WEBVTT - Gerard O’Reilly on Academic Research and Stocks 0:00:02.000 --> 0:00:07.160 This is mesters in Business with Very Results on Bloomberg Radio. 0:00:09.320 --> 0:00:12.560 This week on the podcast, I have an extra special guest. 0:00:12.720 --> 0:00:16.239 Gerard O'Riley is a double threat. He is the chief 0:00:16.320 --> 0:00:21.919 investment officer as well as the co CEO of Dimensional 0:00:21.960 --> 0:00:26.560 Funds UH. They are a factor giant, managing about six 0:00:26.920 --> 0:00:31.560 and fifty billion dollars and total assets. This is really 0:00:31.560 --> 0:00:36.960 a master class in how to think about investing, how 0:00:37.000 --> 0:00:41.360 to be systematic, how to approach it from a evidence 0:00:41.440 --> 0:00:48.240 based scientific basis, how to incorporate the best of academic 0:00:48.320 --> 0:00:51.720 research into your process. One of the things that I 0:00:52.000 --> 0:00:56.560 found really interesting was the d F A focus on costs, 0:00:57.040 --> 0:01:02.880 convenience and customization. Not every giant investment firm UH takes 0:01:02.880 --> 0:01:07.600 that approach. Really, I've interviewed a number of folks from 0:01:07.680 --> 0:01:12.240 d F a UM, from David Booth to Gene Parma 0:01:12.400 --> 0:01:16.720 and throughout the rest of the organization. I think you 0:01:16.760 --> 0:01:22.039 will find this to be absolutely fascinating and and really informative. So, 0:01:22.120 --> 0:01:25.720 with no further ado, my conversation with d F as 0:01:25.880 --> 0:01:32.520 Gerard O'Reilly, this is mesters in Business with Very Redults 0:01:33.000 --> 0:01:38.920 on Bloomberg Radio. My extra special guest this week is 0:01:38.959 --> 0:01:43.120 Gerard O'Reilly. He is the Chief Investment Officer and co 0:01:43.360 --> 0:01:47.800 CEO of Dimensional Fund Advisors, a leader and factor based 0:01:47.840 --> 0:01:51.280 investing for the past forty years. D f A has 0:01:51.320 --> 0:01:56.960 about employees across thirteen offices globally and full disclosure my 0:01:57.120 --> 0:02:00.880 firm Results. Wealth Management is a client of Dimensional Funds 0:02:00.880 --> 0:02:04.480 and we manage a substantial chunk of our assets with 0:02:04.600 --> 0:02:08.680 their products. They manage six hundred and fifty billion dollars 0:02:08.680 --> 0:02:13.040 in assets, about eight percent of that is equity. Girard O'Reilly, 0:02:13.520 --> 0:02:17.360 Welcome to Bloomberg. Thank you, Barry, and thanks for the invitation. 0:02:17.400 --> 0:02:19.239 I've been looking forward to speaking with you for some time, 0:02:19.480 --> 0:02:22.360 so so have I. You have such an interesting background. 0:02:22.760 --> 0:02:26.000 I was really excited to talk to you, especially given 0:02:26.760 --> 0:02:30.280 you have a PhD in aeronautics from the California Institute 0:02:30.320 --> 0:02:34.440 of Technology. What what were your original career plans. Well, 0:02:35.080 --> 0:02:40.079 I've always liked mathematics, and as an undergrad in Ireland 0:02:41.080 --> 0:02:44.600 studied mathematics and physics and so on extensively. I was 0:02:44.639 --> 0:02:47.560 thinking about what to do next and said, well, cal 0:02:47.639 --> 0:02:50.720 Tech does a lot of great stuff in fluid mechanics 0:02:51.200 --> 0:02:53.360 and particular in aeronautics and So I didn't have a 0:02:53.400 --> 0:02:56.480 specific set of career plans. I just know that's the 0:02:56.480 --> 0:02:58.480 subject that I wanted to study and that I enjoyed. 