Should You Be 100% Trend?

That question came up in a podcast with Meb Faber and Jerry Parker. Parker is a very well known and long tenured trend (managed futures) manager. He's been in the space as a hedge fund manager since the early 80's but has transitioned to the ETF wrapper with the Blueprint Chesapeake Multi-Asset Trend ETF (TFPN) which started trading a year ago and more recently, tying up with Meb to manage the newly listed Cambria Chesapeake Pure Trend ETF (MFUT).

Meb is a huge believer in trend. The Permanent Portfolio-inspired Cambria Trinity ETF (TRTY) allocates 35% to trend. Saying TRTY is Permanent Portfolio-inspired is my impression, I don't know that Meb has ever described it that way. 

If you do some digging, you will find research that says yes, we should allocate 100% to trend. Part of the logic is that by owning what is in strong trends and shorting what is in weak trends, you avoid disaster. Occasionally, there are disasters in stocks and bonds and trend programs should be able to sniff that out. Often they do sniff that out. Meb said that one reason trend did so well in 2022 was that it was short bonds. Jerry said that his money is entirely allocated to trend but he wouldn't tell anyone else to do that not because he doesn't believe in it but because he doesn't think most people can handle the periods where it will lag. This is an important acknowledgement of something we repeat here all the time which is that no strategy can always be best, that even great strategies will struggle at times.

We've gone over the extent to which the 2010's were by and large terrible for managed futures. Could such a terrible run repeat? Probably not but there are plenty of instances where that has been the case including Japan, value stocks and Cisco and Intel are both trading below where they were 24 years ago. 

One question I started asking about managed futures years ago, back when I was side-gigging at AdvisorShares, was how much of a drag, zero percent interest rates were. Most of the actual assets of these funds are in T-bills. A zero percent yield means zero return of course and 5%, like T-bills are paying now, means a lot of return. Anytime I've asked that of people that probably understand the strategy better than I do, they said no. The first person I asked was Kurt Voldeng who kept HFRI data and ran a hedge fund replication ETF for AdvisorShares and I swear I think he thought I had rocks in my head for even asking. 

So maybe the 2010's were a coincidence. This question of "yield on collateral" came up very briefly in the podcast but they didn't address it. I can't tell whether Meb thinks it might be a real factor or not but that it came up means I am not the only one to have raised this issue. 

Managed futures did have a couple of fine years in the 2010's in the context of being a diversifier which is how we view the strategy here. It doesn't have to have high rates to offer diversification but currently collateral is earning 500 more basis points than it used to. 

Meb cited what I think he called a fascinating stat and if he didn't use that word I will. According to Katie Kaminsky at Alpha Simplex, 95% of the assets in managed futures are from institutional investors. It made me feel even luckier to have stumbled across the strategy back in 2007.

So should we be 100% trend? For me the answer is an easy no. I believe it is a fantastic diversifier. Even through the 2010's when it mostly struggled, I repeatedly blogged that it has a negative correlation to equities and equities continued to go up. Meb mentioned the possibility of a reversion to the mean for equities and that is of course possible, maybe it's even probable but equities continue to be the thing that goes up the most, most of the time which is of course a trend in its own right. 

Circling back the Cambria Trinity, it allocates 25% each to equities and bonds, 35% to trend and 15% to alternatives. I've said before that I think that is not enough exposure to equities and the 25% to bonds takes more interest rate risk than I want to take but I like the concept and we can learn some things from the strategy. 

Tweaking the Trinity idea as portfolio 3 above does, gives a competitive return versus 60/40 with a standard deviation of only 8.25 which was even lower than Trinity. In 2022 Trinity was down 3.32% and our spin on the idea was only down 3.51%. Our Trinity tweak also went down quite a bit less than Trinity in the 2020 Pandemic Crash. I used ACWI in the back test because Meb is has a pretty heavy weighting to global stocks in TRTY. BTAL is a client and personal holding.

TRTY has a little over 1% in a Bitcoin ETF which I think is interesting. My guess is that Bitcoin would be part of the alternative sleeve but regardless of which sleeve, you could also call it an allocation to its own sleeve, asymmetry. I've owned Bitcoin for a while for it's asymmetric potential. It also fits into the discussion earlier this week about uncorrelated return streams. Bloomberg had the following chart.

Aside from the asymmetric potential of maybe going to zero or a bazillion, I also believe it is an uncorrelated return stream. I think the chart supports the idea. It does its own thing. A few days ago, Jack Mallers the CEO of ZAP and very much a Bitcoin evangelist said on Bloomberg that he thinks Bitcoin will go to somewhere between $250,000 and $1 million "over the next 12 months or so." That sort of carnival barking makes me cringe and think the odds of zero are higher than I previously thought. It wreaks of "we need more suckers." I'm still in for the asymmetry but this prediction is absolute nonsense. 

