WEBVTT - Robert Arnott Discusses the Process Behind Research Affiliates 0:00:02.240 --> 0:00:06.800 This is Master's in Business with Barry Ridholts on Bloomberg Radio. 0:00:09.680 --> 0:00:13.000 This week on the podcast, I have an extra special guest. 0:00:13.200 --> 0:00:15.800 His name is Rob are Not and he is the 0:00:15.920 --> 0:00:22.440 creator and chairman of Research Affiliates, a fundamental indexing firm 0:00:22.600 --> 0:00:27.720 that has strategies that manage over two hundred billion dollars 0:00:28.120 --> 0:00:33.360 in assets. Rob is perhaps best known for popularizing the 0:00:33.440 --> 0:00:39.600 concept of fundamental indexing UH, often called smart data, although 0:00:39.640 --> 0:00:44.440 he has uh takes issue with with that, um moniker. 0:00:44.479 --> 0:00:47.839 He thinks it's just become a marketing term. Uh. This 0:00:47.920 --> 0:00:51.000 is really a fascinating conversation. Every time I speak with Rob, 0:00:51.040 --> 0:00:54.680 I learned something new. Uh. This time I discovered that 0:00:54.800 --> 0:00:57.880 he has co authored a number of papers with the 0:00:57.920 --> 0:01:01.200 great Peter Bernstein, which was something I I didn't previously know. 0:01:02.000 --> 0:01:06.960 If you were at all interested in things like fundamental indexing, 0:01:07.120 --> 0:01:11.440 what's wrong with market cap waiting? How large that sector 0:01:11.520 --> 0:01:16.319 can grow? The advantages of factor based investing. And I 0:01:16.400 --> 0:01:19.560 also learned that there are now over five hundreds such factors. 0:01:19.600 --> 0:01:21.920 I had no idea. I knew it was hundreds, I 0:01:21.920 --> 0:01:24.840 had no idea was that many. Uh. This is really 0:01:24.880 --> 0:01:28.920 a masterclass in this space. So with no further ado. 0:01:29.440 --> 0:01:36.920 Here is my conversation with Rob or Not. My extra 0:01:36.959 --> 0:01:40.399 special guest this week is Rob are Not. He is 0:01:40.520 --> 0:01:44.640 the founder and chairman of Research Affiliates, perhaps best known 0:01:44.760 --> 0:01:49.320 for popularizing the concept of smart beta, better known or 0:01:49.320 --> 0:01:54.200 better described as fundamental indexing uh rafie does not consider 0:01:54.320 --> 0:02:01.400 market cap, focusing instead on four primary factors sales, profit, book, value, 0:02:01.440 --> 0:02:05.120 and dividend are not formed the firm in O two 0:02:05.160 --> 0:02:10.920 in Newport Beach, California. Investment strategies developed by Research Affiliates, 0:02:10.960 --> 0:02:14.360 or Raphy as it's better known, manages over two hundred 0:02:14.360 --> 0:02:18.280 billion dollars in assets. Rob are Not, Welcome back to Bloomberg. 0:02:18.680 --> 0:02:22.360 Thank you. So I mentioned you started Research Affiliates and 0:02:22.400 --> 0:02:25.360 I'm used to just calling it raffi is the nickname 0:02:25.400 --> 0:02:29.520 everybody uses. Uh back in two thousand two. Could you 0:02:29.800 --> 0:02:34.360 have imagined then that it's barely fifteen years later and 0:02:34.639 --> 0:02:37.720 your strategies are now managing over two hundred billion dollars? 0:02:37.720 --> 0:02:41.840 Did you have any expectation of how wildly successful smart 0:02:41.880 --> 0:02:45.960 data would become? Those are two different questions. Did I 0:02:46.080 --> 0:02:51.560 expect to be this successful? Yes, nobody starts a business 0:02:51.960 --> 0:02:58.000 without a healthy dose of optimism. Okay, perhaps sometimes undeserved, 0:02:58.200 --> 0:03:01.040 but but even still, that's a big chunk of change. 0:03:01.120 --> 0:03:05.679 Did you expect to hit those sort of numbers this quickly? Um, Well, 0:03:05.720 --> 0:03:08.880 I don't know how quick that is. Sixteen years is 0:03:08.919 --> 0:03:15.320 a long time. But as for RAFI the fundamental index, 0:03:16.080 --> 0:03:18.960 UH and it's growth. I couldn't have anticipated that because 0:03:18.960 --> 0:03:21.519 we hadn't come up with the idea yet. When when 0:03:21.520 --> 0:03:24.560 did that first roll out? We first thought about the 0:03:24.600 --> 0:03:29.480 idea of indexing in non capitalization ways in two thousand three. 