0:02:58.800 --> 0:03:02.600 So I set off or cal Tech and and I 0:03:02.680 --> 0:03:06.760 really enjoyed my time. They're working on various different projects, 0:03:06.840 --> 0:03:11.480 many theoretical in nature, but very mathematically oriented. So when 0:03:11.520 --> 0:03:14.600 you're looking at aeronautics in the United States, there aren't 0:03:14.600 --> 0:03:16.760 a whole lot of career paths out of that other 0:03:16.840 --> 0:03:20.600 than academia or going to Nassau or one of the 0:03:20.639 --> 0:03:25.519 defense um companies. What led the shift from aeronautics to 0:03:25.680 --> 0:03:29.520 finance as a career, Well, I wanted to learn some 0:03:29.560 --> 0:03:32.519 more about finance. Number one. I hadn't taken a finance 0:03:32.560 --> 0:03:36.040 course ever in my life before joining Dimensional, and Dimensional 0:03:36.120 --> 0:03:39.640 was the firm that I joined straight out of college. Also, 0:03:40.120 --> 0:03:43.120 you know, the academic path wasn't one that appealed to me. 0:03:43.160 --> 0:03:47.480 I really enjoyed grad school, but I preferred to tackle 0:03:47.640 --> 0:03:50.040 something that was I would say more than here and now, 0:03:50.280 --> 0:03:54.839 where your projects that you're working on have impact very 0:03:54.960 --> 0:03:58.080 quickly on you know, the end customer, the end consumer, 0:03:59.160 --> 0:04:02.800 and then also where on the engineering side. You know, 0:04:02.840 --> 0:04:04.400 you mentioned in the U S well, I'm not a 0:04:04.480 --> 0:04:06.840 US citizens, so it's hard for non US citizen to 0:04:06.840 --> 0:04:08.760 work in the aeronautics field here in the US because 0:04:09.200 --> 0:04:13.520 largely required security clearance. So it's looking around. A friend 0:04:13.520 --> 0:04:16.920 of mine was working for Dimensional, knew that person at Caltech, 0:04:17.560 --> 0:04:21.480 and that person was talking about you know, Dimensional has 0:04:21.520 --> 0:04:25.800 all these great academic connections. They really take finance from 0:04:25.800 --> 0:04:29.400 a scientific perspective. Went down, checked it out and said, 0:04:29.920 --> 0:04:31.640 this sounds interesting. I really want to give this a 0:04:31.680 --> 0:04:33.880 shot for a period of time. So so let's talk 0:04:33.920 --> 0:04:38.920 a little bit about those academic um connections. Ken French 0:04:39.000 --> 0:04:41.799 has been at Dartmouth for a long time. His colleague 0:04:42.160 --> 0:04:45.919 Gene Farmer, Nobel Prize winner at University of Chicago. Another 0:04:46.160 --> 0:04:50.520 Nobel Prize winner, Robert Merton, also at Dimensional funds. What's 0:04:50.560 --> 0:04:55.039 it like working with all these Nobel Prize winning economists 0:04:55.440 --> 0:04:58.560 could be a little intimidating to some folks. It's always 0:04:58.560 --> 0:05:01.520 intimidating when you start off working with somebody who's very, 0:05:01.600 --> 0:05:03.960 very talented and you're getting to know them for the 0:05:04.000 --> 0:05:07.360 first time. But it's a privilege and it's great fun 0:05:07.760 --> 0:05:12.680 because those folks, you know, have have worked incredibly hard 0:05:12.880 --> 0:05:15.960 to hone their craft, hone their skills, and when you 0:05:16.400 --> 0:05:18.880 think about Ken or Gene or Bob or Meer and 0:05:18.960 --> 0:05:21.520 any of those folks, they're very, very generous with their 0:05:21.520 --> 0:05:25.040 time and so they're willing to teach because they're in academia, 0:05:25.600 --> 0:05:27.840 and if you're willing to work hard, they're willing to 0:05:27.880 --> 0:05:30.000 put the time and effort into you. So I started 0:05:30.000 --> 0:05:34.200 off with no background in finance and got to learn 0:05:34.279 --> 0:05:37.520 finance from some of the most amazing minds in the field. 