Mostly Voices [0:00]

A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines. If a man does not keep pace with his companions, perhaps it is because he hears a different drummer. A different drummer.

Mostly Mary [0:20]

And now, coming to you from dead center on your dial, welcome to Risk Parity Radio, where we explore alternatives and asset allocations for the do-it-yourself investor. Broadcasting to you now from the comfort of his easy chair, here is your host, Frank Vasquez.

Mostly Uncle Frank [0:36]

Thank you, Mary, and welcome to Risk Parity Radio. If you have just stumbled in here, you will find that this podcast is kind of like a dive bar of personal finance and do-it-yourself investing. Expect the unexpected. It's a relatively small place. It's just me and Mary in here. And we only have a few mismatched bar stools and some easy chairs. We have no sponsors, we have no guests, and we have no expansion plans.

Mostly Voices [1:10]

I don't think I'd like another job.

Mostly Uncle Frank [1:14]

What we do have is a little free library of updated and unconflicted information for do-it-yourself investors.

Mostly Voices [1:25]

Now who's up for a trip to the library tomorrow?

Mostly Uncle Frank [1:29]

So please enjoy our mostly cold beer served in cans and our coffee served in old chipped and cracked mugs along with what our little free library has to offer.

Mostly Voices [1:51]

But now onward to episode 314.

Mostly Uncle Frank [1:56]

Today on Risk Parity Radio, we're just going to tend to your queries.

Mostly Voices [2:04]

You're not going to amount to jack squat! And so without further ado, here I go once again with the email. And?

Mostly Uncle Frank [2:13]

First off, First off, an email from Paul. Look how you've grown up, Paulie.

Mostly Voices [2:20]

I know. I don't understand. When I was a kid, you two were old ladies. Now I'm old and you two are still old.

Mostly Uncle Frank [2:28]

And Paul writes. Happy New Year, Frank.

Mostly Mary [2:36]

Over my holiday break, I have been playing with the portfolio charts and portfolio visualizer tools. While analyzing a variation of the Golden Butterfly portfolio that plots nicely on the portfolio charts risk and return tool, I was alarmed to see it has an overall market correlation of 0.93 when running a backtest at Portfolio Visualizer. This happens despite a mixture of asset classes that you discuss on your show. When I ran the backtest on the regular Golden Butterfly, it shows an overall market correlation of 0.80. What this tells me is that it is very difficult to significantly reduce portfolio correlation despite using risk parity techniques. My first question, is there an optimal amount of overall market correlation to shoot for in a risk parity portfolio? Clearly, the overall market correlation is going to be much higher than I was expecting. Secondly, my Sharpe and Sortino ratios did not seem to be that much better than a 50/50 or 60/40 portfolio. So does that mean on a traditional measurement of risk my portfolio is not really much of an improvement? What am I missing here? Thirdly, should I just be more focused on sustainable withdraw rates in shorter drawdown periods and market correlation? Risk parity does seem to help here despite the higher than expected market correlation. As always, thanks for providing such an insightful show and being willing to answer questions such as mine. Paul G. P.S. I'm attaching some of my portfolio charts and portfolio visualizer results in case they will assist you with your answers.

Mostly Voices [4:10]

He was with the Interior Ministry. Guy's some kind of Russian Green Beret. You're not gonna believe this. He killed 16 Czechoslovakians. Guy was an Interior Decorator.

Mostly Uncle Frank [4:23]

All right, some interesting questions. Let's go through them. And actually let's start with your third question first, where you ask, should you be more focused on sustainable withdrawal rates and shorter drawdown periods than market correlation? And the answer is yes, because that is actually what we are trying to optimize here. That's what we actually care about with respect to a retirement portfolio, because the idea is you want to spend more money rather than less money out of one of these portfolios. And so that's why we focus on that metric. That metric also is a aggregation or conglomeration, if you will, of a lot of these other metrics, including returns, drawdowns, et cetera, so on and so forth.

Mostly Voices [5:12]

Fax mentis, incendium, Gloria, Calpurnia, et cetera, et cetera. Memobis, punitor delectatim. It's all there, black and white, clear as crystal.

