WEBVTT - How to Build a Portfolio That Outperforms For a Century 0:00:11.360 --> 0:00:14.680 Hello, and welcome to another episode of the All Thoughts podcast. 0:00:14.800 --> 0:00:18.520 I'm Tracy Alloway. My co host Joe Wisenthal couldn't make 0:00:18.560 --> 0:00:22.480 it today, So I'm looking at a chart of the 0:00:22.680 --> 0:00:28.200 SMP five hundred, and let's see, it's almost exactly one 0:00:28.320 --> 0:00:33.199 year post the big dip in the SMP five hundred, 0:00:33.760 --> 0:00:37.080 the massive crash that we saw back in March, and 0:00:37.120 --> 0:00:42.680 of course we've seen risk assets come roaring back. But 0:00:43.640 --> 0:00:46.559 with the recovery in financial assets, we have a lot 0:00:46.600 --> 0:00:50.479 of questions over how long can this continue? Are we 0:00:50.640 --> 0:00:53.320 on the verge of some sort of big change in 0:00:53.360 --> 0:00:58.560 the market as central banks unleash fiscal stimulus and interest 0:00:58.640 --> 0:01:03.000 rates remain at or close to the zero bound. Are 0:01:03.040 --> 0:01:06.240 we going to get inflation? Are we going to get stagflation? 0:01:06.680 --> 0:01:09.520 Are we going to see perhaps even deflation? Is there 0:01:09.560 --> 0:01:11.679 any way we're ever going to get out of this 0:01:12.200 --> 0:01:15.760 regime of low inflation? And I think whenever we have 0:01:15.959 --> 0:01:18.959 these big turning points in markets, or when we have 0:01:19.080 --> 0:01:21.639 lots of people talking about the potential for big turning 0:01:21.680 --> 0:01:25.200 points and markets, we also have a lot of opinions 0:01:25.240 --> 0:01:28.080 on how those are going to go and what type 0:01:28.120 --> 0:01:32.200 of portfolio would perform best, And of course we've seen 0:01:32.200 --> 0:01:36.640 a lot of handwringing recently around growth versus value sixty 0:01:36.760 --> 0:01:40.319 forty what happens when both bonds and stocks fall at 0:01:40.360 --> 0:01:44.360 the same time. It's on today's show we're going to 0:01:44.400 --> 0:01:47.840 be discussing exactly this idea. How do you build a 0:01:47.880 --> 0:01:52.919 portfolio that can withstand these regime changes and basically outperform 0:01:53.120 --> 0:01:57.520 over a really long time horizon. And I'm excited to 0:01:57.520 --> 0:02:01.200 say our guest for this episode is Chris Cole, the 0:02:01.280 --> 0:02:05.120 founder of Artemis Capital. He's appeared on all lots previously, 0:02:05.640 --> 0:02:10.520 digging deep into the low volatility regime of the past years. 0:02:10.800 --> 0:02:15.600 It's great to have him back. He publishes some excellent research. Chris, 0:02:15.600 --> 0:02:18.840 thanks for coming on the show again. Thanks Tracy, it's 0:02:18.840 --> 0:02:21.880 great to be back on the show. So, uh, I 0:02:21.919 --> 0:02:24.799 guess my first question is you know the reason we're 0:02:24.800 --> 0:02:27.960 having this discussion is you published a paper called the 0:02:28.040 --> 0:02:31.840 Allegory of the Hawk and Serpent, How to grow and 0:02:31.880 --> 0:02:36.639 protect wealth for a hundred years now. Most investors aren't 0:02:36.720 --> 0:02:40.200 really looking at their portfolios on a hundred year basis. 0:02:41.000 --> 0:02:44.120 What sparked your interest in that kind of time frame? 0:02:45.240 --> 0:02:49.280 What's really interesting about in particular the last forty years, 0:02:50.240 --> 0:02:55.840 is that there's a tremendous recency bias that market participants have. 0:02:56.800 --> 0:03:03.000 The last forty years are incredibly unusual comparative to overall history, 0:03:03.639 --> 0:03:06.239 and I believe, as I make the case in the paper, 0:03:06.720 --> 0:03:11.800 that recency biases is now a systemic risk. The price 0:03:11.840 --> 0:03:15.640 appreciation for a classic sixty forty portfolio over the last 0:03:15.760 --> 0:03:18.399 ninety three years has come from just the twenty two 0:03:18.480 --> 0:03:22.560 years between night four to two thousand seven. So what 0:03:22.639 --> 0:03:26.880 we've had is this incredible out performance of both stocks 0:03:26.919 --> 0:03:30.600 and bonds that has been from a reinforcing cycle, and 0:03:30.639 --> 0:03:33.320 we use the allegory of the serpent for that, and 0:03:33.360 --> 0:03:36.119 that's been led by falling interest rates. Rates fell from 0:03:36.160 --> 0:03:39.960 seventeen percent to zero percent. There's been incredible favorable demographics 0:03:39.960 --> 0:03:43.160 as baby boomers to come into the workforce, falling taxes. 