41% respectively. The leverage is usually obtained by using treasury futures. 

There are a few funds that offer one version or another of risk parity; Fidelity Risk Parity Fund (FAPYX) which started trading in 2022, the indexed Risk Parity ETF (RPAR) which started trading in 2020, the Ultra Risk Parity ETF (UPAR) which is a levered up version of RPAR, Wealthfront Risk Parity (WFRPX) and the AQR Multi-Asset Fund (AQRIX) which used to be AQR Risk Parity but does do things with leverage that resembles risk parity. Actually quite a few AQR funds do this. 

One point made in the paper was that running a risk parity program requires regularly (constantly?) reassessing the risk/volatility of the assets held and reweighting accordingly. This makes sense. Look at the 78/333 blend. It seems like there'd be a lot of leeway to make changes if stocks fell 20% over the next two weeks. In that sort of instance it might make a lot of sense to ratchet up the equity exposure either by buying more stocks, selling down the fixed income or both. 

"Risk parity portfolios are particularly vulnerable when their active weighting algorithms fail to predict shifts in asset correlations." If the paper is correct about the need to do this, it undercuts the premise for indexing risk parity which is what RPAR and UPAR do. As index funds, presumably there is no sort of algo assessing correlations or anything else. RPAR has underwhelmed just about every time we have looked at it. 

The authors noted that risk parity did very well for a long time but that "bond based risk-parity failed miserably in 2022." In fund form, it started doing badly long before 2022 which is corroborated by AQR's change to AQRIX in 2019. It had been struggling for a while at that point and so they changed it. 

The table/chart goes back to FAPYX' inception. In the same period Vanguard Balanced Index Fund (VBAIX) which is a proxy for a 60/40 portfolio compounded at 10.89% with a standard deviation of 12.43%. AQRIX has some risk parity attributes but maybe it doesn't really belong? Either way  FAPYX, RPAR and WFRPX are all tough sells. This year they are doing a little better price-wise with indexed RPAR bringing up the rear with a 5% gain through August. 

Just because the investment case for the funds is weak doesn't mean we can't learn from the risk parity paper. Here are some stats adding trend to stocks and bonds in a risk parity program weighted at 67% equities, 268% bonds and 63% in trend versus the 78/333 blend for stocks/bonds we mentioned above.

Stocks/bonds/trend also had a much lower kurtosis. Kurtosis is a very fancy word where the higher the number, the greater the risk of an outlier result like 2022. A lower number is a mathematical representation of smoothing out the ride via fewer/smaller outlier results. They also compared stocks combined with bonds to stocks combined with just trend (no bonds) and stocks/bonds did much better which is surprising. That is explained by the terrific run in bonds for most of the period studied. I have been contending for a couple of years, all that was good about bonds with duration is now gone. 

Then the paper looked at what it called optimized trend which is a combination of trend (managed futures) and carry. We've spent some time trying to figure out carry as ReturnStacked implements it and the paper gives what might be a more useful idea. Trend will go long markets like commodities and currencies and so on that are in favorable trends and sell short markets in unfavorable trends. Weaving carry in to optimize trend limits the strategy to going long markets in favorable trends and with a favorable roll yield aka positive carry aka backwardation. To short a market in an unfavorable trend that market would also have to be in contango, have a negative carry. That gave improved results as shown here.

Surprisingly though, the kurtosis was inferior to stocks/bonds/plain vanilla trend. The weightings for this model were 64%, 251% and 54% to optimized trend. Finally they work commodities in and the result appears to be better leaving commodities out.

Note that the kurtosis of this blend is a a fair bit higher with commodities included. And the weightings to this last one were 51%, 225%. 35% to commodities and 46% to optimized trend.

I am not aware of a managed futures fund that adds in carry in the manner discussed above. I spent some time looking but didn't make a day of it. If you know of one, please leave a comment. 

The manner in which leverage was used in the paper is not accessible to retail sized investors in brokerage accounts which is fine with me. Many terrible market episodes have been caused by misusing leverage and while I am sure risk parity and the like is "different" (I am being sarcastic and snarky)....there can be a use for some leverage in a diversified portfolio like a sub 10% portfolio weighting in a fund that uses leverage. Such a fund is unlikely to end in financial catastrophe even if the fund in question blows up. That is a far cry from 333% in bonds. 

