Adam: 00:00:25We wrapped up the last episode naming the fact that bond yields are at generational lows. That if markets are even reasonably efficient, other financial assets have been discounted to that same very low rate and therefore the expected returns on basically every major global asset class are lower than average. So it prompts the question, how do investors, institutions etc., meet their liabilities and their funding objectives? I think what may be really counterintuitive is that diversification itself has very interesting properties that may, even if the underlying markets themselves do exhibit low returns over the next few decades, or a decade or more, the rebalancing among a diverse population of well-balanced bets may actually generate enough of a tailwind to be able to achieve those long term liabilities.

Mike:00:01:26And guess what? If markets of the world don’t have low returns over the next number of years, the rebalancing premium will still be realized on top.

Rodrigo:00:01:36In fact, it’ll be higher if you look at the data. So this is not anything that needs to be too contentious, not about a view, it’s not about anything but mathematical properties. We wrote a piece on Shannon’s Demon a while back that looked at two asset classes that were non correlated to each other with high volatility, that were both losing money. And when you rebalance daily against each other, you had a portfolio that made positive returns. That is an extreme case. But we’ve done some work from the asset classes that we know of and their correlations and volatilities, that there is a really valuable rebalancing premium that just requires things to move differently and have volatility. That’s it. Those are the necessities.

Adam:00:02:16The key here is to have a large number of independent moving parts in the portfolio. This brings us to the question of how do you measure diversification? We want to have a large number of unique sources of risk and return. How do you measure that? Well, it turns out, it’s very intuitive. We just examine the volatility of a portfolio if all of the holdings in the portfolio are perfectly correlated. Well, if all of the holdings are perfectly correlated, then how many bets are there in the portfolio? There’s just one, right? So what is the volatility of the portfolio if all of the markets in the portfolio represent just one bet, and then we measure the actual portfolio volatility assuming the true diversification properties of the portfolio. If they’re not all one bet, rather they all are not perfectly correlated, there’s some diversification. The ratio of the volatility if they were all perfectly correlated to the volatility of the portfolio assuming diversification is called the diversification ratio.

And it turns out that you can optimize a portfolio to maximize this diversification ratio. If you do, you’ve got the maximally diversified portfolio. An interesting property of this is that the number of independent sources of risk and return in a portfolio is equal to the square of this diversification ratio. So if the diversification ratio is two, for example, then that would indicate that there are four independent sources of risk and return in the portfolio.

Rodrigo:00:03:59And just to put this into context of the markets that we can all invest in, again, we’re going back to the difference between line items and then having that X ray vision we’ve been talking about often, and understanding the true amount of bets in your portfolio. A while back we’ve kind of shown this a few times in our research, but just trying to identify how many bets there are in different groups of investments. We first looked at 10 different industry classes within the S&P 500, 10. So the headline number is 10 but the amount of effective bets are 1.45. Twenty-five factor portfolios, 1.91 bets. Thirty-eight industries, 2.92. Forty-nine industries, 3.68. Twelve different global asset classes? I think we did this to exchange traded funds. We’re looking at around five now. We’re talking five unique on average bets. But the world that we live in for obvious reasons, is the future space that includes all types of commodities, all types of bonds, equities, and like.

Now we’re looking at an average amount of bets than fall in line between 13 and 14 bets per portfolio over time. These change, this is just the average. But that’s how crucial it is to… It’s not about how many of those 48 different futures contracts only gave us 14 bets. So it’s important to be able to just like we talked about, in the episode about Philip Tetlock, knowing what your Brier Score is, what your ability to be correct in your future estimates. We also think that advisors, investors, anybody who is a fiduciary should know at any given time what the Diversification Ratio is. How many bets? How many actual diversified bets they have in their portfolio. And we’re going to continue to expand on why that’s important.

Mike:  00:05:41Adam, bringing this back to the portfolio, how does the number of unique bets translate into some excess returns in the portfolio? Can you expand on that a little bit?

Adam:00:05:51Yeah. This is obviously where the rubber really hits the road, because nobody cares about diversification itself. That’s not an objective on its own. The objective is to make the most amount of money with the least amount of risk. So let’s just go back to Harry Brown’s Permanent Portfolio and let’s simplify it so that it’s stocks, bonds, and gold. On average over the very long term, stocks, bonds, or gold have about a zero correlation to one another. So if you hold them in equal risk weight in the portfolio, how many bets that we have? Three.