0:03:43.200 --> 0:03:46.720 Taxes have fallen to near a hundred year lows, globalization 0:03:46.800 --> 0:03:49.960 on precedent of monetary policy, and we now have some 0:03:50.000 --> 0:03:53.120 of the highest levels of both governments and corporate debt 0:03:53.320 --> 0:03:56.680 in American history. So as a result of this, there's 0:03:56.720 --> 0:04:00.120 been this incredible out performance of stocks and bonds over 0:04:00.120 --> 0:04:04.720 the last forty years. But the trillion dollar question that 0:04:04.800 --> 0:04:09.200 I think is important for any allocator is is this repeatable? 0:04:10.120 --> 0:04:13.320 And I actually believe that the factors that drove this 0:04:13.520 --> 0:04:17.080 generational boom in the stocks and bonds are are now reversing. 0:04:17.680 --> 0:04:21.000 We're now on a framework where debts at all time highs, 0:04:21.400 --> 0:04:24.359 the middle class hasn't seen real wage growth since the 0:04:24.400 --> 0:04:28.400 nineteen seventies, and demographics are really poor, and interest rates 0:04:28.440 --> 0:04:32.640 can't go any lower. So investors expecting the games in 0:04:32.640 --> 0:04:36.240 the last four years are likely to be extremely disappointed. 0:04:36.600 --> 0:04:40.479 And so to kind of understand what the next twenty 0:04:40.560 --> 0:04:42.560 years are going to look like and how to build 0:04:42.560 --> 0:04:46.800 a portfolio that will last um and can manage through 0:04:46.839 --> 0:04:49.400 this period of secular change, what I did is I 0:04:49.440 --> 0:04:54.360 went back through history and I recreated all of these 0:04:54.600 --> 0:04:59.160 financial engineering strategies, UH using I think defensible assumptions over 0:04:59.200 --> 0:05:02.360 the last years. And we know that you know, history 0:05:02.400 --> 0:05:06.359 doesn't repeat, but it rhymes. And my my goal was 0:05:06.400 --> 0:05:11.359 to figure out what portfolio can consistently perform to every 0:05:11.400 --> 0:05:14.440 market cycle, whether it's secular growth, whether it's inflation, whether 0:05:14.440 --> 0:05:16.599 it's deflation. And I think I came up with some 0:05:16.680 --> 0:05:19.479 really interesting answers in my paper last year in the 0:05:19.480 --> 0:05:22.640 follow up paper that I just released that really answers 0:05:22.640 --> 0:05:26.600 the questions as to what type of portfolio really sustains 0:05:26.680 --> 0:05:30.719 wealth and capital appreciation and limits draw downs. And the 0:05:30.760 --> 0:05:33.240 answer that I got is radically different from the type 0:05:33.279 --> 0:05:36.919 of portfolio that many institutions and retail investors are currently 0:05:37.040 --> 0:05:41.760 are currently allocating too. I have so many questions already, 0:05:41.920 --> 0:05:45.160 but one that jumps out at me is going back 0:05:45.200 --> 0:05:49.320 a hundred years and trying to reconstruct modern portfolios against 0:05:49.480 --> 0:05:52.720 financial assets in the nine twenties and nineteen thirties. How 0:05:52.760 --> 0:05:55.920 exactly did you do that? Because I'm looking at the paper, 0:05:55.960 --> 0:05:58.359 and you know, you look at things like naked call selling, 0:05:59.080 --> 0:06:02.760 how that perform during the Great Depression? I'm just curious 0:06:02.800 --> 0:06:08.360 how you recreated, uh, those portfolios. Absolutely, it's a It's 0:06:08.360 --> 0:06:11.480 an incredible question. And I think one of the one 0:06:11.480 --> 0:06:13.200 of the things that I think is very important to 0:06:13.240 --> 0:06:18.080 understand about about this exercise is that we don't necessarily 0:06:18.080 --> 0:06:20.920 say that these portfolios are realized performance, but they are 0:06:20.960 --> 0:06:25.520 best effort at understanding how a given financial engineering strategy 0:06:25.600 --> 0:06:27.320 might have performed in the past. And I think there's 0:06:27.320 --> 0:06:30.200 a lot of defensible assumptions you know, first of all, 0:06:30.279 --> 0:06:33.400 we start with a wealth of observable data. There is 0:06:33.520 --> 0:06:37.640 data that we can gather off from history, namely the 0:06:37.800 --> 0:06:41.400 composite SMP five hundred prices, from stock data the top 0:06:42.200 --> 0:06:45.560 companies at a given point in time. There's obviously gold data, 0:06:45.600 --> 0:06:49.680 interest rate data, there's data on commodities and that that's 0:06:49.680 --> 0:06:51.800 a starting point um and we use that data from 0:06:51.800 --> 0:06:55.720 the global financial database. From there, what we can do 0:06:55.920 --> 0:06:58.960 is we can construct some UH and replicate some basic 0:06:59.120 --> 0:07:03.960 strategies like risk parity volatility targeting UH sixty forty portfolios. 