Where true risk parity is out (for me anyway) and I'm not seeing with the funds, the question is whether some attributes of risk parity can work into a diversified portfolio to make it more robust? Again, the leverage used in the paper isn't accessible so I built the following to replicate RPAR. I'm not using RPAR yet because this replication allows us to go back much further than how long RPAR has been trading. The math is only off by a shade using leverage via UST and a little bit of SSO, remember RPAR is leveraged.

Taking the idea from the paper of swapping in managed futures or reducing it, I built the following.

Portfolio 2 splits the duration sleeve of RPAR with half in managed futures and half in catastrophe bonds which provide income without really taking duration risk. Portfolio 3 puts the entire duration sleeve into managed futures. Both Portfolios 2 and 3 are reduced proportionally to take leverage out.

Splitting the duration into catastrophe and trend had slightly better growth, a considerably lower standard deviation and much better Sharpe Ratio. This is probably attributable to trend having some weak years in the 2010's. The CAGR and volatility strikes me as something you might shoot for with a strategy like the Cockroach Portfolio that we've looked at many times or maybe what the Cambria Trinity ETF (TRTY) is trying to do. Both 2 and 3 were up in 2022 while the RPAR replication was down 11%. 

I find this to be interesting but anyone needing normal stock market growth in order for their retirement plan to work, probably isn't going to get it from any of these portfolios. 

The next table adds the actual RPAR in as the benchmark which shortens the time frame and the output is puzzling. In the replication of Portfolio 1, I tried to avoid any sort of qualitative improvement to get a better result. 

The Replication is based on this from RPAR.

In 2020, RPAR did much better than the replication, did a little worse in 2021, 2023 and 2024 YTD and in 2022 RPAR lagged the replication by 11%. If you see an error on my part please leave a comment. 

I've been studying risk parity for a long time. The idea is very intriguing, volatility weighting different asset classes even though the parity part of trying to equal weight the volatility is not something I believe in doing. A portfolio that equal weights assets' volatility is not going to have a normalish weighting to equities which I think is important for most investors. Normalish is a wide range, 40-70% maybe, well 40% might be a little light but you get the idea. 

The volatility and correlation characteristics of the alternatives we talk about regularly here can help solve the issue of equity volatility management. I'm sure it's obvious that is where I am coming from, I write about it almost every day. One concept from way back that we talk about here is taking bits of process from various sources to build your own process. The above probably captures the little bit that risk parity contributes to my process. Portfolios I manage don't really look anything like risk parity but there is influence. 

And since I mentioned TRTY above, here that fund is with the same Portfolio 2 both compared to VBAIX.

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. 

Fundamental, Dividend or Quality?

Yesterday we took a quick look at the state of value investing, prompted by an article in Barron's. Another Barron's article looked a little deeper, looking at value ETFs primarily but also a couple of so called fundamental ETFs. Fundamental ETFs weigh out things like revenue, cash flow, debt and other metrics that seem to overlap with value. Typically the process underlying a fundamental ETF will have a few more moving parts than value funds which look at various types of ratios like price to sales or price to earnings. 

One of the ETFs listed was the Schwab Fundamental US Fundamental Large Company ETF (FNDX) which we've never looked at here. One of the comments, always read the comments, said that Schwab US Dividend Equity ETF (SCHD) was better. The reader said that the ten year return was identical but that SCHD paid out more income than FNDX. The reader was right for ten years. For five years, FNDX compounded at 15.51% versus 13.60% for SCHD but yes, SCHD paid more in dividends for the five year period as you might expect. Depending on your tax situation though, more dividends may not be better.

This table captures ten years.

For the same period, the S&P 500 compounded at 12.94% with a standard deviation of 15.24%. All four of the funds in the table seem like they are sort of related, value-ish. 

Check that. The older quality factor funds' performance doesn't really differentiate all that much from market cap weighting. Invesco Quality (QUS) was about 240 basis points better than market cap weighting in 2022, but you can see in the bar chart the others were all dramatically better. Despite the crisis alpha in 2022, there was none to be found in the 2020 Pandemic Crash from FNDX, SCHD or SPYV but somehow the Invesco Quality ETF was better by 5-6%. In the mini crash at the end of 2018, FNDX and SPHQ did worse than the S&P 500, SPYV was about the same as market cap weighting and SCHD was the best performer for that event.