0:03:30.320 --> 0:03:34.760 We formalized our research and UH completed development of the 0:03:34.800 --> 0:03:37.400 idea by mid O four. We were live with assets 0:03:37.440 --> 0:03:40.280 by the end of O four. First fund was launched 0:03:40.280 --> 0:03:44.800 by Pimco and OH five, first published index was by 0:03:44.840 --> 0:03:50.120 Footsie UM in November of oh five. So barely a 0:03:50.160 --> 0:03:53.720 dozen years. That's even faster. Scale up to not a 0:03:53.720 --> 0:03:56.800 lot of fun scale up to two and a dozen years. No, 0:03:56.920 --> 0:03:59.640 that's true, that's true, but keep in mind it's not 0:04:00.000 --> 0:04:02.560 a hundred billions of assets under management. We licensed the 0:04:02.600 --> 0:04:10.000 idea three distribution partners, so we have the help of Pimco, Schwab, Investco, Nomura, 0:04:11.080 --> 0:04:14.440 foot See and Russell UH and couldn't get any names 0:04:14.440 --> 0:04:18.920 that anybody recognized. We keep looking, We keep hoping somebody 0:04:19.160 --> 0:04:22.839 known actually signs onto the idea. So that let's let's 0:04:22.880 --> 0:04:26.520 digress and talk about that a moment. That's a fascinating 0:04:26.560 --> 0:04:31.400 model where now I know you guys also manage money directly, 0:04:31.560 --> 0:04:34.960 not anymore, not anymore. Oh so that's that's a change 0:04:34.960 --> 0:04:37.000 since the last time you were here. When did you 0:04:37.120 --> 0:04:43.400 stop directly managing money? We decided in two thousand fourteen 0:04:44.400 --> 0:04:51.440 that UM five of our assets under quote management close quote, 0:04:51.440 --> 0:04:54.760 where money was money that we actually managed. Fift pcent 0:04:54.800 --> 0:04:58.440 of our revenues came on that, and the tail was 0:04:58.480 --> 0:05:03.240 wagging the dog. We were winding up with distribution partners, 0:05:03.680 --> 0:05:06.200 fearful that we were going to compete directly with them. 0:05:06.320 --> 0:05:11.320 That makes sense, and we decided we don't need these complications. 0:05:11.400 --> 0:05:14.400 Let's stick to what we're best at. And so we 0:05:14.440 --> 0:05:19.800 approached Pimco, our largest affiliate, and in effect said would 0:05:19.839 --> 0:05:23.480 you like this eight billion dollar book of business UM 0:05:23.560 --> 0:05:27.000 And they said, they said, at a price tag of zero, 0:05:27.080 --> 0:05:29.760 that's not a bad deal. I think we'll go for you. 0:05:29.760 --> 0:05:30.800 You know, if you would have come to me, I 0:05:30.800 --> 0:05:32.400 would have I would have given you the same deal 0:05:32.400 --> 0:05:35.599 they gave you. Oh my goodness, that didn't cross my mind. 0:05:35.600 --> 0:05:37.960 I wish, I wish I had thought of that. So 0:05:37.960 --> 0:05:40.520 so this is kind of fascinating. So what I find 0:05:40.640 --> 0:05:45.200 so intriguing about your model in addition to the intellectual property, 0:05:45.680 --> 0:05:49.680 but essentially, you're in the business of licensing these indices 0:05:49.960 --> 0:05:52.760 to other people to do you know, the roll up 0:05:52.800 --> 0:05:55.920 your sleeves and do the trading, do the clients everything 0:05:56.000 --> 0:06:01.919 do which is complex and filled with risk and expensive. 0:06:02.040 --> 0:06:06.839 So you really have the sweetest part of of the pie. 0:06:07.240 --> 0:06:12.720 For us, it's the sweetest part because, uh, it's what 0:06:12.760 --> 0:06:14.960 we love to do, and it's what you you, I 0:06:15.200 --> 0:06:18.480 ostensibly would imagine you do best. Let the other people 0:06:18.520 --> 0:06:22.240 who are experts in trading and what have you. Um 0:06:22.279 --> 0:06:26.640 I think not being involved in trading back office call 0:06:26.800 --> 0:06:30.760 desks is a wonderful thing. Um I think there are 0:06:30.760 --> 0:06:34.479 people who are really good at that, and I want 0:06:34.560 --> 0:06:39.000 them to do what they're really good at. In effect, 0:06:39.560 --> 0:06:44.480 being in the business of product innovation and licensing, we 0:06:44.560 --> 0:06:50.000 face an unusually high hurdle because our clients are distribution partners. 