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.

Tom Petty Had Thoughts On Portfolio Diversification

Let's continue yesterdays discussion about uncorrelated return streams. Yesterday, we talked about Ray Dalio's investing utopia of blending together 15-20 uncorrelated return streams in pursuit of a beefed up version of his more commonly known All-Weather portfolio. I included the following matrix of various alts and a few fixed income funds that have low to negative correlations to equities and low and negative correlations to each other. 

The matrix goes a long way toward true diversification as opposed to some other ideas/portfolios/funds whose idea of diversification is to own a lot of long and intermediate term bonds. Dalio's All-Weather allocates 55% to long and intermediate term bonds. 

If something is supposed to diversify equity exposure with a negative correlation and the stats back that up, then it is easier to understand why it lags when the stock market goes up a lot. When something is supposed to have a very unvolatile, fairly consistent but low return then it is easier to understand why it lags when stocks go up a lot. Same with something that is truly uncorrelated, it just does its own thing, irrespective of what the stock or bond market is doing. Alt 3 and Alt 4 in the matrix are examples of that. Bitcoin might fit into this description too.

I thought it would be useful to look at some example of diversification attempts that maybe don't work out as well on the matrix below, I blocked out the names of funds that are doing poorly, piling on is not the point.

All of the funds look to offer an enhancement on a balanced portfolio or otherwise be a core portfolio holding. They all blend different assets together with some sort of strategic goal. The timeframe is short because a couple of the funds are newer. While the products are well diversified, maybe HEQT not so much, the diversification is arguably not very effective for the three unnamed funds. 

Something like the Permanent Fund (PRPFX) does look different than VBAIX but the performance is close, I'd say that if it was lagging too, and it has a lower standard deviation. PRPFX is supposed to be a substitute for VBAIX and it meets that expectation. It will not always outperform of course, nothing will but it works. 

The three unnamed funds are trying to do the same thing but somehow they are coming up short. There is a lot of academic research shows large allocations to longer duration bonds leads to very good results and the three unnamed funds have that idea embedded. I frequently talk about the importance of fully understanding something before deviating away from it. The context is usually the 4% rule for retirement withdrawals but it also applies to the research that concludes we should be heavy in long bonds. Any research done on this includes a 40 year period where interest rates went from the mid-teens down to zero. This cannot be repeated. That is a simple observation to make, portfolios will not get a tailwind from bonds that is anywhere near was it was from 1981-2021. The correlation of fixed income to equities has become more volatile and less reliable. It really is that simple. 

In a way, this makes portfolio construction more difficult but to the extent it is more difficult, the ETF and mutual fund industry are offering more accessible sophistication to address the issue. The difficulty is more like being able to sift through the funds that don't quite get the job done in favor of the ones that do. It is important to understand why successful funds are successful.

I write all the time about the effectiveness of using AGFiQ US Market Neutral Anti-Beta (BTAL) to help manage equity volatility.

The two portfolios have identical standard deviations but the blue line allocates 79% to equities versus just 60%, it compounded 291 basis points better than 60/40 and in 2022 it was only down 10.07% versus 16.87 for 60/40. It only takes a little bit of work to find these. As Tom Petty might have said, the sifting is the hardest part. 

BTAL is a client and personal holding. 

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 Many Uncorrelated Return Streams Can You Find?

Eric Crittenden sat for a wide ranging interview on the Show Us Your Portfolio podcast. Eric is the brains and manager of the Standpoint Multi-Asset Fund (BLNDX/REMIX). I've written about this fund many times, I've owned it since it's first or second day of trading and added it for clients shortly thereafter. I met Eric and his partner Matt Kaplan a couple of times in the earliest days of the fund. I'm in touch with Matt occasionally. 

The podcast was a little over an hour and I'd say the first half was more focused on learning about the fund, the strategy and the influences and the second half was a broader conversation about many different topics. 

BLNDX is a combination of equities and managed futures, blended together in such a way as to seek out what Standpoint describes as an all-weather return. 

It certainly would be fair to say that four plus years is a a short time frame but it set an expectation and has lived up to it so far. Not every fund we look at here has done that. 

Eric talked a lot about diversification. He was critical of the typical retirement planning glide path of owning stocks and bonds and adjusting between the two as people get older. Paraphrasing, he said the diversification benefit of bonds against stocks is unreliable. They aren't far enough apart in terms of correlation. He didn't say there's no place for bonds but this led to an interesting part of the conversation.