In trying to sort out why there was pretty much no crisis alpha in 2018 and 2020 while there clearly was in 2022, my hunch is that 2018 and 2020 were shorter panics driven by exogenous shocks and an important part of 2022's story was the big lift in interest rates which changes all sorts of valuation methods and shines a spotlight on certain fundamental attributes. Please leave a comment if you have a different theory.

While the there is differentiation in performance, less so with quality, there doesn't appear to be reliable crisis alpha with these. That is important for setting expectations. Looking at the above bar chart, value whiffed completely in 2020 and dividends missed completely in 2023. The partial year of 2014, only quality, the green bar, was close to market cap weighted.  

Looking at the ten year compounding numbers, I wouldn't dismiss any of these, they are all valid, and will capture the effect for the most part. I wouldn't buy any of them though hoping for a repeat of 2022 however. Maybe they will all do as well in the next bear market or maybe not. Dividend funds generally got pounded far worse in the Financial Crisis than market cap weighting because dividend indexes were heavier in financials. 

The correlations of all the funds we're talking about is 0.91 or higher. There is no way to know if fundamental or dividends will offer any protection in the next event. More reliable protection in serious market events would come from holding assets with negative correlations or no correlation. 

The 90/10 blend outperformed SCHD and FNDX with less volatility. It held up better than VOO in 2022 but did not offer a ton of crisis alpha that year. It did help some in the 2018 mini crash and helped a lot in the 2020 Pandemic Crash. The appeal from the 90/10 blend, in addition to a smoother ride and some crisis alpha, is that if the market takes a real hit and a client has to take money out near the low (it happens), BTAL can be sold while it is up without permanently impairing capital. The need to protect against a large decline with BTAL is less, after a large decline has already happened. 

Obviously, you probably have your own ideas on how to handle all of these variables. 

I'll close out with some blends involving Invesco S&P 500 Momentum ETF (SPMO). Some interesting things happen with this.

Blending SPMO with FNDX and SCHD as shown above, outperformed the S&P 500 for ten years with less volatility. Both also did deliver crisis alpha in 2022 which is good but I just wouldn't rely on that in the future. 

And adding BTAL into the mix is perhaps more interesting still.

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. 

Be Careful Being Too Academic

The Up & Down Wall Street column in Barron's looked at how Friday's PCE report validated the FOMC's decision to cut by 50 basis points this month. The conversation drifted into a look at value stocks' performance which teed up Cliff Asness to make the case for value. Cliff is obviously bigger than the typical thought leader but there is an element of why say in 100 words what you can say in 1000 words to his writing style. 

In limited space he made an argument for why to invest in value now. Nick Colas from DataTrek contributed a great retort saying “we must trade the market we have, not some academic ideal.” It's a great forest for the trees observation. If you have a diversified portfolio then you have some value stocks in there already. From the top down, value has lagged for an extended period but it hasn't been catastrophic.

If you look at the holdings of any large cap value fund, you'll see plenty of names that have done well, better than the S&P 500, in line with growth. At some point value will collectively outperform again and if you don't want to be in the business of guessing when that might happen, then for anyone who owns individual stocks, you should probably own at least few names already, If you own SPY or a similar fund then you already do have value stock exposure. 

Corey Hoffstein sat for a long podcast with The Algo Advantage that is worth listening to. It went it great detail about managed futures replication versus implementing the full strategy as well as different fund structures. There's an interesting dynamic with this because ETFs are clearly the more attractive wrapper for investors, it's not even close, but Corey made full managed futures implementation in an ETF sound like a bad idea based on several technical factors related to things like ETF market makers being able provide creates for new assets and no limit to how big an ETF can get. Mutual funds can close of course but if a managed futures ETF grows to several billion then it might have a difficult time trading in some smaller markets making it unable to do what it says it will do. I don't know if any managed futures trading program trades Pakistani rupees as an example, but if they do, at too big a scale, it might have to stop trading that market was the thought process. 