0:06:50.440 --> 0:06:52.920 They don't need us, They have their own R and 0:06:53.000 --> 0:06:59.400 D teams and so for our business to succeed, UM, 0:06:59.640 --> 0:07:02.360 they have to see us as an extension of their 0:07:02.360 --> 0:07:05.080 own R and D team, complimentary to what they do, 0:07:05.960 --> 0:07:11.320 and UM, not all organizations can think that way. You 0:07:11.360 --> 0:07:14.240 have to be willing to go against the not invented 0:07:14.280 --> 0:07:17.840 here syndrome. So that sort of puts a little more 0:07:18.000 --> 0:07:22.320 heft on the name of the firm research affiliates, because 0:07:22.400 --> 0:07:29.360 essentially you provide intellectual um property and research strength to 0:07:29.480 --> 0:07:31.720 the affiliates you work with. I never really thought of 0:07:31.760 --> 0:07:36.000 it in quite those terms, but really that's what Raffie does. 0:07:36.360 --> 0:07:39.520 That's exactly right. Was that intentional or did we you 0:07:39.600 --> 0:07:42.560 just happen to stumble on that? Um, a little bit 0:07:42.600 --> 0:07:45.440 of both. I was running first Quadrant at the time 0:07:46.360 --> 0:07:48.960 and had decided if I'm ever going to start my 0:07:48.960 --> 0:07:52.120 own business, I had to do it sooner rather than later, 0:07:52.920 --> 0:07:57.960 and so I was running both in parallel for two years. 0:07:59.160 --> 0:08:02.560 Running both in parallel for two years meant I had 0:08:02.600 --> 0:08:06.080 to be studiously avoid conflicts of interest. So initially we 0:08:06.160 --> 0:08:13.040 started out saying, let's subadvise, let's license our ideas, and 0:08:13.160 --> 0:08:15.520 we found that that business model works, so we stuck 0:08:15.560 --> 0:08:20.000 with it. Quite fascinating. Let's talk a little bit about 0:08:20.360 --> 0:08:26.640 factors investing um Fama French famously identified first the three 0:08:26.680 --> 0:08:29.360 factor model, then the five factor model, then the seven 0:08:29.400 --> 0:08:33.040 factor model. Are there still factors out there that have 0:08:33.240 --> 0:08:38.800 yet to be discovered? Two answers to that question. Firstly, yes, 0:08:39.200 --> 0:08:43.520 there will be. They've already been five factors published five 0:08:44.640 --> 0:08:49.400 and um there will be hundreds more. Now the more 0:08:50.200 --> 0:08:53.000 pertinent question is how much of this is data mining 0:08:54.080 --> 0:08:57.280 finding relationships that prevailed in the past that have no 0:08:57.400 --> 0:09:00.520 reason to prevail in the future, And how much of 0:09:00.559 --> 0:09:07.440 it is um truly factors that drive price that can 0:09:07.480 --> 0:09:12.080 be truly a reliable source of access return, Meaning is 0:09:12.120 --> 0:09:19.280 there enough outperformance here that can be captured by a portfolio? Exactly? So? 0:09:19.440 --> 0:09:22.960 Of those five So if these were really good factors, 0:09:23.000 --> 0:09:25.840 why wouldn't we create a if not a five hundred 0:09:25.920 --> 0:09:29.319 factor portfolio. Hey, let's put these in order of the 0:09:29.400 --> 0:09:33.360 strongest out performance generators, and here are the fifty or 0:09:33.800 --> 0:09:37.440 five best factors. Why can't we do that? Well? I 0:09:37.440 --> 0:09:40.240 wrote a paper a couple of years ago called how 0:09:40.280 --> 0:09:43.720 can smart Bay to Go Horribly Wrong? And I remember 0:09:43.760 --> 0:09:49.160 that it was massively controversial. I thought the controversy was 0:09:49.440 --> 0:09:54.040 amusing because if I'd written a paper entitled how can 0:09:54.080 --> 0:09:59.120 Stockpicking go Horribly Wrong? And I offered the the sage 0:09:59.520 --> 0:10:03.880 inside that if a stock soars and its fundamentals haven't, 0:10:04.080 --> 0:10:09.120 so it's valuation multiples have sword, it's past returns will 0:10:09.160 --> 0:10:14.080 look artificially wonderful, and if there's any mean or version 0:10:14.080 --> 0:10:17.880 on valuations, it's future performance will, to use a technical term, 0:10:18.000 --> 0:10:24.040 suck so um. People would have read that paper and said, 0:10:24.040 --> 0:10:28.000 what's this nitwit talking about? Everybody knows that by saying 0:10:28.080 --> 0:10:34.200 exactly the same thing about factors and smart beta strategies, 0:10:35.600 --> 0:10:39.640 I was pilloried for suggesting that the same thing could 0:10:39.679 --> 0:10:44.240 apply for strategies. Now, if you go back historically, you 0:10:44.320 --> 0:10:48.880 find that the alpha of many of the smart beta 0:10:48.960 --> 0:10:55.439 factors have has not been tested in terms of how 0:10:55.520 --> 0:10:58.439 much of that access return came from rising valuation multiples. 