0:07:04.000 --> 0:07:08.120 Those are relatively easy to replicate using that that kind 0:07:08.160 --> 0:07:12.840 of base level data. Now, Artemis is a long volatility 0:07:12.880 --> 0:07:17.120 trading shop. We what we do is we we provide volatility, 0:07:17.240 --> 0:07:21.320 long volatility and defensive solutions for our investors. So one 0:07:21.360 --> 0:07:24.200 of the most type of popular strategies that has been 0:07:24.240 --> 0:07:28.520 employed by many institutions are volatility overwriting strategies, either for 0:07:28.600 --> 0:07:31.840 income or defensive purposes. And what we wanted to do 0:07:31.960 --> 0:07:36.440 is to test these going back ninety three years. Of course, 0:07:36.800 --> 0:07:41.200 how do you test a volatility strategy prior to the 0:07:41.200 --> 0:07:45.240 existence of the options market, And that's that's not necessarily 0:07:45.320 --> 0:07:48.240 something that's easy to do. So obviously you have to 0:07:48.320 --> 0:07:50.560 make some assumptions and it becomes a bit of an 0:07:50.560 --> 0:07:54.360 intellectual exercise, but I think the assumptions are very defensible 0:07:54.560 --> 0:07:58.520 if one understands them. What we first did is we 0:07:58.520 --> 0:08:04.080 we took options data that exists from the President. We're 0:08:04.120 --> 0:08:08.720 able to solve for a volatility service. Volatility surface describes 0:08:08.800 --> 0:08:12.440 the pricing of volatility at various out of the money points, 0:08:12.680 --> 0:08:17.000 both for calls inputs. So that's the realized data, the 0:08:17.040 --> 0:08:20.000 real data that we have. Now we don't have that 0:08:20.160 --> 0:08:23.600 data going back obviously in the nineteen thirties or the 0:08:23.640 --> 0:08:28.960 nineteen seventies, but we do have data on how equity 0:08:29.000 --> 0:08:33.960 markets performed. We can calculate, for example, realized volatility on 0:08:34.120 --> 0:08:37.760 ten thirty sixty day time frames. We can calculate the 0:08:37.880 --> 0:08:42.200 rolling performance and next draw downs of equity markets over 0:08:42.240 --> 0:08:47.160 those times, so those are observable inputs. So using our 0:08:47.480 --> 0:08:52.000 arbitrage SBI ball surface that we solve for from real data, 0:08:52.320 --> 0:08:55.160 we then were able to run a multi multi variable 0:08:55.280 --> 0:09:02.040 regression to actually look at the last years and fit 0:09:02.400 --> 0:09:07.160 a volatility surface based on observable market data. And then 0:09:07.200 --> 0:09:11.439 what we did is we used that fitted data to 0:09:12.200 --> 0:09:18.120 an essence, run various option trading strategies using a theoretical 0:09:18.720 --> 0:09:22.840 volatility fit, and then we looked and compared that to 0:09:23.120 --> 0:09:26.640 some of the most popular CBO vol indices, like the 0:09:26.800 --> 0:09:29.840 buy right index and the put right index. And what 0:09:29.880 --> 0:09:31.920 we were able to do is we were able to 0:09:31.960 --> 0:09:39.480 replicate those indices using our theoretical volatility indications to generally 0:09:39.520 --> 0:09:43.400 over a point eight five correlation and almost exacting performance. 