It was very technical but worth the time. The reason to mention it is that it sounded like ReturnStacked hopes their next fund will combine bonds and merger arbitrage. I write about these a lot because I am both fascinated and skeptical. With a nod to Colas who articulated it very well, these might just be too academic.

The table goes back to the October, 2023 inception of ReturnStacked Stocks & Managed Futures (RSST). It's not that capital efficiency can't work but somehow, anytime I look, these particular funds are underwhelming. 

Managed futures has generally struggled lately just kind of grinding around which as I said almost all the way through the 2010's is what you might expect a strategy that is usually negatively correlated to equities to do. Stocks up a lot, a thing that is negatively correlated to the thing that is up a lot has a good chance of being down. That doesn't invalidate managed futures but you might not want managed futures weighing down your equity exposure as it arguably has done under the hood of RSST. The fund is marketed as a core equity holding with managed futures "on top." If they think that is what it has actually delivered, not arguing with them, then I guess that is not how I want to access either segment. 

RSST has outperformed client/personal holding Standpoint Multiasset (BLNDX) but with a higher correlation to the equity market and a much higher standard deviation than BLNDX. I think they have much different attributes. 

The picture with the managed futures/bond blend (RSBT) is even worse. RSBT has AGG-like bond exposure and while I want no part of that type of exposure, anyone willing to accept it inside of RSBT had AGG's gain wiped out. Looking back to when RSBT started trading in March, 2023. Portfoliovisualizer has its total return at -5.03% versus +5.52% for AGG.

I've been skeptical of Returnstacked's ability to pull these blends off. RDMIX which is related has been no great shakes and Corey's old 75% Stocks/75% Bonds leveraged mutual fund did very badly before it closed. There are fund providers who do pull this off including AQR. The ReturnStacked guys run intellectual circles around me so maybe I'll be proven wrong about these funds but we've looked at them repeatedly as they've launched and the results just haven't been very good.

I checked in on the YieldMax Short Tesla Option Income ETF (CRSH). The fund is synthetically short Tesla common stock and sells puts against the short position, covered puts.

The chart compares CRSH to Tesla common stock and TSLY which is the covered call version of CRSH. CRSH has paid out $3.59 in dividends so far, it hasn't been around too long. The dividends would add 15.8% back into the 37% price decline. Holding on to a fund like CRSH is a tough way to make a living. The odds of this going down 90% on a price basis and then reverse splitting are pretty high. TSLY already did this back in February after just 15 months of trading. If you have to own any of these very high yielders, I would suggest reinvesting the dividends, not spending them.

The theme to these three ideas today is to be cautious with academic and complex investment strategies. Plain vanilla equity exposure is about as simple as it gets. Managed futures isn't necessarily simple but a small weighting to a managed futures mutual fund, for anyone who believes in that exposure, is simpler way to access the space than going in via a multiasset fund. And think long and hard about leveraging up to build a portfolio. 

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. 

Deconstructing Covered Call Funds & The Fires in Ecuador

A few things today starting with wildfires in Ecuador.

Ecuador has been a favorite retirement destination for American expats for quite a while. We've written about Ecuador in this context several times and followed up in the last couple of years as political and social volatility has escalated. It seems to have mostly impacted Quito while it has primarily been Cuenca that has been the expat destination. Now the wildfires which Bloomberg says are threatening Quito. 

Wildfires have also become far more common in other expat destinations, notably Portugal, Greece and to a lesser extent, Spain. One component of expat living that I hit on in every post about it is the suggestion of keeping a paid off home in the US, renting it out and taking in that income stream. It provides an escape hatch if you need to come back for any reason. The odds of being priced out five or ten years after you leave are pretty high. 

The reasons someone would need or want to come back could include health reasons or some sort of serious political upheaval. Now, we need to add climate change to that list? Regardless of where anyone stands on climate change, there have been more wildfires beyond the American west and, knowing a little about this subject, it is not obvious they really know how to manage wildfires in some of these places. When you roll up on a couple of acres on fire, it can be as simple as put the wet stuff on the red stuff, but the incidents occurring in these places, like Ecuador and Portugal now, are far more complex. 

If you move someplace you think is beautiful and then it burns down, maybe you wouldn't want to come back to the US but it would be nice to have the option. 