0:05:37.640 --> 0:05:41.640 So it was just it was great Ken, Jean and Bob. 0:05:41.920 --> 0:05:45.000 I've never heard of those three gentlemen referred to quite 0:05:45.000 --> 0:05:47.200 in that way, but I guess when you work with 0:05:47.279 --> 0:05:50.400 them as frequently as you do UM, it becomes Ken, 0:05:50.480 --> 0:05:54.560 Jean and Bob. So, so what are the parallels between 0:05:54.640 --> 0:06:00.919 academia and UM working in finance professionally? And then I 0:06:00.960 --> 0:06:04.359 have to ask what are the parallels between aeronautics and 0:06:04.400 --> 0:06:09.680 fluid dynamics and finance and investment. Well, working in academia, 0:06:10.279 --> 0:06:12.560 you know, you're always trying to solve a problem. You're 0:06:12.560 --> 0:06:16.480 looking for interesting problems to solve that haven't really been 0:06:16.480 --> 0:06:19.440 tackled before, or an aspect that you're working on hasn't 0:06:19.440 --> 0:06:22.479 been tackled before, and you're saying, well, can I bring 0:06:22.600 --> 0:06:26.719 something new to the table, something innovative, and that's incredibly 0:06:26.760 --> 0:06:30.599 rewarding and incredibly interesting. Working in finance is no different. 0:06:31.000 --> 0:06:35.880 You're looking for new problems to solve. Those problems are 0:06:35.960 --> 0:06:39.200 largely driven by what it is your clients are looking for, 0:06:39.320 --> 0:06:44.400 what types of investment solutions do they require to solve 0:06:44.440 --> 0:06:47.480 the investment problems that they have, and then you're coming 0:06:47.520 --> 0:06:49.880 up with innovative ways to solve those problems. So in 0:06:49.880 --> 0:06:53.240 that respect, there's a lot of similarities. The time scale 0:06:53.240 --> 0:06:55.400 and the time frames are a little bit tighter and 0:06:55.480 --> 0:06:58.960 faster when it comes to finance than in academia. In 0:06:59.000 --> 0:07:02.320 academia it may be multi years, and there's multi year 0:07:02.440 --> 0:07:04.560 projects that happen in finance, but you want to be 0:07:04.600 --> 0:07:07.320 able to live or something for your clients in shorter 0:07:07.360 --> 0:07:10.760 time frames than that. When you think about engineering or 0:07:11.360 --> 0:07:15.120 mathematics or physics, and then how does how do those 0:07:15.160 --> 0:07:17.920 skill sets translate over to finance. Well, again, it's all 0:07:17.960 --> 0:07:22.320 about problem solving and what you're looking for is how 0:07:22.360 --> 0:07:27.640 do I number one pose the question correctly? How do 0:07:27.680 --> 0:07:30.680 I ask the right question? Because that's as important as 0:07:30.720 --> 0:07:32.960 trying to solve the problem, you have to set it 0:07:33.040 --> 0:07:35.280 up in the right way. And that's true whether it's 0:07:35.320 --> 0:07:38.120 in mathematics or physics, or engineering, or it's in finance. 