So you can actually model out the expected diversification return that you can get, a Rebalancing Premium that you can get from holding three uncorrelated markets in a portfolio and scaling to for example, 10% portfolio volatility. What you observe is that for three independent markets or three independent bets on the portfolio, you can extract about 1% a year, this is a 1% a year in excess tailwind, purely from being diversified, and rebalancing back and forth between these three different markets as they gyrate up and down. Well, the rebalancing premium that you should expect is a function of the number of independent bets that you have in the portfolio. It turns out that once you go from three independent bets to 5, 7, 9, or more independent bets in the portfolio and you’re rebalancing regularly back to the appropriate weights, you’re able to generate a 3, 4, 5 percent a year excess return just from rebalancing alone. That is the real unsung magic of the type of Risk Parity Portfolio that we advocate for.

Mike:00:07:35If I can translate, the more unique sources of return and risk that you can have in the portfolio with some volatility obviously, the greater the potential free lunch of diversity is and the greater the rebalancing tailwind is.

Adam:00:07:55That’s exactly right. Think about a typical 60/40 portfolio. Now, we talked about the fact that 95% of the risk is in the equity component, a 60/40 portfolio if you measure the number of bets is like 1.05 bets. So you’re getting almost no tailwind from diversification. If you have an 80/20 portfolio, 80% bonds 20% equities, now you’ve got equal risk in both of those different markets. Well, now you’ve got about two independent bets, two independent bets gives you about 0.8% a year in rebalancing premium if you scale that portfolio to 10% volatility. Well, okay, now you’re cooking, it goes to show the importance of both having diversity. You’ve got to have a number of assets that move in different directions at different times for different reasons and holding them in appropriate ways so that you’re maximizing the number of independent bets in the portfolio.

Rodrigo:00:08:49Like anything in our industry there’s many different ways to skin that cat. And in this case, risk parity is a cat. So when you think about the broad idea of risk parity is to have as many bets as possible in a truly diversified manner. But when we X-ray a lot of these portfolios we find that there’s very little thought put into the commodity basket. It is just whatever index you can find or equal weight. There’s more bets in the commodity basket than people tend to understand. And when you look at that Deutsche Bank Commodity Index and the like, we’re putting out a report in the next couple of weeks, it shows that different commodity indices actually, because of their weighting scheme have drastically different unique bets to give and therefore the rebalancing premium is much lower than it could be.

A big issue with risk parity as we see it out there today is that commodities are part of it. The reason commodities are part of it is because there is evidence to suggest that passive commodities don’t have a positive risk premium. But once you take into account the diversity, maximize the rebalancing premium, now you have a basket that can provide as much as three, three and a half percent excess return just by thinking about that problem better. Now you have a portion of your portfolio that can defend against inflation, and provide a passive rebalancing premium that rivals that of the equity risk premium.

Mike:00:10:19So the commodity space is a space that is ripe with diversity. Soybeans are not like crude, which is in turn not like nickel. And nickel is not like copper. Copper is not like natural gas. And so you have this space that has the incredible opportunity to provide these unique bets to your portfolio. Now, you need to integrate that fully into the portfolio rather than letting it sit in a commonly accepted market cap weighting or something like that. Really, you want to, as you mentioned earlier, use the maximum diversification calculation in that space in order to maximize the tail when it comes back to the portfolio. Adam, the Rebalancing Premium Paper was something you took a lot of time with. What were we seeing as you dissected this commodity market? What were the increase in bets? What was the tailwind like from this area?

Adam:00:11:19I think most people will find it pretty astonishing just how large this premium can be if it’s done properly. And just backing it up, if you look across the major commercial risk parity products, typically they’re constructed, you’ve got an equity sleeve, which is a representative basket of global equity futures markets, you’ve had a fixed income sleeve, which is a representative basket of global bond futures, and you’ve got a commodity sleeve, which as we’ve been talking about, typically models one of the major global commodity indices. Now these commodity indices are typically weighted based on flows and or the liquidity of the different commodity contracts. So the priority here is twofold. One is to try and be representative of the inputs to the economic engine. You’re saying that the amount of production of a commodity is representative of its importance in terms of global GDP and global inflation inputs and perhaps more importantly for the stability of these indices, they’re weighting more liquid commodity contracts with substantially more weight, than less liquid commodity contracts.

But when you look through an X-ray, the exposures in these major commercial indices, you see that manifests in a very high degree of concentration. So many of them have over 30% of the portfolio in the energy complex alone. And the remainder is a heavy concentration in metals, maybe some grains. And when you look even more deeply to try and measure the number of independent bets that are expressed in these major commercial indices, we see that because of those, that level of concentration, you’re only getting two to three independent bets from the holdings in these indices. When, if you use the exact same constituents of the indices, you’re not adding any other commodity markets, you’re looking at only those commodity markets that are contained in those indices. But you simply create the maximally diversified version of that portfolio of commodities, you’re able to extract six to seven independent bets from those same commodities. That translates from a tailwind of maybe 2% per year in expected rebalancing premium, to 3 to 4% per year in rebalancing premium.