I sat in on a webinar about model portfolios which turned out to just be a sales pitch, I was hoping to take in a little about portfolio construction process. The webinar whiffed on that but toward the end there was one useful nugget, more of a confirmation of what we talk about here all the time. 

One of the presenters talked about predictability being a feature of the models they build. I have no idea if their models offer any predictability or not but it is a priority in how I do my job and like I said, we talk about here all the time. We talked about it just yesterday with this graphic.

Adding tech should increase volatility and longer term should add to performance and utilities should do the opposite. There are times in the cycle where more volatility is good and other times where it is not ideal. If you don't think you can predict when you'd want less volatility (most people won't be able to do this) then you need to figure out how to build a portfolio that can let you endure the entire cycle. 

I didn't mention this yesterday when I posted that image but in 2022. the XLU blend was down 16.23%, SPY was down 18.19% and the XLK blend was down 19.14%. That is a great example of the predictability referenced in that webinar and what I mean when I am talking about it. The utility blend should hold up a little better in a serious decline. It may not always do so, there is no guarantee, but that is a reasonable expectation to have. 

Then I sat in on a second webinar put on by ProShares about their suite of 0dte covered call funds. I've disclosed owning the ProShares S&P 500 High Income ETF (ISPY) which is part of this suite. I bought it when it first listed to test drive it for possibly adding across the board in client accounts. When you listen to these presentations you really need to sift through the sales pitch aspect of them and this one was no different. 

A big part of the pitch here is better upcapture than the monthly products. We talk all the time about Global X S&P 500 Covered Call ETF (XYLD) which is a monthly call writing strategy and how little it participates in the stock market's upside. According to Yahoo Finance, in the last five years XYLD is down 16% on a price basis. If you spent all the dividends, you'd be down 16%. To be clear, that is not the total return but we are talking about upcapture and XYLD hasn't had any. In the same five years, the S&P 500 is up about 88%. 

In the sales deck they put up a table that only went through June 30th and showed ISPY being very close to the S&P 500. I thought I heard them to say on a price basis but when I looked at all the of the charts they said total return. Maybe I misheard but I tried to make sure I had it right, they were talking about total return.

The above is the index underlying ISPY not the fund, so it goes back further. The total return (price plus yield) in that graphic shows upcapture of just under 90%. Does it stand up in the real world?

Yahoo only tracks price return, not total return. ISPY had paid out $3.10 so far this year which would tack another 7.6% in total return in addition to the 13.2% of price gains. So the NAV upcapture is about 2/3rds and the total return, based on Yahoo's numbers is close to even with the index. I threw the Global X S&P 500 Covered Call & Growth ETF (XYLG) into the chart. That fund sells monthly calls on half the portfolio. In addition to the 13% price gain, you'd add in another $1.06 in dividends which works out to another 3.7% to the total return which is a little less than the total for ISPY. Rounding it out, XYLD has paid $2.89 in dividends this year which adds another 7.3% in total return to the 4.5% from the chart.

Here are the volatility numbers for all four through 8/31.

There was a very interesting chart in the presentation that they spent almost no time on about how to size ISPY into a portfolio.

I have no idea if they actually meant to cut plain vanilla equity exposure in half to add this type of fund in but that is way too much for me. My interest would more complementary that as a core.

There's a lot going on with the next backtest.

Portfolio 1 is 85% SPY and 15% XYLG which goes back much further than ISPY. Portfolio 3 is just SPY. Portfolio 2 is a combo of Invesco Momentum (SPMO) and XYLG that is risk weighted almost the same as 100% in SPY. Blending SPMO and XYLG that way gave a return that annualized out 50 basis points better than 100% SPY which is interesting. Additionally, that blend was only down 11.52% in 2022 thanks to a relatively good year for momentum as we discussed yesterday.

And the portfolio income from this backtest.

We only have five years (partial and full) to study and Portfolio 2 was the best performer in two years and the worst performer in three times. The overall back test is appealing but it would be emotionally challenging to hold. Every valid portfolio we could possibly come up with will struggle at times, there's no getting away from that. 

So, is any of this worth it? You may read this post and decide it is not worth it and that's ok, you'd have put in some time to study and then ruled something out. That's productive in my opinion. 