0:09:44.440 --> 0:09:46.800 So from there, what we're able to do now that 0:09:46.840 --> 0:09:49.600 we have this kind of in sample history, we were 0:09:49.600 --> 0:09:54.560 then able to apply that methodology to create a theoretical 0:09:56.080 --> 0:10:02.240 volatility imply volatility service going all the way back. Now, 0:10:02.480 --> 0:10:07.679 there are limitations in this because what you're naturally assuming 0:10:08.520 --> 0:10:14.000 is that the way market participants price volatility over the 0:10:14.080 --> 0:10:20.400 last thirty years would be very similar, would would be 0:10:20.559 --> 0:10:22.400 similar to the way they would price it in the 0:10:22.480 --> 0:10:25.720 thirties and the seventies and the fifties. Now we don't 0:10:25.760 --> 0:10:28.520 know that for certain, right we don't. We're willing to 0:10:28.600 --> 0:10:33.080 make that assumption to an essence, give give ourselves a 0:10:33.160 --> 0:10:38.439 fairly realistic assessment on how strategies like UH put writing 0:10:38.760 --> 0:10:42.720 or strategies like naked call selling would perform during those 0:10:42.720 --> 0:10:47.560 time periods, and we received some very interesting results that 0:10:47.800 --> 0:10:53.920 really give allocators a a sense on how these strategies 0:10:53.920 --> 0:10:58.600 would perform outside of the incredible regime of the last 0:10:58.640 --> 0:11:05.640 forty years. So you replicate the volatility performance from you know, 0:11:06.200 --> 0:11:08.760 over a hundred years ago or ninety three years ago, 0:11:08.800 --> 0:11:12.640 as you mentioned, you find that a technique that a 0:11:12.679 --> 0:11:16.840 lot of investors who have been using in recent years 0:11:16.880 --> 0:11:22.640 to pump up returns um again going short volatility doesn't 0:11:22.679 --> 0:11:25.880 perform as well, or hasn't performed as well way back then, Like, 0:11:25.960 --> 0:11:29.839 why exactly did that happen? What were the market conditions 0:11:29.960 --> 0:11:33.840 in place that you think allowed for the under performance 0:11:33.840 --> 0:11:38.280 of short fall. Well, you know, obviously over the many years, 0:11:38.280 --> 0:11:41.360 I've been a critic of short volatility and that strategy, 0:11:41.400 --> 0:11:43.839 and I also believe that long volatility is one of 0:11:43.880 --> 0:11:46.640 the one of the most under allocated assets that is 0:11:46.679 --> 0:11:49.360 out there today, and I think this paper and some 0:11:49.440 --> 0:11:53.679 more analysis that we provided give support to that. So 0:11:54.040 --> 0:11:56.959 I think saying short volatility strategies and what I mean 0:11:57.000 --> 0:12:00.120 by that these are strategies that sell put options, or 0:12:00.120 --> 0:12:04.280 maybe they sell call options, might be call overwriting by 0:12:04.360 --> 0:12:08.959 right programs. These are strategies that that institutions have employed 0:12:09.400 --> 0:12:12.679 in many ways to generate excess yield. Over the last 0:12:12.760 --> 0:12:16.920 four years, they performed exceptionally well, largely because we've been 0:12:16.960 --> 0:12:21.520 in an environment that has emphasized stability. Every single time 0:12:21.520 --> 0:12:25.520 equity markets draw down, central banks are able to respond, 0:12:25.920 --> 0:12:30.520 and that has produced an extremely mean reverting environment. That's 0:12:30.520 --> 0:12:33.000 not always been the case. And let me kind of 0:12:33.040 --> 0:12:36.120 explain why. Let's look at a period like the nine 0:12:37.679 --> 0:12:42.240 over the entire decade of the nineteen thirties, volatility realized 0:12:42.320 --> 0:12:48.559 at about that's incredible. So you know, Vaul was realizing 0:12:49.360 --> 0:12:54.720 two thousand eight. Imagine two thousand eight for an entire decade. Obviously, 0:12:54.760 --> 0:13:01.199 that does terrible things. Two portfolio that's continuously selling options meality. 0:13:01.240 --> 0:13:03.640 But one of the surprise takeaways and I think people 0:13:03.640 --> 0:13:07.120 would would appreciate appreciate this after the last year. You know, 0:13:07.160 --> 0:13:10.400 like the original paper came out before the COVID crisis 0:13:10.760 --> 0:13:12.840 and in many ways kind of predicted a lot of 0:13:12.840 --> 0:13:15.760 the problems that we're experienced in the COVID crisis. Uh 0:13:15.800 --> 0:13:19.720 in the in the reflation afterwards, Well naked call selling 0:13:20.000 --> 0:13:24.640 was among the worst strategies that we looked at you 0:13:24.640 --> 0:13:27.600 would think naked put selling would be or put put 0:13:27.600 --> 0:13:31.160 writing would be the worst strategy. Naked call selling was terrible. 