0:07:38.560 --> 0:07:42.440 Then how do I gather data to help me address 0:07:42.480 --> 0:07:45.280 and find the answer to this question? And that's true 0:07:45.440 --> 0:07:48.160 of both fields. And then how do I interpret the data? 0:07:48.400 --> 0:07:50.280 What are the tools and the models that I can 0:07:50.400 --> 0:07:52.520 use such that I'm going to be able to organize 0:07:52.560 --> 0:07:55.280 these data in such a way to draw inferences about 0:07:55.720 --> 0:07:58.720 how I want to act going forward. And that's true 0:07:58.720 --> 0:08:02.640 of both, Madam, physics, engineering, and finance. I think the 0:08:02.640 --> 0:08:06.640 big differences are the laws of physics tend not to 0:08:06.720 --> 0:08:11.840 change over time, but the laws of that government finance 0:08:11.880 --> 0:08:16.760 can change through time. There are many repeatable experiments in physics, 0:08:17.080 --> 0:08:20.720 there are no repeatable experiments in finance. But there is 0:08:20.800 --> 0:08:23.080 a kind of a common truth in both, which is 0:08:23.120 --> 0:08:28.120 that in finance and investing, people demand return for bearing uncertainty. 0:08:28.200 --> 0:08:30.840 That doesn't change through time, but how you go about 0:08:30.880 --> 0:08:33.840 implementing that can change through time because the laws are changing. 0:08:35.000 --> 0:08:37.960 On behalf of Isaac Newton, I'm going to raise an 0:08:37.960 --> 0:08:41.120 objection that at least our understanding of the laws of 0:08:41.120 --> 0:08:46.360 physics have changed over time. So so maybe the underlying 0:08:46.440 --> 0:08:49.360 laws themselves are the same, but our perception seems to 0:08:49.400 --> 0:08:52.600 have evolved. I think that's that's a good way to 0:08:52.600 --> 0:08:54.720 put it, in a nice precise way to put it, 0:08:54.760 --> 0:08:59.640 is that the underlying drivers don't change, but our perception changes. 0:08:59.679 --> 0:09:02.760 And that's it's an interesting observation because their perceptions change 0:09:03.160 --> 0:09:07.439 because the models that we use to explain and understand 0:09:07.640 --> 0:09:11.079 those underlying drivers evolve over time. All models are incomplete, 0:09:11.400 --> 0:09:13.120 none of them are true, none of them are perfect 0:09:13.120 --> 0:09:15.800 descriptions of reality. And that's true of physics and it's 0:09:15.840 --> 0:09:19.600 true of finance. But you can improve those models over time. 0:09:20.000 --> 0:09:22.120 You can improve the data that you can collect over time, 0:09:22.400 --> 0:09:25.200 and that enhances your understanding over time. I'm a big 0:09:25.200 --> 0:09:28.160 fan of George Box. I love the quote all models 0:09:28.160 --> 0:09:31.040 are wrong, but some are useful, and it sounds like 0:09:31.160 --> 0:09:35.240 you very much embrace that philosophy as well. And it's 0:09:35.280 --> 0:09:38.720 an important philosophy to embrace when you're, you know, working 0:09:38.720 --> 0:09:42.120 in the field of finance, because ultimately, what you're doing 0:09:42.240 --> 0:09:45.720 is you're investing money on behalf of others. It's their 0:09:45.720 --> 0:09:49.439 life savings. Often so it's what they've made sacrifice to 0:09:49.880 --> 0:09:52.880 put together so they can afford a better retirement or 0:09:52.960 --> 0:09:56.319 something that's important to them in their future. And if 0:09:56.320 --> 0:09:59.240 you ever believe that the model is reality, you're probably 0:09:59.240 --> 0:10:01.880 going to build non robust solutions and do them a disservice. 