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. 

How To Win With Factors

John Rekenthaler wrote an article at Morningstar about smart beta or factor funds or as Morningstar calls them strategic beta funds that looks at the space in a manner that is shortsighted. A spoiler, I'm not making an argument for factor funds really but the article missed an opportunity to talk about them in a constructive manner. Excerpt:

What I overlooked, though, was whether smart-beta investors would choose well. After all, smart-beta funds are a cousin of sector funds, in holding one segment of a marketplace. And sector fund shareholders have fared poorly, owing to their habit of buying high and selling low. They chase performance.

There's a few things to unpack there. Chasing heat, chasing performance is a behavioral issue having nothing to do with how an index is filtered toward specific attributes. We'll get to sectors in a bit, but investors do chase performance at the sector level too as well as individual stocks and any other segment of the capital markets. 

There are quite a few different factors and down below we'll look at momentum, quality, low volatility, buybacks and.....and....market cap weighting. Market cap weighting is a factor like any other factor. It is the most common factor but it is a factor. 

Going back when factor ETFs started to really proliferate I talked about how no single factor could possibly be the best for all times. First a table of the five factors I mentioned above.

Looking back at eight years which is as far back as the newest fund goes, Buybacks (PKW) was the best performer in three individual years, Minimum Volatility (USMV) and Momentum (SPMO) were the best in two years each, Quality (QUAL) was the best in one year and Market Cap Weighted (SPY) wasn't the best performer in any of the eight years. Note that we have usually looked at the iShares Momentum ETF (MTUM) for that factor but it underperforms SPMO at almost every turn so going forward, we'll use SPMO for blogging purposes.

Looking at the entire period, SPMO was the clear winner even before this year. Minimum Volatility offered the smoothest ride which is what is should do. QUAL looks identical somehow to SPY and PKW appears to be much more volatile than SPY.

It is possible to tease out some expectations about when a certain factor might outperform? For example, in a down market value will probably do better than growth as was the case in 2022. Small cap tends to do better earlier cycle than large cap but it has been a while since that has come into play. If you spend the time and find the patterns then trying to allocate around those indicators is probably better than flipping a coin but not that much better. My sense is that these sorts of indicators used to be more reliable than they are now and this is not something I would try to chase. 

Chase is the word the Rekenthaler used in the excerpt above. To the title of this post, if there is a way to win with factor funds, besides market cap weighting, it lies in making an allocation to one or more and then sticking with it save for rebalancing when/if necessary. That's not a very satisfying answer, pick one or pick a combo and stick with it but that's it. And I define winning as riding through with a strategy you can sleep with and then having enough money when you need it. No matter what factor you choose, you are guaranteed to lag the others some portion of the time. Based on the table above, the one factor you could choose, including market cap weighting will lag more often than not. Market cap weighting was never the single best in the period we studied with those funds yet it was the second best overall and would have been just fine if it fared worse than second best overall. 

For anyone wanting to venture out beyond the market cap weighted factor, what should they do? I'm not sure. You could just look at the chart and say momentum but you'd need to be prepared to lag by kind of a lot sometimes and not necessarily when you'd expect to lag. SPMO did very well on relative basis in 2022 thanks to a late year comeback which is surprising. 

Although I haven't found it yet, I do believe there is a way to combine two or maybe three factors to get consistently, not universally, better results. Cliff Asness is a fan combining momentum and carry (not an equity strategy) as powerful solution compared to traditional 60/40. We've looked at that before using AQR Multi-Asset (AQRIX) as proxy for carry which was suggested by Microsoft Copilot. When I asked it again today it spat out the iShares Commodity Carry Strategy ETF (CCRV) which is just one slice of carry. 

The timeframe is short because of CCRV but the results are very interesting. Yes, the volatility does uptick but the improvement in performance is noteworthy and even though it was just one test in 2022, both momentum/carry portfolios appeared to have crisis alpha. 

It's a similar story with sectors. Some have the tendency to be more volatile than the broad index which will usually, not always, be good on the way up and hurt on the way down. 