0:10:58.720 --> 0:11:03.400 We might as well have a an Apple factor that 0:11:03.720 --> 0:11:09.960 simply says Apple has outperformed magnificently. It is a powerful factor. 0:11:10.600 --> 0:11:15.400 You have a long Apple, short everything else factor, and 0:11:15.840 --> 0:11:18.680 because of the past performance, we know it's going to 0:11:18.720 --> 0:11:22.240 continue to work. Let's digress a little bit and and 0:11:22.320 --> 0:11:26.480 define factors for people who may not be familiar with 0:11:26.640 --> 0:11:29.040 Gene Fama, who won the Noble Prize a few years 0:11:29.040 --> 0:11:32.600 ago for his for a lot of his work. UM 0:11:32.640 --> 0:11:35.320 the original and its Gene Farm in Ken French at 0:11:35.400 --> 0:11:40.439 Dartmouth Farm is a Chicago. The original Fama French factor 0:11:40.480 --> 0:11:45.520 paper was small capitalization value, and I'm trying to remember 0:11:45.600 --> 0:11:49.200 was that momentum, quality, market market beta market. So that's 0:11:49.240 --> 0:11:51.520 the three, okay, and then when we moved to five, 0:11:52.480 --> 0:11:57.920 we added um uh, quality and momentum. That's the next lot, 0:11:58.360 --> 0:12:00.400 and then seven. I don't even know what the next 0:12:00.440 --> 0:12:04.640 two are. I'm not certain, but I think it's illiquidity 0:12:04.640 --> 0:12:08.040 and investment. Right. So, so all of this comes back 0:12:08.040 --> 0:12:11.800 to let's let's keep it simple. Low cost stocks over 0:12:11.840 --> 0:12:15.320 long periods of time, low uh better value stocks, not 0:12:15.440 --> 0:12:19.080 expensive stocks. I don't mean low price tend to outperform 0:12:19.280 --> 0:12:24.440 expensive stocks over the long haul, exactly right now. Advocates 0:12:24.440 --> 0:12:27.120 of efficient markets will say it's got to be because 0:12:27.120 --> 0:12:29.360 of some kind of hidden risk, because you can't get 0:12:29.440 --> 0:12:33.320 something for nothing. Um. I would push back against that 0:12:33.400 --> 0:12:36.920 and say it's not risk with value. Might be risked 0:12:36.920 --> 0:12:39.920 with small cap because there and then there's a liquidity issues, 0:12:40.200 --> 0:12:44.360 but with value, it's the psychology it's the behavior of 0:12:44.800 --> 0:12:47.559 who wants to buy this. Dominoes is one of my 0:12:47.600 --> 0:12:51.040 favorite examples. Domino's had a big, a whole run of 0:12:51.040 --> 0:12:54.720 issues late nineties early two thousands. The stock did poorly. 0:12:54.920 --> 0:12:57.160 People were thinking, all right, well, that chain is pretty 0:12:57.240 --> 0:13:00.199 much done. And Dominoes over the past I think it's 0:13:00.200 --> 0:13:04.600 either fifteen or twenty years, has handily outperformed Apple. May 0:13:04.640 --> 0:13:08.600 be getting the timeframe wrong. So and that's the psychology 0:13:08.640 --> 0:13:11.120 of who wants to touch that? And if you look 0:13:11.240 --> 0:13:15.680 from the trough in two thousand nine, Uh City ever 0:13:15.800 --> 0:13:20.080 so briefly dipped below a buck, be of a dipped 0:13:20.120 --> 0:13:25.640 to roughly two bucks, and since then those have handily 0:13:25.640 --> 0:13:31.560 outperformed Apple. Um so, but they've outperformed it from a 0:13:31.640 --> 0:13:38.200 starting point of being thought on death's door. So value, 0:13:38.559 --> 0:13:43.280 it seems to me, does have a behavioral basis. Basically, 0:13:43.520 --> 0:13:46.840 when you have a value orientation, you're buying what's unloved, 0:13:47.200 --> 0:13:51.760 what people want to shun. That should be rewarded. Now 0:13:52.000 --> 0:13:56.000 is it a hidden risk factor only psychologically? Well, let 0:13:56.000 --> 0:13:57.880 me let me push back a little bit over there. 