0:13:31.760 --> 0:13:35.480 Why was naked call selling so bad? Well, if we 0:13:35.679 --> 0:13:39.400 go back to the Great Depression, you have these incredible 0:13:39.600 --> 0:13:44.920 drawdowns and equity markets. Central banks responded either by cutting rates, 0:13:45.520 --> 0:13:50.640 by implementing programs, or by devaluing goal and you had 0:13:50.640 --> 0:13:57.440 equally insane rallies in the market that were as violent 0:13:58.320 --> 0:14:03.200 as the draw downs. So that that's something I think 0:14:03.240 --> 0:14:07.120 that's so important for people to understand and really was 0:14:07.240 --> 0:14:11.320 foreshadowing the We have this huge drop in March two 0:14:11.360 --> 0:14:14.680 thousand twenty, we had this big explosion in April. Well, 0:14:14.760 --> 0:14:17.080 if we look at the Great Depression, you know, after 0:14:17.120 --> 0:14:22.760 a brutal three year decline of the market, rallied seventy 0:14:22.800 --> 0:14:27.600 two percent in just one point five months in ninety two, 0:14:28.080 --> 0:14:29.960 and that was after the signing of the Banking Act. 0:14:30.840 --> 0:14:35.400 In nineteen thirty three, the market had an eight percent 0:14:35.520 --> 0:14:40.160 rebound in just four point five months after Roosevelt devalued 0:14:40.200 --> 0:14:44.080 the dollar. So you have these violent rallies that occur 0:14:44.560 --> 0:14:49.240 during these periods of deflationary crises, and if you're selling 0:14:49.360 --> 0:14:55.560 call options into those violent rallies. You can clearly understand 0:14:55.560 --> 0:14:59.800 how bad that is. If you're doing covered call over writing, 0:15:00.000 --> 0:15:03.080 it's clear how bad that is. Another period that was 0:15:03.120 --> 0:15:07.080 really violent was the seventies where you had this kind 0:15:07.080 --> 0:15:10.560 of right tail skew realization. Now, what does that mean. 0:15:11.320 --> 0:15:18.240 Prior to the devaluation of gold in markets, markets used 0:15:18.320 --> 0:15:23.080 to trend. Equity markets used to trend. They were auto correlated. 0:15:23.120 --> 0:15:26.480 And what that meant is that if yesterday was up, 0:15:26.560 --> 0:15:29.040 today it was likely to be up, and the next 0:15:29.120 --> 0:15:32.080 day was likely to be up. So we reached this 0:15:32.200 --> 0:15:37.000 kind of secular peak in trending of equity markets in 0:15:37.000 --> 0:15:42.240 the nine seventies, and after the devaluation of gold, which 0:15:42.280 --> 0:15:47.880 empowered central banks to be reactive, we began a multi 0:15:48.080 --> 0:15:54.440 decade period where mean reversion ruled and last year represented 0:15:55.280 --> 0:15:59.320 the highest peak in mean reversion. Another way of saying 0:15:59.360 --> 0:16:02.280 that is nega voutle correlation. But really it's like if 0:16:02.360 --> 0:16:04.360 yesterday was up, today was likely to be down, and 0:16:04.440 --> 0:16:08.840 vice versa. The mean reversion and markets reached all time 0:16:08.920 --> 0:16:12.640 highs in over a hundred years of history in trending markets. 0:16:12.680 --> 0:16:17.920 That's really bad for short vault selling strategies because volatility 0:16:18.000 --> 0:16:21.520 is comprised of well, naturally vall. You have the vega, 0:16:21.800 --> 0:16:25.800 that's the volatility. But there's another component to options, which 0:16:25.840 --> 0:16:28.440 is the gamma. In another way of saying, gammas trend. 0:16:29.360 --> 0:16:33.200 So in for the greater part of seventy years, option 0:16:33.320 --> 0:16:37.640 buyers would have profited from trend. But over the last 0:16:37.720 --> 0:16:41.960 forty years, mean reversion has ruled, and that's largely been 0:16:42.160 --> 0:16:45.120 been connected to the decline and interest rates and the 0:16:45.200 --> 0:16:49.240 proactivity of central banks. So that's another reason why some 0:16:49.320 --> 0:16:55.480 of these short volatility strategies dramatically underperformed. And we're incredible, uh, 0:16:55.560 --> 0:16:58.680 I mean not only underperformed, I mean resulted in complete 0:16:59.040 --> 0:17:18.720 and cataclismic loss of capital for about seventy years. I'd 0:17:18.720 --> 0:17:21.280 love to get your take on how the volatility strategy 0:17:21.359 --> 0:17:25.520 is actually performed last year, like in March, and whether 0:17:25.640 --> 0:17:28.879 or not we've seen them build back up in the 0:17:28.960 --> 0:17:31.320 months since, because I think that might help us get 0:17:31.320 --> 0:17:33.800 a sense of like the direction that we're going in now. 0:17:34.320 --> 0:17:37.359 There's a question as to what portfolio is most robust 0:17:37.840 --> 0:17:42.080 and how do you build a robust portfolio. And one 0:17:42.080 --> 0:17:45.520 of the conclusions that we had doing this ninety years 0:17:45.520 --> 0:17:49.800 study of history was that to achieve a portfolio that 0:17:49.920 --> 0:17:53.240 is optimal, what investors should do is that they should 0:17:53.240 --> 0:18:00.600 prioritize long term correlations between asset classes over access returns. 