The chart is kind of difficult to look at but you've got most of the Sector SPDR ETFs there. I left out energy and materials because the demand profiles can sometimes be procyclical and at other times countercyclical making them less reliable in this context. Technology (XLK) is the easiest to see. It pretty reliably goes up more on the way up and down more on the way down. In 2022, XLK was 900 basis points worse than the S&P 500. If you buy XLK or some other tech sector ETF you are adding volatility. XLY shows a very similar, reliable pattern. I cut the chart off at late 2021 because XLY has struggled since that point. Amazon is the largest holding and mostly traded sideways and Tesla, the second largest holding is down 32% since the end of 2021. 

The note with XLY makes the point of needing to know what is under the hood of any ETF you own. XLY is still a client holding. Looking forward, I would expect it to be up more on the way up and down more on the way down. It was down twice what the S&P was in 2022 but up considerably more than the index from when I bought it in 2008. Communications is another that tends to fall into the same groove as tech and discretionary but it has struggled lately because of a couple of large drawdowns in Google. I've had better luck with using an individual name for this sector instead of a sector ETF.

At the other end of the spectrum are Staples (XLP) and Utilities (XLU). They will generally go up less and down less pretty reliably although they are vulnerable to rising interest rates. You can see how they've lagged long term but in 2022, XLP was down 0.83% and XLU was up 1.42%. DON'T BUY STAPLES AND UTILITIES FOR LONG TERM OUTPERFORMANCE. If you want to invest to the sector level, I do, think in terms of expectations and probabilities. Tech is going to do what I said most of the time and occasionally it will miss. Utilities will do what I said most of the time and occasionally it will miss. That description has nothing to do with the heat chasing that Rekenthaler is talking about although of course you can chase heat too. 

That last table goes back almost ten years and is exactly the sort of long term result you'd expect. Tech adds volatility and adds to long term performance. Utilities, reduces volatility and acts as a drag on returns. Guessing what these would do year to year is just that a guess beyond the fact that stocks go up about 72% of the time.

Sorry, but I think Morningstar can do better at explaining this stuff.

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. 

Adapting Bob Elliott's Simple Game Plan Portfolio

Today's rabbit hole comes courtesy of Bob Elliott from Unlimited Funds. The starting point was the futility of trying to time the market. He didn't use what I think is a great example choosing all out sitting in cash or all in in a portfolio of 30% equities, 55% bonds and 15% commodities. That allocation was referred to in the post as 'assets.' I don't know how he came to choose that allocation, maybe the Unlimited Hedge Fund ETF (HFND) was similarly allocated at one or more points but doesn't appear to be now based on the fact sheet. Regardless, being invested in 'assets' outperformed cash while obviously being more volatile.

I'm not going to make an argument for actual market timing but I think we can dovetail to a concept that I take from John Hussman, others probably do something similar. There are ways to take an inventory of when risks are elevated or maybe less so. A simple example from previous posts a long time ago is that there is less risk of a large decline in equities immediately after a large decline in equities. When a chart goes up in parabolic fashion, risk of a decline goes up. There are many other equally simplistic rules of thumb that fit in this discussion. 

Quantifying it just a bit, when an index like the S&P 500 goes below its 200 day moving average (DMA), it indicates a problem with demand for equities. The problem may or may not turn out to be serious but still a problem of some sort. In terms of putting on some defense based on a breach of the 200 DMA and to Bob's point, going all out is going to be a very bad bet more often than not. All out is a whole different matter versus putting on a little defense, like adding an inverse fund or a long VIX fund or something else that would be likely to go up when stocks go down. 

Also related to the 200 DMA is how far the index is from the 200 DMA. Although it doesn't happen very often, a 20% gap between the index and the 200 DMA has historically not been sustainable and a good time to buy when the index is 20% below the 200 DMA or a good time to get defensive when the index is 20% above the 200 DMA. Not get out entirely, just get defensive a little whether that means reducing exposure or adding negatively correlated assets.

I've acted on this just a couple of times, it doesn't happen often, with just an incremental change to the portfolio. The mindset should not be that you are trying to time the market like you are hitting a bottom because that is unknowable in real time. The mindset should instead be that you are buying low. If you buy something after a 25% decline, you are buying low but it absolutely could go lower. It is hopefully a more comfortable mindset for a task, buying after a large decline, that is generally uncomfortable. 