0:13:58.960 --> 0:14:02.920 Two companies in financial crisis look terrible. One of them's 0:14:02.960 --> 0:14:05.680 a I G. The other's Lehman brothers. You can buy 0:14:05.800 --> 0:14:10.319 a I g rescued and and since pretty decent, not great, 0:14:10.360 --> 0:14:13.640 but pretty decent returns off of the bottom and Lehman, 0:14:14.360 --> 0:14:16.360 I guess we could call that a value trap. Although 0:14:16.400 --> 0:14:18.760 there were elements of fraud and RepA one oh five, 0:14:19.240 --> 0:14:21.680 there was a whole different set of problems with Lehman. 0:14:21.800 --> 0:14:25.560 But the risk with value is am I buying something 0:14:25.600 --> 0:14:27.680 that's going to be a zero? That's known as a 0:14:27.760 --> 0:14:31.040 value trap. It's a stock that looks cheap on its 0:14:31.040 --> 0:14:33.680 way to zero. Now it's hard to have a whole 0:14:33.720 --> 0:14:36.240 sector that looks cheap on its way to zero. I 0:14:36.360 --> 0:14:40.680 coined the term anti bubble in two thousand nine to 0:14:40.920 --> 0:14:44.800 describe what I perceived as the inverse of a bubble, 0:14:44.800 --> 0:14:48.640 where an entire sector of the economy is priced as 0:14:48.680 --> 0:14:52.760 if they're all headed for oblivion, when in fact, every 0:14:52.800 --> 0:14:57.800 failure clears the runway for the survivors to have higher profits, 0:14:57.880 --> 0:15:01.320 higher margins, and greater success us ahead. So, unless you 0:15:01.400 --> 0:15:07.760 wanted to um except the notion of armageddon the end 0:15:07.800 --> 0:15:13.120 of the economy, uh you, it made sense to think 0:15:13.200 --> 0:15:17.880 that collectively this sector was being dismissed when it shouldn't be. 0:15:18.160 --> 0:15:21.360 There's a whole group of people that are the armageddon traders. 0:15:21.400 --> 0:15:25.040 I think the Ft called them the plastic bears. They'll 0:15:25.040 --> 0:15:28.120 scream arm again, but then they'll he here's what we 0:15:28.160 --> 0:15:31.360 can sell you while we're waiting for armagedon um. The 0:15:31.400 --> 0:15:35.240 other thing about anti bubble is so fascinating. You could 0:15:35.240 --> 0:15:38.800 say there's an anti bubble in what was it? Home 0:15:38.840 --> 0:15:41.840 builders and oh five and mortgage brokers and bankers and 0:15:41.920 --> 0:15:44.960 oh six and pretty much anything else in finance and 0:15:45.000 --> 0:15:50.000 O seven and each almost every there's almost always some 0:15:50.120 --> 0:15:53.480 sort of bubble in the market somewhere and some sort 0:15:53.520 --> 0:15:56.640 of anti bubble in the market somewhere. You go back 0:15:56.640 --> 0:16:01.520 to early two thousand sixteen emerging mark, It's deep value 0:16:02.160 --> 0:16:06.760 was priced at two to three times cash flow, and 0:16:07.000 --> 0:16:09.600 it's not as if the emerging economies of the world 0:16:09.640 --> 0:16:14.840 were collectively all going to go bust, so it represented 0:16:14.880 --> 0:16:18.760 an extraordinary opportunity. Rafi, the fundamental index in emerging markets 0:16:18.800 --> 0:16:22.320 was briefly trading for a Schiller Pe ratio below six. Wow, 0:16:22.360 --> 0:16:25.800 that's amazing. Let's talk about a study you did way 0:16:25.800 --> 0:16:30.640 back when looking at the SNP five index, which theoretically 0:16:30.880 --> 0:16:34.920 is a passive, and we'll put a footnote on exactly 0:16:34.960 --> 0:16:41.680 what cassive means. In this you found from to the 0:16:41.760 --> 0:16:44.280 last year there was a full year of data available, 0:16:44.800 --> 0:16:48.680 stocks added to the index underperformed those that were kicked 0:16:48.680 --> 0:16:53.400 out by an average of twenty three percentage points over 0:16:53.440 --> 0:16:56.960 the ensuing twelve months. That's not a bad gap in return. 0:16:57.120 --> 0:16:59.640 Oh my goodness, that's amazing. So so what are the 0:17:00.000 --> 0:17:03.640 occations of this and what does this mean about the 0:17:03.800 --> 0:17:08.320 stock picking acumen of the editors on the committee of 0:17:08.359 --> 0:17:11.320 the SMP Index who select the stocks that go into it. 0:17:11.760 --> 0:17:15.040 I don't fault the index committee for the way they 0:17:15.080 --> 0:17:18.520 make choices. They they are under. Their clients are the 0:17:18.600 --> 0:17:21.679 people who license the index, and the people who license 0:17:21.800 --> 0:17:25.720 the index want the index to include all the names 0:17:25.720 --> 0:17:28.639 