0:18:01.520 --> 0:18:06.440 And so we devised a portfolio that's radically different than 0:18:06.680 --> 0:18:11.360 what many institutional portfolios have. That what this portfolio does 0:18:11.640 --> 0:18:16.000 is it diversifies assets based on market regimes. And by 0:18:16.040 --> 0:18:19.960 market regime, I mean regimes like inflation, deflation, and growth. 0:18:20.560 --> 0:18:25.080 It diversifies assets based on market regime rather than asset classes. 0:18:25.640 --> 0:18:30.440 So is an asset class diversification tool. Or the other 0:18:30.480 --> 0:18:34.760 diversification diversification tool used by many investors is trailing volatility 0:18:34.760 --> 0:18:38.160 and correlations. That's what's used by the strategies like risk Perry. 0:18:38.480 --> 0:18:42.920 So the portfolio that we really recommended to perform consistently 0:18:43.680 --> 0:18:51.000 over over ninety three years is obviously about about equity, 0:18:51.920 --> 0:18:56.040 about high quality bonds, and this is where it gets 0:18:56.040 --> 0:19:01.840 interesting approximately gold and precious metals. And I'll call that 0:19:01.880 --> 0:19:04.840 feat alternative. Some people might actually put crypto in that today. 0:19:05.000 --> 0:19:10.040 Obviously you can't test crypto going back trend and momentum strategies. 0:19:10.080 --> 0:19:13.080 These would be managed future strategies that profit off trends 0:19:13.080 --> 0:19:19.320 and commodities or currencies, and then finally long volatility and 0:19:19.359 --> 0:19:25.200 defensive hedgie. What ends up happening in these different asset classes. Obviously, 0:19:26.080 --> 0:19:31.200 obviously equities performed during periods of secular growth. Fixed income 0:19:31.359 --> 0:19:37.919 performs during periods of relatively stable inflation and UH and deflation. 0:19:38.359 --> 0:19:40.840 But you have a limit on fixed income at zero 0:19:40.880 --> 0:19:43.359 bound obviously, and that that that's occurred in the history 0:19:43.359 --> 0:19:47.520 before it happened kind of in the nine Now, long 0:19:47.600 --> 0:19:53.400 volatility and trend and momentum perform in periods of deflation, 0:19:54.080 --> 0:19:58.280 tremendous deflation, and tremendous inflation, So deflation like the thirties, 0:19:58.320 --> 0:20:02.360 inflation like the seventies, and obviously fat alternatives like precious 0:20:02.400 --> 0:20:05.439 metal perform in periods like the nineteen seventies where you 0:20:05.440 --> 0:20:09.879 have tremendous stagflation and negative real rates. So you have 0:20:10.160 --> 0:20:14.080 you're diversifying based on these market regimes. Now, we published 0:20:14.080 --> 0:20:16.680 this paper at the beginning of last year. We didn't 0:20:16.680 --> 0:20:19.680 have the opportunity to see in the future about what 0:20:19.720 --> 0:20:22.800 would happen throughout two thousand twenty. I released a new 0:20:22.840 --> 0:20:26.640 paper that talked about how this recommended portfolio performed through 0:20:26.680 --> 0:20:30.840 two thousand twenty. It was exceptional performance, because, as you know, 0:20:31.040 --> 0:20:35.320 two thousand twenty was like an entire business cycle condensed 0:20:35.359 --> 0:20:40.800 into twelve months. The whole business cycle from January to 0:20:41.000 --> 0:20:45.920 March that was like a nineteen thirties deflation. Then from 0:20:46.000 --> 0:20:50.920 April to about August, we had this incredible fiat devaluation 0:20:51.040 --> 0:20:55.440 with ten trillion dollars in global stimulus and this speculative 0:20:55.440 --> 0:20:59.280 asset growth. That's that's uh, you know, some almost kind 0:20:59.280 --> 0:21:04.520 of similar to type of scenario. As we entered the 0:21:04.520 --> 0:21:09.119 the wintertime, we went into a reflationary scenario that began 0:21:09.160 --> 0:21:12.440 to resemble kind of the onset of stagflation in the 0:21:12.480 --> 0:21:15.600 late sixties. I mean