He cited Ray Dalio's "investment utopia" of having 15-20 uncorrelated return streams for what I think Eric was connecting to all-weather. He said that BLNDX brings in 8-9 uncorrelated return streams. Dalio's idea of 15-20 would appear to go far beyond his All-Weather idea that has been written about extensively, I've written about it a few times too. Dalio's common All-Weather portfolio is defined as 30% in equities, 55% in long/intermediate fixed income, 7.5% in gold and 7.5% broadly diversified commodities. 

Backtesting 18 years at https://tools.archindices.com/portfolio-backtest, Dalio's All-Weather compounded at 5.21% versus 7.61% for 60/40 but had a lower standard deviation, 8.31% to 12.22% and the Sharpe Ratios were within 1 basis point of each other. Shortening up the timeline to the 2020's, All-Weather has lagged 60/40 by 500 basis points annually due to the rough two and half years for longer term bonds. I'm not sure how close Eric and I are on bonds but I see no reason to have more than a token exposure to bonds with duration. 

The idea of many uncorrelated return streams is something I've been implementing in client portfolios since before the financial crisis, it is an idea we talk about all the time here but 15-20 is quite a bit more than I believe I target. I don't doubt Eric's word that his fund targets 8 or 9, I'm just not convinced that the best way, as an end user, to view it. Thinking of BLNDX that way becomes an argument for a huge allocation to the fund.

The matrix is from Portfoliovisualizer. It tracks my alternatives ownership universe and three short term fixed income funds I use. A couple of the alts though look just like short duration fixed income like merger arbitrage. There's a lot of uncorrelated return streams in that table. 

This is the sort of thing you don't really need until you do like in 2022. Unlike Dalio's All-Weather, I don't want to get too far from a "normal" equity allocation unless there is a client mandate for one reason or another. At times, having this sort of influence in the portfolio is a life-saver and at other times it will try patience. Eric made another point that is relevant here. Whatever it is you do, there will be a tradeoff. A portfolio is not going to keep up with Nvidia without a ton of volatility, it cut in half in 2022. 

I tried to find what 15-20 uncorrelated return streams Dalio has in mind but couldn't find the info. Please leave a comment or a link if you have better luck than me finding it. Even without knowing what Dalio had in mind, the matrix reiterates a point we make here frequently which is the fund space, ETFs and mutual funds, are continuing to increase the potential sophistication that retail sized accounts can add to their portfolios. This correlation discussion looks at some complex portfolio construction and it's just a few funds away for anyone who believes in this sort of thing and I realize not everyone does. Maybe most people don't, not sure. 

Eric made one other point to address in this post that was along the lines of trying to create a 75/50 portfolio. As a reminder, 75/50 means getting 75% of the upside with only 50% of the downside. It is very hard to do but if you play around with the numbers, it would work out to a fantastic long term result. In the early days of thinking about creating BLNDX and having conversations about his idea, Eric says he got feedback that he described as 80/40 but that people expected that every day. Since its inception, BLNDX is a lot better than 75/50 or 80/40 in aggregate but not every period. 

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.

Forget About Factors?

In its latest post, Finomial may have poured cold water on factor investing. One way that factors are used is as sort of rearranging the S&P 500 to weight it differently or sampling a smaller chunk of the 500 that have some trait in common. The more common ones are momentum, quality, low volatility, even value and growth could be considered factors too. Other ones that I would add to the list include dividends, buybacks, various option overlay funds might count too and there are also multifactor funds.

These tend to be long only funds. Finomial talked about long/short being better representatives of the factor investing, citing a couple of AQR funds as fitting this bill but those funds seem more like differentiated return streams than hoped for improvements versus market cap weighting. 

My take on factor investing is that I have long been intrigued but it feels like a puzzle I've been unable to solve. I made the following table  with the help of portfolio visualizer. 

To my point about being an unsolved puzzle, there's no pattern or consistency to when a given factor will outperform. Intuitively you might think that momentum would outperform when the S&P 500 is up a lot. From 2015 forward, the S&P 500 was up 20% or more four times. The iShares Momentum ETF (MTUM) outperformed only one of those four, in 2017. In 2018, the S&P 500 was up 18% and MTUM outperformed and it is outperforming so far in 2024. 

I've mentioned the quote from Cliff Asness about momentum plus carry being a very strong combination. I'm intrigued by that idea and have tried to replicate it with funds but haven't gotten there yet. Maybe there is a better momentum fund? Below are some of the bigger ones.

The market is shockingly top-heavy these days and SPMO has benefitted from that more than the others. SPMO will continue to benefit from the current narrow breadth for as long as that lasts. When or if the music stops, SPMO is very likely to get pasted. MTUM's weight in NVDA is close to the plain S&P 500 weight in the name and the others ones certainly are skewing high but that top ten for SPMO is very much a live by the sword weighting. 