that are hot, beloved, and are embarrassing not to have 0:17:28.680 --> 0:17:32.399 in the index. Just added was Twitter as of this 0:17:32.480 --> 0:17:37.479 tape in okay Um. And if there's a stock that 0:17:37.560 --> 0:17:44.880 has been brutalized, unloved, dirt cheap and nobody wants it, 0:17:44.880 --> 0:17:46.919 it's embarrassing to have it in the index. So of 0:17:46.920 --> 0:17:50.639 course they're going to take those out. Now, what happens 0:17:50.720 --> 0:17:54.800 is two things. Firstly, they announce a change, and they 0:17:54.840 --> 0:17:57.240 announced the date the change will take effect. That gives 0:17:57.280 --> 0:17:59.880 the index funds a grace period in which to trade 0:18:00.480 --> 0:18:04.360 where they're going to move those stock prices, and those 0:18:04.400 --> 0:18:07.600 stocks will still be in the index, so it won't 0:18:07.600 --> 0:18:12.760 create a performance drag. There. Everybody, please front run our trades. Well, 0:18:12.800 --> 0:18:16.480 there's a hedge fund community that does exactly that. So 0:18:16.600 --> 0:18:21.320 the trades are made, the positions built, and then given 0:18:21.320 --> 0:18:25.040 over to the index funds on the effective date, preferably 0:18:25.160 --> 0:18:30.320 at or near the close. The stocks added when you 0:18:30.359 --> 0:18:33.879 compare them with the discretionary deletions deletions that aren't related 0:18:33.920 --> 0:18:39.760 to corporate actions, um outperformed by nearly nine percent during 0:18:39.880 --> 0:18:43.120 that grace period. So when index funds say we don't 0:18:43.160 --> 0:18:47.959 move stock prices with our trading, that's that's rubbish. Um. 0:18:48.080 --> 0:18:50.399 They let other people move it on their behalf, and 0:18:50.440 --> 0:18:53.359 they could say, oh, look we haven't moved this. Well 0:18:53.480 --> 0:18:55.879 there's that, and then there's also the simple fact that 0:18:55.920 --> 0:19:00.120 the ads beat the discretionary deletions by nine during a 0:19:00.119 --> 0:19:04.919 period of days. Well, that's a big move. So you 0:19:05.119 --> 0:19:09.760 also have part of that return difference in the subsequent 0:19:09.840 --> 0:19:14.760 year is simply mean reversion taking the price impact of 0:19:14.800 --> 0:19:18.560 the index funds back out, and part of it is 0:19:18.680 --> 0:19:22.560 quite simply. The stocks added are extravagantly expensive on average, 0:19:22.960 --> 0:19:26.240 and the stocks dropped are usually a deep discounts um 0:19:26.280 --> 0:19:31.480 over nine of the ads are companies trading at premium multiples. 0:19:31.520 --> 0:19:35.159 Over the discretionary deletes are trading at a discount. And 0:19:35.200 --> 0:19:40.119 the average gap between valuation multiples of the ads and 0:19:40.200 --> 0:19:43.240 of the deletes is more than three to one. That's amazing. 0:19:43.359 --> 0:19:47.520 So let me make the case for quote unquote passive indexing. 0:19:48.280 --> 0:19:50.920 I think smart data. Let me caveat this. Smart data. 0:19:51.000 --> 0:19:53.359 Is that the wrong phrase. It should be fundamental indexing. 0:19:53.800 --> 0:19:56.960 Passive investing is a wrong phrase. It should be low 0:19:57.040 --> 0:20:01.080 cost indexing. But here's the pro indexing argument, and I 0:20:01.119 --> 0:20:04.160 want you to take this argument apart. This is the 0:20:04.240 --> 0:20:08.920 most cost effective, least expensive way to get exposure to equities. 0:20:09.600 --> 0:20:11.879 It has the lowest amount of turnover, the least amount 0:20:11.920 --> 0:20:17.480 of tax consequences, and everybody seems to have the greatest 0:20:17.480 --> 0:20:21.480 difficulty outperforming the SNP five hundred. So we're gonna keep 0:20:21.560 --> 0:20:27.080 using that as a benchmark going forward. Discuss the Bill 0:20:27.119 --> 0:20:30.320 Sharp wrote a piece in the nine nineties entitled the 0:20:30.359 --> 0:20:34.800 arithmetic of active investing, and it's a very simple thesis. 