that's where you have struggles and 0:21:15.680 --> 0:21:19.040 interest rates. Interest rates began to rise at perform, but 0:21:19.119 --> 0:21:22.240 then you have this expectation of a reset deepening of 0:21:22.240 --> 0:21:25.719 the yield curve, and commodities began to perform. I mean, 0:21:25.840 --> 0:21:30.240 lumber reached all time highs copper was exploding. So if 0:21:30.240 --> 0:21:33.399 we look at how this portfolio that we call the 0:21:33.440 --> 0:21:39.000 Dragon portfolio formed in that deflationary period in the first 0:21:39.000 --> 0:21:43.399 part of the year, well, long volatility strategies were the 0:21:43.480 --> 0:21:49.639 huge winners. They actually we're able to fill the gap 0:21:50.520 --> 0:21:55.480 where equities drew down. So actually for this portfolio, that 0:21:55.600 --> 0:22:03.280 first quarter gained about where sixty portfolios and risk parity 0:22:03.320 --> 0:22:08.040 portfolios had huge underperformance because they were overreliant on bonds 0:22:08.920 --> 0:22:12.760 as their defensive protection. So long volatility strategies is so 0:22:12.840 --> 0:22:16.359 well and protected and helped you make money during that quarter. 0:22:17.080 --> 0:22:19.679 Then during the FIATA valuation and the kind of growth 0:22:19.720 --> 0:22:23.040 period after Central Bank stepped in, gold took its turn 0:22:23.119 --> 0:22:26.440 performing in the summer, along with definitely crypto and other 0:22:26.480 --> 0:22:29.200 assets like that, and then we had a huge outperformance 0:22:29.200 --> 0:22:33.480 and equities. Then by the fall, we began to see 0:22:33.520 --> 0:22:36.720 gold began to sell off, and we began to see 0:22:37.280 --> 0:22:40.880 equities continued their upward trajectory. Fixed income began to sell off, 0:22:41.400 --> 0:22:46.400 but trending commodities managed futures began to outperform, profiting from 0:22:46.400 --> 0:22:49.760 the trends in commodities markets and the trends in these 0:22:49.760 --> 0:22:53.879 other asset classes. So we had an entire business cycle 0:22:54.080 --> 0:22:59.800 condensed into twelve months, and at each point these assets, 0:23:00.080 --> 0:23:04.399 these market regime diversification benefits of the portfolio became really clear. 0:23:05.280 --> 0:23:09.000 And this concept of this dragon portfolio that we introduced 0:23:09.480 --> 0:23:13.399 actually would have returned close to last year where the 0:23:14.280 --> 0:23:20.040 portfolio and risk parity portfolios only returned about on average 0:23:20.400 --> 0:23:24.840 with over three times the max draw down. For people 0:23:24.840 --> 0:23:27.960 who read our paper, this should not be a surprise. 0:23:28.760 --> 0:23:31.080 It should not be a surprise because we saw this 0:23:31.200 --> 0:23:34.800 happen in previous market cycles. But you have to look 0:23:34.840 --> 0:23:39.200 at the periods outside the last forty years to understand that. 0:23:40.000 --> 0:23:43.000 So I get the contention that if you're building a 0:23:43.080 --> 0:23:47.000 portfolio that's going to outperform in these big regime changes, 0:23:47.040 --> 0:23:51.159 then in when you basically had a compressed business cycle, 0:23:51.440 --> 0:23:56.680 as you mentioned, it would do phenomenally well. But most 0:23:56.720 --> 0:24:00.719 people would agree that was an extremely unusual year, and 0:24:00.760 --> 0:24:05.520 I think most investment professionals are probably more used to 0:24:05.600 --> 0:24:10.880 trying to pinpoint the regime changes as they come, rather 0:24:10.960 --> 0:24:14.880 than build a portfolio that's going to outperform all types 0:24:14.920 --> 0:24:18.919 of regimes for a hundred years. Right Like, there outlook 0:24:19.000 --> 0:24:21.880 is always going to be you know, ten, twenty, maybe 0:24:21.920 --> 0:24:24.000 thirty or forty years, but it's going to be much 0:24:24.160 --> 0:24:28.080 much shorter than what you're talking about. So I guess 0:24:28.119 --> 0:24:32.200 my question is like, how useful is this for your 0:24:32.240 --> 0:24:38.159 average financial advisor or investor, and how do you encourage 0:24:38.200 --> 0:24:41.560 people to think on a longer term horizon, or think 0:24:41.560 --> 0:24:45.159 about diversification across regime changes rather than just trying to 0:24:45.200 --> 0:24:50.400 pinpoint when a particular change is taking place in markets. Yeah, 0:24:50.440 --> 0:24:52.240 I think one of the one of the phrases that 0:24:52.280 --> 0:24:56.880 we say is do not fear, do not predict, prepare. 