0:10:02.120 --> 0:10:07.480 So having a healthy skepticism around all models and basically 0:10:07.559 --> 0:10:11.280 all data sources that you see is important because it 0:10:11.360 --> 0:10:14.760 leads you to, well, what if I'm wrong, do I 0:10:14.800 --> 0:10:17.120 still have a good in solution even if this model 0:10:17.679 --> 0:10:20.880 it turns out to be incorrect? And I think that's 0:10:20.880 --> 0:10:22.960 that's a good way of looking at the world. So 0:10:23.080 --> 0:10:25.640 let's talk a little bit about your career. You began 0:10:25.679 --> 0:10:28.120 at d f A in two thousand four in the 0:10:28.160 --> 0:10:32.120 research department. A little more than a decade later your 0:10:32.160 --> 0:10:36.120 chief investment officer, and not that many years after that 0:10:36.200 --> 0:10:41.400 you become co chief executive officer. That's a pretty rapid 0:10:41.480 --> 0:10:44.559 career path. Explain to us, if you would, the concept 0:10:44.640 --> 0:10:46.959 of co CEO S or co c i O S 0:10:47.320 --> 0:10:52.440 and how you managed to um advance so rapidly in 0:10:52.480 --> 0:10:56.320 a firm that was led by David Booth for so 0:10:56.360 --> 0:10:59.160 many decades. Yeah, So let me let me start with 0:10:59.200 --> 0:11:02.520 the ladder. How how how do you advance? And my 0:11:02.600 --> 0:11:05.880 viewpoint on success is there's a combination of three things, 0:11:05.920 --> 0:11:08.520 and I'm not sure which one is most important, but 0:11:08.559 --> 0:11:12.200 they probably all are equally important at different stages. One 0:11:12.559 --> 0:11:14.640 is a little bit of luck, a little bit of 0:11:14.720 --> 0:11:19.000 luck in the things that you've learned up to that 0:11:19.040 --> 0:11:22.080 point in time when the opportunity comes, a little bit 0:11:22.120 --> 0:11:25.680 of luck. For example, finding Dimensional was well suited to 0:11:26.280 --> 0:11:28.679 the way that I thought about the world. And then 0:11:28.720 --> 0:11:32.559 there's some talent. Do you have the right skill set 0:11:33.160 --> 0:11:36.880 that will be helpful in that particular organization? And it 0:11:36.920 --> 0:11:40.240 turns out that a quantitative and analytical type skill set 0:11:40.440 --> 0:11:44.320 was very helpful for an organization like Dimensional in our clients. 0:11:44.880 --> 0:11:47.960 And then hard work. Are you willing to do whatever 0:11:48.000 --> 0:11:53.319 it takes to complete projects to move the ball forward 0:11:53.559 --> 0:11:57.440 to help your clients succeed. And when you have all 0:11:57.480 --> 0:11:59.840 three of those, I think good things can happen. And 0:12:00.120 --> 0:12:02.440 was fortunate that had a little bit of each one 0:12:02.480 --> 0:12:06.200 of those when I came to Dimensional, and Dimensional has 0:12:06.240 --> 0:12:09.200 been a growing firm for money money decades and when 0:12:09.200 --> 0:12:11.920 I came in two thousand or four, we had about 0:12:11.960 --> 0:12:15.960 fifty billion under management and you know, that grew rapidly, 0:12:16.000 --> 0:12:17.960 so there was a lot of opportunities for those folks 0:12:17.960 --> 0:12:20.679 that were willing willing to step up. And so I 0:12:20.800 --> 0:12:24.320 consider myself fortunate and very happy by how that's turned out, 0:12:24.320 --> 0:12:26.440 because I've had a blast doing it and it's been 0:12:26.520 --> 0:12:29.280 it's been rewarding. And then in terms of the co 0:12:29.520 --> 0:12:31.280 c i O s and co CEOs, we do a 0:12:31.320 --> 0:12:34.479 lot of CODs. We have cos of different department heads. 0:12:34.480 --> 0:12:39.839 From my particular case, Dave Butler is the other co CEO. 0:12:40.480 --> 0:12:43.800 And it tends to work well when you have people 0:12:43.840 --> 0:12:46.920 who number one get along well with each other, they 0:12:46.960 --> 0:12:50.280 respect each other and each other's ideas, and then they 0:12:50.280 --> 0:12:53.600 have maybe complementary skill sets. And so the way that 0:12:53.720 --> 0:12:57.920 Dave and I have worked in that job together I 0:12:57.960 --> 0:13:00.960 think has been much more all my preference I would have. 0:13:01.080 --> 0:13:02.880 I'm much preferred to have done it with him then 0:13:02.920 --> 0:13:06.880 without them, because you can do some dividing and conquering. 