0:20:34.840 --> 0:20:39.000 The market is capitalization weighted. The index fund span almost 0:20:39.040 --> 0:20:42.720 all of the market and is capitalization weighted. You take 0:20:42.920 --> 0:20:47.840 that portfolio, all of the indexers out, and what's left 0:20:47.960 --> 0:20:52.080 is what active managers collectively own. Well, it's the same portfolio, 0:20:52.200 --> 0:20:56.880 giver take some wiggle room. So if it's the same portfolio, 0:20:57.160 --> 0:21:01.400 active managers should have the same per formances index funds 0:21:01.880 --> 0:21:06.119 minus costs, and the costs are higher for active managers. 0:21:06.760 --> 0:21:12.640 That arithmetic is not just true, it's a truism that 0:21:12.720 --> 0:21:17.639 means that if you're choosing active managers with absolutely no skill, 0:21:17.880 --> 0:21:22.920 you should expect to earn index returns minus um when 0:21:22.920 --> 0:21:26.560 you choose an active manager. This doesn't mean in choosing 0:21:26.600 --> 0:21:29.240 active managers as a waste of time. What it does 0:21:29.320 --> 0:21:32.119 mean is you'd better have a good answer to the 0:21:32.200 --> 0:21:35.960 question if this active manager is a winner, so there's 0:21:35.960 --> 0:21:38.560 a loser on the other side of their trades, who's 0:21:38.600 --> 0:21:41.199 the loser and why are they are willing loser for 0:21:41.240 --> 0:21:45.000 fundamental index The answer to that is really simple. This 0:21:45.080 --> 0:21:48.359 is a strategy that contratrades against the market's biggest and 0:21:48.440 --> 0:21:52.160 most extravagant bets, and so the loser on the other 0:21:52.160 --> 0:21:55.920 side of the trade is the performance chasing lemmings who 0:21:55.920 --> 0:22:00.920 are legion. Here's the Bill Miller pushed back. Most active 0:22:00.960 --> 0:22:05.560 managers are as you've described, with a very low active 0:22:05.600 --> 0:22:09.680 share and essentially their closet index ers. So why on 0:22:09.760 --> 0:22:12.200 earth should anybody pay a high fee when you could 0:22:12.200 --> 0:22:16.000 pay a low fee and get the same portfolio. Fair 0:22:16.080 --> 0:22:19.960 fair criticism, It's a fair criticism. There's a lot of 0:22:20.000 --> 0:22:24.080 active managers hiding the bushes near the benchmark. Most active 0:22:24.080 --> 0:22:27.720 managers are constantly looking over their shoulder at the benchmark 0:22:27.920 --> 0:22:31.680 and worrying about beating it. The beauty of fundamental index 0:22:31.880 --> 0:22:36.800 and of smart beta as it was originally defined. Smart 0:22:36.840 --> 0:22:39.679 beta originally meant strategies that break the link with price, 0:22:39.760 --> 0:22:43.920 that don't pay any attention to market capitalization or a 0:22:44.160 --> 0:22:47.640 price in setting the weight of a stock. This term 0:22:47.680 --> 0:22:52.720 has been stretched to the point of meaninglessness. But um, 0:22:52.840 --> 0:22:57.560 under that definition, you do have the advantage that you're 0:22:57.600 --> 0:23:01.240 going to be contratrating against the market's biggest bats um, 0:23:01.280 --> 0:23:06.000 whether you're equal weighting or fundamental index or minimum variants, 0:23:06.080 --> 0:23:09.399 you're going to be having an anchor a target weight 0:23:09.600 --> 0:23:12.240 that isn't related to price, so as the price sores 0:23:12.280 --> 0:23:16.320 and tumbles, you're gonna be selling and buying. It's built 0:23:16.359 --> 0:23:21.360 in structural cell high Bilow discipline. Fascinating. We were talking 0:23:21.800 --> 0:23:28.680 earlier about traditional passive investing or or low cost indexing. 0:23:29.320 --> 0:23:31.840 Let's let's do a little bit of a compare and 0:23:31.960 --> 0:23:38.040 contrast with UM fundamental indexing. And we touched on this. 0:23:38.320 --> 0:23:40.639 I want to give you a quote from Burton Malchiol. 0:23:41.359 --> 0:23:44.840 Smart beta strategies are riskier than index funds and not 0:23:45.040 --> 0:23:50.160 right for individual investors. What is professor Malkiel getting wrong there? 0:23:51.720 --> 0:23:56.800 Malkiel comes from the efficient markets community. He believes he 0:23:57.160 --> 0:23:59.840 wrote random Walk down Wall Street back in I think 0:23:59.840 --> 0:24:07.880 the sixties. UM. It's interesting to note that if you 0:24:08.040 --> 0:24:14.879 believe that UM the the share prices equal a fair 0:24:14.960 --> 0:24:19.439 value that we cannot see plus or minus and error 0:24:21.080 --> 0:24:24.840 adding up to the price, And if you believe the 0:24:24.920 --> 0:24:28.080 market is constantly hunting for those errors and trying to 0:24:28.200 --> 0:24:32.800 fix them so that the error is mean reverting, then 0:24:33.200 --> 0:24:39.320 contratrating against big price moves has a structural alpha that 0:24:39.480 --> 0:24:42.640 is in fact the key driver of the value effect. 