Again, holding the same investment universe constant, but simply thinking more deeply about how to better diversify that portfolio of commodities. Now extend that same thinking to a portfolio that doesn’t separate the equity component from the fixed income component from the commodity component, but instead thinks of all of these markets as an opportunity to maximize the diversification in the total portfolio. Well, now you’re able to go from a total number of bets in the neighborhood of four or five, for the way that most commercial risk parity products construct their portfolio to 9, 10, 12, 13 independent bets in the portfolio. You go from an expected rebalancing premium in the neighborhood of one to one and a half percent a year, two unexpected rebounds in premium in the three and a half to 4% a year. Obviously, in an extremely low return environment, the ability to add 3% per year more in excess return to the portfolio is fundamentally transformative in terms of expectations.

Mike:00:14:48To put it in context, the equity risk premium is only in the three to 4% range. And so you are literally adding another risk premium to your suite of opportunity sets within a well-diversified portfolio to achieve all of the goals that one needs to achieve with the liabilities of those assets. I would add, it’s interesting when you think of the objective function of the asset owner, their objective function is to meet those liabilities. It’s not to create an ETF or exposure that has capacity. There’s an opportunity there, it’s a little bit more work. But the ETF providers are not trying to maximize diversity in your portfolio to meet your obligations. They’re there to maximize in many cases slightly different ways. But they’re there to maximize the capacity of that particular ETF or throughput.

Adam:00:15:46And we should be clear, if you’re CalPERS or CPP, you will not be able to take advantage of this rebalancing premium, there just isn’t the capacity within the markets to be able to do that. If you want commodity exposure, you will need to go to one of these major indices and take advantage of the liquidity. And even then, because of CFTC limits, you can’t own enough commodities to make a difference in the portfolio. But if you’re a small institution, or a family office, or just a general investor with multi-generational wealth, then you have an opportunity to invest a little bit differently. You don’t need to model your portfolio after CalPERS or CPP, you can model your portfolio after more thoughtful approaches. And this is one of the things you can take advantage of to add tailwind to your portfolio that these larger players just cannot access.

Rodrigo:00:16:31 I would add that large players need to get that protection from inflation from TIPS because it is a much broader market, more liquidity, they can get direct access to that. When we think about inflation, a lot of people think inflation is one thing, but there are many different types of inflation that manifest in many different ways, that will change the pricing in different commodities with different magnitude at different times. When you just say, “I’ll have my inflation hedge here with my TIPS and I’m going to do it. I’m a medium sized institution, I’m going to do something that CalPERS is doing,” you’re missing out on the profit and protection that you may get from having six different unique bets in the commodity space that may pop in that inflation expectation shift. Again, a level deeper than you can go because there’s liquidity. That’s not just diversification premium, it’s a protection that you get from inflation and so taking risk parity to a deeper level is important here.

Mike:00:17:29This is not about predicting. It’s about being maximally prepared. So, however the inflation shock may manifest, by having a thoughtful maximally diversified set of exposures coming from the commodity complex, gives you the best opportunity to make sure you’re there in whatever area of main inflation, that unexpected inflation is manifesting from. So you don’t have to say, “Well, I think it’s going to be the 70s and it’s going to be an oil shock, or I think it’s going to be some fiat debasement of currency and gold or whatever CPI is represented in TIPS.” So you’re really saying, “Well, let me just maximally diversify to all these potential shocks and come what may, and I hope there’s volatility so I can get this rebalancing bonus as well.” And that’s the objective you have to meet the long term obligations that you have, or to have wealth that is sturdy through all of these decades that we talked about earlier, that the different types of regimes that are going to manifest with these two dimensions, inflation and growth and the shocks they’re in.

Rodrigo:00:18:41I think to wrap up with is, now we’ve set the table. Now we’ve put something together that has immense value on its own merit of diversification without any sort of prediction. But it also means that when you decide to tilt a portfolio away from maximum diversification, you can’t just look at, what’s my expected return of this overweight. Now, you have to recognize that the moment you deviate, you’re also giving up some rebalancing premium.

Mike:  00:19:09Not only that, because you have moved away from the maximally diversified portfolio, you’ve introduced volatility drag to the existing portfolio, which the bet not only has to overcome, it has to have a return that’s in excess of that and contribute to the portfolio more than that because you’ve got a volatility drag on this less diversified portfolio. It’s a high hurdle.

Rodrigo:00:19:30That’s right. And so here we are, this is a new challenge. Now you have to make sure that your predictive abilities, your tilts are really going to overcome all of that before you go ahead and tilt away from that.

Adam:00:19:44Yeah. I guess we’re going to start talking about tilts, are we?

Rodrigo:00:19:46We are.

Mike:00:19:47I’m tilted. Cue the music.