0:24:57.920 --> 0:25:01.920 So if you are able to perfectly time and predict 0:25:03.240 --> 0:25:07.800 the regime changes, and I tell you I can't, you know, 0:25:08.280 --> 0:25:11.280 but if you're able to do that, you're Stanley Druckon Miller, 0:25:11.440 --> 0:25:16.199 You're George Soros, and you should be a billionaire, right. Um, 0:25:16.320 --> 0:25:22.000 The average retail investor, the average institutional investor, is not 0:25:22.160 --> 0:25:26.479 able to necessarily time those changes perfectly. And even if 0:25:26.480 --> 0:25:28.439 they're able to do so, if you're managing a fifty 0:25:28.440 --> 0:25:31.400 billion dollar portfolio, if you're an institution, or a ten 0:25:31.440 --> 0:25:35.119 billion dollar portfolio. I you're an institution, it's difficult to 0:25:35.160 --> 0:25:38.960 tilt the portfolio. You have to choose some portfolio allocation. 0:25:39.760 --> 0:25:42.760 And you know, my point is that the average pension 0:25:42.800 --> 0:25:49.560 fund in America is approximately equity linked investments and approximately 0:25:51.080 --> 0:25:55.120 in kind of fixed income and alternatives in cash. What 0:25:55.280 --> 0:25:58.240 many of these institutions have done is they've crowded in 0:25:58.280 --> 0:26:02.560 two assets like private equity kind of pretending that they're diversifiers. 0:26:03.600 --> 0:26:05.560 But if you look at some of the Cambridge studies, 0:26:05.800 --> 0:26:09.000 private equity has tremendous correlation to the business cycle. It's 0:26:09.040 --> 0:26:12.560 not a diversifying asset by market regime. No, they've jumped 0:26:12.600 --> 0:26:15.520 into vcs' is the same thing you jump into real estate. 0:26:15.560 --> 0:26:18.600 It's the same problem. These are all asset classes that 0:26:18.640 --> 0:26:21.959 are correlated to business cycles. Their long GDP asset classes 0:26:22.000 --> 0:26:26.840 that are correlated to one growth regime. Now, in the past, 0:26:27.040 --> 0:26:31.320 you could rely on fixed income to provide diversification, but 0:26:31.960 --> 0:26:35.679 when fixed incomes at the zero bound, it fails to 0:26:35.720 --> 0:26:39.200 provide great diversification. And anyone who studies the nineties would 0:26:39.200 --> 0:26:42.480 have seen that problem. When we recreated risk parity portfolios 0:26:42.520 --> 0:26:45.439 which rely heavily on fix income and lever the fixed income. 0:26:45.840 --> 0:26:48.680 There's tremendous underperformance in the nineties when rates were near 0:26:48.720 --> 0:26:53.400 the zero bound. So these institutions are assuming the last 0:26:53.440 --> 0:26:57.720 forty years will repeat, and they're in essence levering equity linked, 0:26:57.760 --> 0:27:00.680 not levering, but they're they're relying on this expectation of 0:27:00.720 --> 0:27:04.600 equity link performance and these kind of false diversifiers to 0:27:04.680 --> 0:27:07.919 reach their seven point to five percent return targets. I 0:27:07.920 --> 0:27:10.800 think this is a huge mistake because if you remove 0:27:10.880 --> 0:27:13.600 the last forty years, their performance is closer to to 0:27:13.680 --> 0:27:18.040 five percent annualized abysmal, you know, using the allocations that 0:27:18.080 --> 0:27:21.439 they have. So if they're not able to mate that 0:27:21.520 --> 0:27:25.000 seven point five percent return target, you know, recency bias 0:27:25.080 --> 0:27:26.919 becomes a problem that all of us are going to 0:27:26.920 --> 0:27:29.840 pay for. And the reason being there's about one point 0:27:29.840 --> 0:27:33.880 four trillion dollar US state and local pension deficit right now, 0:27:34.080 --> 0:27:36.560 and that's assuming that they are able to hit their 0:27:36.560 --> 0:27:41.360 return targets. If there's under performance, as one would expect 0:27:41.400 --> 0:27:45.320 given current valuations and the current correlation mix of their investments, 0:27:46.160 --> 0:27:50.040 you can expect that deficit to rise to anywhere between 0:27:50.200 --> 0:27:52.520 three trillion dollars all the way to up to nine 0:27:52.560 --> 0:27:56.240 trillion dollars. So in the last stimulus bill, there's a 0:27:56.240 --> 0:28:00.720