Mostly Uncle Frank [5:35]

Now, going back to your first question, is there an optimal amount of overall market correlation to shoot for in a risk parity portfolio? The answer is really no, because that is not really what you care about ultimately. When you're looking at those sorts of things, I mean, you're looking at the correlations of the components of the portfolio overall as compared with each other, you would expect most of these portfolios to have positive correlations with the overall markets simply because most of what is in these portfolios and the main return drivers of them is still stock market funds. So those are going to be 100% correlated with the overall market, at least over long periods of time. And since they also are your primary return drivers, that is also going to skew the metric positive. So getting correlations with the overall market of about 0.8 or so is kind of ordinary or usual for something like this. If your overall goal was to reduce that correlation, you would hold some kind of mostly alternatives portfolio like Larry Swedroe holds in his personal accounts. I'll link to a video about that, but he's got like 25% in stocks and the rest in alternative assets. Something like that definitely would show a very low correlation with the overall markets. But for most people, that would be psychologically very difficult to hold because your returns would not bear much relationship to what is going on in the overall markets. So I wouldn't be concerned with that statistic. All right, going to the second one, my Sharpe and Sortino ratios did not seem to be much better than for 50/50 or 60/40 portfolio. So does that mean on traditional measurements of risk, my portfolio is really not much of an improvement? What am I missing here? Well, it depends on what time frame you're looking at. I've looked at the attachments you made, and I've also recreated something like that to put in the show notes. But when you put in that allocation to commodities that truncated the data set, and so it started in 2007. And so if you look at the period from 2007 until now, that's mostly just a big upslope in the stock market after the 2008 crash. And in recent times, you just have a bunch of volatility. So if you're just looking at that period, you will see portfolios that are mostly stocks just being the best performers on Sartino or sharp ratio bases. Over longer periods of time, actually will see higher sharpened sortino ratios out of these kinds of portfolios. But you do have to be including some of these bad periods for standard portfolios, notably things like the 2000, early 2000s and the 1970s. I ran another back test from portfolio visualizer taking out that allocation to commodities. So giving you a longer data set back to 1978, comparing a portfolio like the one you constructed with a 6040 portfolio. And you'll see the portfolio you constructed is just slightly better than the 6040 all the way along and has a higher Sharpe and a higher Sortino ratio and a higher compounded annual growth rate. So again, while those can be interesting, I wouldn't read too much into them because they don't exactly translate into the sustainable or safe withdrawal rate calculation. The Sartino ratio does a bit more because it focuses on negative performances, whereas the Sharpe ratio is more focused on positive performances. But in order to put all of this stuff together, you'd actually have to include something called skew. which is a measure of how bad are the bad performances versus how good are the good performances and which side has the more volatility to it. And if you look at the metrics tab in Portfolio Visualizer, it'll give you all kinds of different metrics like that, but I'm not sure that's ultimately that useful. But again, ultimately, I think the problem you were having is that that Commodities selection in Portfolio Visualizer has a very limited data set that only goes back to 2007. So you'd be better off if you're going to use something like that to actually pick a fund to go to the ticker symbol analyzer and you could get more data out of that. Although honestly, commodities funds tend to underperform versus a managed futures fund, which is essentially a better commodities fund in some ways, although it trades other things besides straight commodities. The problem with commodities funds over time is that they tend to have a low return profile over long periods of time, over decades of time, and it's very variable. Sometimes it's very high and sometimes it's very negative. And so over a long period of time, often the performance is only a few percentages. like 3% or 4%, whereas if you look at something like a managed futures fund, the general return profile is something more in the 6% to 7% range. But that just is one thing to flag whenever you're using Portfolio Visualizer since the data sets are different for different components. And if you put a component in with a short data set, it will make the whole analysis shorter. and you need to look at what it prints out because it'll tell you if the analysis is limited, which component in there was the limiting component. And sometimes we'll give you an alternative fund or component to substitute for that to get a longer data series to analyze, which we will be talking about in answer to the next question. Well, Laddie, frickin' da.

Mostly Voices [11:51]

So hopefully that will help you some and thank you for your email. Amazing thing about snakes is that they reproduce spontaneously. What do you mean? They have both male and female sex organs. That's why somebody you don't trust, you call a snake. Second off. Second off, we have an email from Greg. Greg, honey, is it supposed to be this soft? And Greg writes, hello there.

Mostly Mary [12:23]

I was comparing VSIAx and VBR on Portfolio Visualizer, and it was showing that over a 30-year period, VSIAx can be expected to perform more than twice as well as VBR. Both started from $100 and from the 50 percentile results, FSIAx ends up with $2,610 and VBR ends up with $1,137. Simulation model configurations were the same, so that's not it. From my understanding, VSIAx and VBR are basically identical aside from one being a mutual fund and one being an ETF. I know in general mutual funds can be less efficient due to cap gain distributions, But as VSIAX hadn't had one since 2000, I'm not worried about this. Is there something I'm overlooking here? Some details about the funds or something under Portfolio Visualizer's hood? Look forward to your answer. I love the pod and have learned so much from it, Greg.