0:24:43.000 --> 0:24:47.840 If you look at value strategies, the alpha comes from 0:24:47.840 --> 0:24:52.000 the rebalancing, not from the cheapness of the stocks. Well, 0:24:52.080 --> 0:24:54.119 isn't that the same thing. When we say a stock 0:24:54.240 --> 0:24:57.760 is cheap, we're essentially saying it's trading at a discount 0:24:57.840 --> 0:25:00.919 to its fair value, and once that rebalance o cards 0:25:00.960 --> 0:25:04.160 it should snap back to that and there are gains. Well, 0:25:04.280 --> 0:25:08.360 just as an example, fundamental index overweight stocks that are 0:25:08.400 --> 0:25:11.600 trading at discounts proportional to the magnitude of the discount, 0:25:12.080 --> 0:25:16.320 underweight stocks trading at disc premiums proportional to the magnitude 0:25:16.320 --> 0:25:20.040 of the premium. So your systematically downweighting the growth stocks 0:25:20.080 --> 0:25:22.679 and upweighting the value stocks. You're saying, thanks for the 0:25:22.720 --> 0:25:25.160 gains on the growth stocks, let me trim it back 0:25:25.160 --> 0:25:28.439 to its economic footprint, thank you for the discounts on 0:25:28.480 --> 0:25:31.560 the value stocks, reweighting it back up. So there's a 0:25:31.560 --> 0:25:37.240 structural value tilt. Well, why then, does this strategy relentlessly 0:25:37.720 --> 0:25:43.160 beat cap weighted value indexes Russell value EFA, Value IFA 0:25:43.240 --> 0:25:47.480 Emerging Markets value UM winning against the value index is 0:25:47.520 --> 0:25:50.600 seventy or eighty percent of the time, year by year. Uh. 0:25:50.640 --> 0:25:54.320 It does so for the simple reason that value indexes 0:25:54.359 --> 0:25:56.960 are themselves cap weighted so they're going to overweight the 0:25:57.000 --> 0:26:00.520 overvalue and underweight the undervalued value stocks, even within the 0:26:00.640 --> 0:26:05.840 universe of value correct um. Think of it this way. 0:26:06.320 --> 0:26:12.400 If a stock outperforms in the future, then it must 0:26:12.400 --> 0:26:14.960 have been under under price today. If it underperforms in 0:26:14.960 --> 0:26:20.960 the future, it must have been overpriced today. And what 0:26:21.119 --> 0:26:24.719 stocks get the most weight in a cap weighted portfolio 0:26:25.040 --> 0:26:28.360 those that are overpriced. So let's let's address that. Casette 0:26:28.400 --> 0:26:31.760 raises a really interesting question. When we look at cap 0:26:31.840 --> 0:26:35.960 weighted indexes, we end up owning more of the names 0:26:36.000 --> 0:26:40.000 that are outperforming have been out, I've been out before, 0:26:40.720 --> 0:26:44.719 and since we know momentum as a factor, the assumption 0:26:45.000 --> 0:26:48.840 is that those outperformers will, at least for some finite 0:26:48.840 --> 0:26:52.199 period in the future, continue to outperform. So we end 0:26:52.280 --> 0:26:54.639 up owning more of the winners that are keeping winning 0:26:54.680 --> 0:26:57.399 and less of the losers that are keeping losing. Or 0:26:57.480 --> 0:27:03.280 so goes the cap weighted indet argument. What's wrong with 0:27:03.320 --> 0:27:05.960 that claim? Don't we want more exposure to stocks that 0:27:06.000 --> 0:27:09.600 are winning. Here's what's fascinating. Yeah, momentum is a powerful 0:27:09.640 --> 0:27:15.080 factor across multiple asset classes spanning decades, but within stocks, 0:27:15.400 --> 0:27:19.760 momentum was first published by professors jagged Yaian Titman back 0:27:19.800 --> 0:27:26.480 in nineteen two or three, and since nineteen momentum is 0:27:26.480 --> 0:27:30.440 a strategy in the stock market hasn't worked well, hasn't worked, 0:27:30.440 --> 0:27:34.480 it's underperformed some other factors. No high performing stocks have 0:27:34.560 --> 0:27:39.560 underperformed low performing stocks on average since ninety nine. So 0:27:40.480 --> 0:27:43.159