And for MTUM

Since it listed in 2015, SPMO has compounded at 16.48% versus 13.36 for MTUM and 13.81 for VOO. If we only look until the end of 2023, then SPMO compounded at 13.45% versus 11.22% for MTUM and 12.74 for VOO. I think the better result up through 2023 is attributable to only going down 10% in 2022 versus 18% for the S&P 500. 

If you have a diversified portfolio, you have exposure to the the big tech stocks that are doing all the heavy lifting this year but being overweight in names that have gone parabolic takes on enormous risk. We can each decide for ourselves whether the extra risk is worth the incremental return opportunity. Yes there have been a couple of years of real outperformance for each of the factors but there have also been some very poor years too. The concept of factors still intrigues me but Finomial made a great point about their spotty performance.  

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 Complexity

We've talked before about the various model portfolios that ReturnStacked ETFs maintains. One of the simpler ones has 20% equal weights to five different funds. It's simple because of the five funds and the equal weight but it is actually very complex for what the funds do. It benchmarks to the iShares Core Growth Allocation ETF (AOR) which is a 60/40 fund.

They call it ReturnStacked Diversified Risk Premia and it owns the following.

• AQR Long Short Equity (QLEIX)
• AQR Diversified Arbitrage (ADAIX)
• AQR Alternative Style Premia Fund (QSPIX)
• ReturnStacked US Stocks & Managed Futures ETF (RSST)
• Simplify Intermediate Term Treasury Futures Strategy ETF (TYA)

I didn't find a look through on the ReturnStacked website to see what the notional exposures are but the model is leveraged up considerably. As of March 31, QLEIX was long 196% and short 169%. It was also long an addition 29% in futures (mostly US equity futures). ADAIX splits between merger arb, convertible arb and event driven. All three are long and short of course but as of 3/31 they were net long. QSPIX is kind of a risk parity strategy in that it is risk weighted long and short equities, commodities, currencies and fixed income. It was 610% long and 578% short. QSPIX is similar to QRPIX which is another AQR fund that we looked at a few weeks ago. RSST is 100% US equities and 100% managed futures. I've seen TYA explained several different ways but in looking at the fund page it has a 300% notional weight to ten year treasury futures which has the effect of extending the duration considerably. As a result, TYA has almost cut in half since its inception in late 2021. 

I modified their portfolio by swapping out RSST because it is so new and replaced it with an S&P 500 ETF and AQR Managed Futures (AQMIX).

The model did well despite how terribly TYA performed. It's still a very short test because TYA only goes back to 2021. The next slide, Portfolio 2 removes TYA and replaces it with iShares Treasury Floating Rate (TFLO) which has very little price fluctuation. 

Avoiding what was obvious interest rate risk adds more that 400 basis points of return per year. For anyone new, I've been writing about avoiding longer duration fixed income for many years. 

The last slide is a longer period thanks to just looking at the TFLO version compared to AOR. With a longer period to study we can see each one has taken turns outperforming. Over the course of 11 full and partial years, the alt-centric portfolio only outperformed five times. The reason the CAGR is so much higher is mostly attributable to 2022 when that portfolio was up 12.7% versus a decline of 15.6% for AOR.

The ReturnStacked portfolio diversifies among many more asset classes/strategies than just stocks and bonds in AOR. The proved out to be more robust in 2022 but not the down year of 2018 when it lagged AOR by over 200 basis points. Looking back, I suppose just about anyone would take that result but there were a lot of years of having to endure weaker, relative returns. The blue line is a better portfolio but it is another example of a portfolio that will be challenging at times to sit with but of course no portfolio can always be best.

In trying to think through different ways to employ what a fund like RSST does into a very simple portfolio. I had one idea that is sort of a game over, someone who doesn't need normal stock growth, but needs to keep up or maybe stay a little bit ahead of inflation type of strategy. 

A 30% allocation to RSST with the rest in a cash proxy might do that. I used Vanguard S&P 500 (VOO) and AQR Managed Futures to get a longer backtest and got the following. 

It clearly does not keep up with 60/40 but that is not the expectation. The leverage means there is 60% notional exposure but only 30% is actually placed into risk assets. In theory, because of the negative correlation between equities and managed futures, RSST should have a lower standard deviation than straight equities but so far it does not. Longer term VOO combined with AQMIX has a slightly higher standard deviation too but the correlation to equities is on the low side.

The worst year for the portfolio above was 2018 when it dropped 3.35%. For most of the back test, there was no yield. Who knows how long T-bills will stay at 5% but yields aren't going to zero. If T-bills had averaged 3%, it would have added about 2.1% in yield to the total return of the blue line result. I also think there is a good chance that managed futures will do better now that rates are off zero. Most of the period available to backtest was during a very dark winter for managed futures. 

Some interesting theory today even if I'm not sure where it will lead us. 

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.