Mostly Uncle Frank [13:24]

Well, I'm afraid I wasn't able to replicate what you were doing, Greg, because there is not 30 years of data for either VSIAX or VBR in Portfolio Visualizer, at least in the backtester. this has some of the same issues that we talked about in answer to the last question. So if you run VSIAx and VBR side by side in the back tester, you'll see that the data for VSIAx only goes back to 2011, but it does give you the alternative fund to use at Vanguard to get a further back test of data. you'll notice though that VBR and VSIAx perform virtually the same because they are virtually the same fund. Now it does helpfully tell you that you should substitute VIS VX for VSIAx because VIS VX is the oldest version of that fund and Vanguard has often several funds or several ticker symbols for essentially the same fund. But if you put in VIS VX, you'll go back to 2004 because that is the oldest date for VBR. And you'll see that VBR performs a bit better than its older mutual fund counterpart, I think because the fees are lower or something of that nature, but they do perform similarly. Now, if you want to go back even further, Remove VBR altogether and just analyze VIS VX and it will take you back to 1998. That's about as far as you can go with these individual funds. Honestly, if you're really interested in just small cap value as an asset class, go ahead and use the asset class analyzer and portfolio visualizer because it has data that goes back to 1972. And that is probably going to be your best bet.

Mostly Voices [15:22]

That is the straight stuff, O Funk Master.

Mostly Uncle Frank [15:26]

Now, I did have a listener at one point give us ticker symbols for like the oldest variation of all kinds of different funds or asset classes and some of them going back, mutual funds going back to the 80s and 90s. Unfortunately, I cannot find for the life of me that episode, although I've searched for it. If any of you are aware of which one it is, Please let me know so I can dig that out because it was useful for people who wanted to do longer back tests on the ticker symbol versions of either the back tester or the Monte Carlo simulator over there. Sorry I couldn't be more helpful there. As for the specific results you got out of this, I'm not sure how you got those. So if you want to send me the links to them, I will Take a look at them, but other than that, I'm kind of at a loss as to why you get different performances for VSIAx and VBR. Hopefully some of that helps. And thank you for your email.

Mostly Voices [16:37]

Well, I must have some punishment coming. Look, Greg, if you know what you did wrong, I mean, that's more important than any punishment we could think of. I do, Mom. I really do. Last off. Last off, we have an email from Stuart. You spit Stuart out this instant snowball, spit him right out.

Mostly Mary [17:04]

And Stuart writes, Frank, you described my really complicated portfolio in episode 291. I had started listening around episode 276, so I was a quasi-new listener at the time. Your description as a really complicated portfolio took me aback a little. I think of it as basically a few simple principles, but it inspired me to go back and do the podcast sequence starting from episode one. It's taken 14 weeks, but I've just caught back up to where I left off. I think I have a much better appreciation for your intent now. This has been a very nice accompaniment to long drives, runs, walks, and mind-numbing housework.

Mostly Voices [17:45]

You are talking about the nonsensical ravings of a lunatic mind.

Mostly Mary [17:49]

I'll be left at loose ends when I get through them all. Running through the episodes in sequence, I heard some repeated items that struck me. In particular, rebalancing talk is a thread through a number of episodes. I will just share my back testing experience using my own code and daily returns. First off, rebalancing is almost an order of magnitude more important with three times LETFs than with on levered assets. This is because both volatility drag and the rebalancing bonus are proportional to the variance, thus the square of the leverage. However, I find it may take hundreds of rebalances with randomly generated returns based on fixed historical statistics to consistently get a rebalancing benefit, and timing luck dominates with just a few rebalances. As a rule of thumb, someone comfortable with annual rebalancing with unlevered assets might have the same risk management using quarterly with half three times LETFs and monthly rebalancing with all three times LETFs. Second off, there is a tendency for better results when rebalancing around the turn of the month, turn of the quarter, instead of the middle of the month or middle of the quarter, which would have been noticeable using three times LETFs. Historical variances and correlation structures appear to be a bit different for the middle of the quarter versus the end of the quarter, at least since 2010. Portfolio Visualizer tends to be lined up on the turn of the month, turn of the quarter for rebalancing, so backtests may be a little optimistic. Third off, I find that a band approach with 10 or 15% range seems to work well to manage risk for three times LETFs while reducing rebalancing frequency. Last off, these rebalancing effects are mostly too subtle to worry about for the sample portfolios. except perhaps the Accelerated Permanent Portfolio and Aggressive 5050. I also thought I would pass along a recommendation for a book called Smart Portfolios by Robert Carver. It is from 2017. This has many similarities with how you recommend portfolio construction, but is more formal and very much in line with my approach as well. A few of your more technically adept listeners may find it interesting. He writes it for investors with portfolios from beginning to institutional scale. The main benefit I take from the book is his handcrafted portfolio approach, in which he explicitly partitions investments according to a top-down risk assignment approach, then converts the risk allocation to the cash allocation. For example, the top level might be partitioned into equities, bonds, and alternatives, maybe with a 40-40-20 risk partition. Then each partition gets independently subdivided into categories. EG, large cap growth, cowbell each get 50/50 of the risk budget for the equities, continually subdividing down to the desired granularity. He also very carefully walks through cost calculations. Thankfully, the need for considering per trade brokerage commissions has aged poorly and allocation sizing. The calculations are suitable for spreadsheets. I do find it interesting that he considers LETFs to be poison. I hope this is interesting for your listeners.