Included in the blog post was a link to a blog Bob wrote in 2023 about a portfolio he called A Simplified Game Plan (SGP). SGP allocates as follows;

I started to map out a portfolio to match this allocation idea and then it occurred to me that I probably wrote about this before and sure enough I did. I put together the 2024 version before I remembered that I wrote about it in 2023. I labeled each version accordingly.

There were a couple of holdovers including client and personal holding ASFYX. Part of SGP portfolio drifts into All-Weather by Ray Dalio which makes sense because Elliott worked at Bridgewater. Portfolio 3 below is just QDSIX which strikes me as being a variation of all-weather and I used client and personal holding BLNDX as the benchmark which is marketed at an all weather portfolio. As I read what I just wrote, maybe SGP is just flat out a variation of all weather. 

The 2024 version has a much lower volatility and return because it takes a lot less risk with the diversified alpha sleeve. 10% each into two stocks as we did with the 2023 version is very much a live by the sword, die by the sword approach. QDSIX is a fund of AQR funds whose largest allocation is to AQRIX which is sort of risk parity fund. It used to be called AQR Risk Parity. All four offered crisis alpha in 2022. The worst performer was the 2023 version which was down 2.81% versus a drop of 16.87% for the Vanguard Balanced Index Fund (VBAIX) which is a proxy for a 60/40 portfolio. 

In one of the two posts from Bob was this link to a short Ray Dalio video explaining how he gets to believe 15-20 uncorrelated income streams makes for a Holy Grail portfolio. The results of both SGP versions are interesting but do they pass the uncorrelated return stream test?

Well, yeah kind of. It looks like mostly lowish correlations but we've done better in previous blog posts, still though its decent. The 2024 version was only up slightly in 2023 and is not up very much this year either. The portfolio is valid IMO but as is the case with any valid portfolio there will be periods that try an investors patience. 

And just for fun

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.

No Mobility With Portable Alpha

I went down a portable alpha rabbit hole. We've discussed it plenty, using the terms capital efficiency and return stacking more often that portable alpha. It is essentially using leverage to add exposure beyond a 100% portfolio. Portable alpha a little more specifically is the matching up of beta (simple market exposure like via an index fund) with an alpha seeking strategy. Here's a video from AQR to explain it better than I can, that assumes a "top quartile hedge fund" for the alpha seeking component. 

Here's an old paper from Cambridge Associates looking at portable alpha. They list out some different ideas to consider for the alpha component.

• Catastrophe bonds
• Legal claims
• Global macro 
• Fixed income arbitrage
• Fixed income relative value
• Commodity pairs trades
• Ultrashort duration fixed income
• Direct lending 
• Harvesting the option volatility risk premium

There are some asset classes we look at regularly in the list and several we never have looked at. I modeled out six different portable alpha ideas using funds we've talked about before. Each one is levered up 50% allocating 50% to ProShares Ultra S&P 500 (SSO) which equals 100% into the S&P 500 and then I put 50% into the funds as labeled on the images. SSO has been imperfectly close to twice the S&P 500 far more often than not but of course looking forward, the imperfect closeness could completely unravel. The new Tradr ETFs we've talked about might do a better job with longer term tracking. 

The green line on both charts benchmarks the S&P 500. They all outperform but all but one also dial up the standard deviation versus the S&P 500. Three of the six offer slight improvement in risk adjusted return. Only two of the six, the one with QLEIX and the one with LFMIX. were down less than 15% in 2022. While some of the individual funds have offered crisis alpha in the past, that wasn't necessarily a benefit the way we constructed the portable alpha ideas for this post. Three of the six were down more than the S&P 500 in 2022.

We've talked countless times about using leverage to leverage down. All of these clearly leverage up and do so without helping on a risk adjusted basis. When we do these exercises we see that sometimes the leverage boosts returns a little but not always. For my money there's not much of a basis to think this idea can reliably add alpha. When we leverage down though, we regularly see more meaningful improvement in lower volatility and much better risk adjusted portfolio results than going the leverage up route. 

It's not that this doesn't work, more like it is difficult to make it work with funds available to retail sized accounts. With retail accessible funds, we've had far better luck pulling a little from equity and a little from fixed income, or maybe a lot from fixed income, to make the portfolio more robust. 

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.