Mostly Uncle Frank [21:11]

Well, this looks like some good work you've done here, Stuart, because this is kind of a frontier area where there has not been a lot of research, really, as to optimizing rebalancing and thinking about how the choice of date may give you random outcomes in rebalancing. that was the subject of a paper and a couple of podcasts or videos with Corey Hoffstein that I'll link to in the show notes that you'll want to listen to. But basically this was a concern that some of the back tests from other people that he had looked at seemed to be particularly good simply because of when they were rebalancing on the given data set. and that if you change the rebalancing, the performance of the portfolio changed markedly. We really don't have that much issue with what we're talking about generally because we're usually just rebalancing once a year on a calendar basis for most of these things. The other difference that if you're looking at daily data, which it sounds like you've put together, that is going to be much more granular an analysis than looking at the monthly data that you're generally going to find in a place like Portfolio Visualizer. I'm not surprised with some of the things you found that rebalancing leveraged funds more often makes sense. And to me, that makes sense because they have a higher volatility. And so the purpose of rebalancing ultimately is to take advantage of volatility swings in one direction or another. So it makes sense to me that you would want to rebalance those more often and also on a Tolerance band basis and not on a calendar basis. I think the people over at Optimized Portfolios, that website might be very interested in this because they're all about constructing portfolios out of leveraged funds. The one thing I wasn't clear on is when you were talking about the band approach using a 10 or 15% range, whether that was an absolute range, as in going from 25% of the portfolio to 40% of the portfolio, or whether that was a relative range, as in going from 25% to 27.5% or 29% of the portfolio, which would be the relative movement there. So if you hear this, please let me know what you were intending to say there. Now, moving to that Carver book. Yes, I've heard of it, but I have not actually read it myself. It's on a list of things I haven't read yet. There are a lot of good books, though, I see that have come out since about 2017, really. And I'm not sure whether that's just because we have more readily available data sets and tools than we did before, or it has something to do with the change in allowing more types of ETFs to be created. But anyway, I do find that a lot of the books coming out since then are much more informative than the kinds of things you saw back in 2008. or so. I think a lot of these books have picked up on the white papers that had been previously published by hedge funds like AQR and places like that. But I will put a link to that in the show notes so people can check it out for themselves. You didn't really have a question, but I thought there was a lot of nice and useful information here, and I thank you for putting all the work into doing this. Top drawer, really top drawer. Because I think this really is an opportunity that we have as amateur do-it-yourself investors that we really didn't have in the past prior to about 2015 or 2016, which is the availability of all kinds of data that we can work with ourselves and we don't have to sit around waiting for some guru or some academic to publish something. And so I appreciate all the effort that you put into this. The best, Jerry, the best. And thank you for your email. Stuart is one of the family now. We do not eat family members. But now I see our signal is beginning to fade. Just one announcement, this podcast will be going on hiatus for a couple weeks and we will return in February. I'm not sure I'm going to be able to update the website, but I will do my best at some point and will certainly get it done when we come back from hiatus. It's not that I'm lazy. It's that I just don't care. And I'll tell you all about that then. In the meantime, if you have comments or questions for me, you can send them to frank@riskparityradio.com That email is frank@riskparityradio.com or you can go to the website www.riskparityradio.com and put your message into the contact form and I'll get it that way. If you haven't had a chance to do it, please go to your favorite podcast provider and like, subscribe, give me some stars, a review. That would be great. Okay? Thank you once again for tuning in. This is Frank Vasquez with Risk Parity Radio. Signing off.

Mostly Mary [26:51]

Risk Parity Radio, Frank Vasquez, risk parity, risk, parity, Frank, Vasquez in the The Risk Parity Radio Show is hosted by Frank Vasquez. The content provided is for entertainment and informational purposes only and does not constitute financial, investment, tax, or legal advice. Please consult with your own advisors before taking any actions based on any information you have heard here, making sure to take into account your own personal circumstances.