0:13:07.559 --> 0:13:10.280 But also what I find is that as you get promotions, 0:13:10.320 --> 0:13:13.320 and this is a little bit facetious, but you tend 0:13:13.320 --> 0:13:16.280 to become, at least if you judge it by the 0:13:16.320 --> 0:13:21.080 input that you get from your peers, smarter and funnier 0:13:21.600 --> 0:13:24.760 in that the input that you get from your peers 0:13:24.800 --> 0:13:28.439 becomes less informationally rich. But when you have a troop 0:13:28.480 --> 0:13:31.840 heer like Dave and I are are troop peers, anything goes. 0:13:31.960 --> 0:13:35.360 We can have robust, open, honest conversations and with David 0:13:35.400 --> 0:13:38.080 as well, which read us a lot of pressure test 0:13:38.160 --> 0:13:41.280 things before we have to go and talk about them 0:13:41.320 --> 0:13:43.880 with the rest of the firm. And that really, you know, 0:13:43.880 --> 0:13:46.480 I always think iron sharpens iron, that you have to 0:13:46.559 --> 0:13:49.120 have people who you can, you know, spar with on 0:13:49.160 --> 0:13:52.040 a daily basis test your ideas. They'll push you, you 0:13:52.080 --> 0:13:56.439 will push them so that you can improve every day. 0:13:56.520 --> 0:13:59.040 So it's it's worked very very well. We do a 0:13:59.080 --> 0:14:02.760 divide and conquered. We've thirteen global departments at dimensional. Four 0:14:02.800 --> 0:14:06.320 comes straight to me, four go straight to him, and 0:14:06.360 --> 0:14:08.920 then the five in the middle kind of go to 0:14:08.960 --> 0:14:11.000 both of us either through the CEO we have a 0:14:11.040 --> 0:14:14.320 CEO Lisa Dalmer, are directly like legal and compliance come 0:14:14.360 --> 0:14:17.920 to both of us directly, and that way that it's 0:14:18.000 --> 0:14:21.080 it's just worked well. We've been very pleased with what 0:14:21.160 --> 0:14:23.120 we've been able to accomplish over the past five years 0:14:23.120 --> 0:14:26.080 working together. It's I guess you each keep each other 0:14:26.160 --> 0:14:29.960 sharp and keep each other honest. That's right, really interesting. 0:14:30.480 --> 0:14:33.960 So so let's talk about factors a little bit um. 0:14:34.000 --> 0:14:38.880 How did the academic research that that Rex and David, 0:14:39.000 --> 0:14:43.160 the two co founders of d f A, How did 0:14:43.200 --> 0:14:46.160 that become part of the investment process. So I guess 0:14:46.200 --> 0:14:49.560 there's a couple of salient points there. One is factor 0:14:49.640 --> 0:14:52.920 research in itself, and we talked a little bit earlier 0:14:52.920 --> 0:14:55.880 on about models and what they're useful for and how 0:14:55.920 --> 0:14:58.520 you draw inferences from them. I really look on factor 0:14:58.560 --> 0:15:01.920 models as way to organize historical data so you can 0:15:01.960 --> 0:15:06.360 try to understand better what really drove differences and returns 0:15:06.400 --> 0:15:09.880 across different groups of securities, different groups of stocks, different 0:15:09.920 --> 0:15:13.960 group of bonds, and from those you can glean very 0:15:13.960 --> 0:15:19.560 important insights about the drivers of expected returns, the drivers 0:15:19.720 --> 0:15:24.760 of differences and risk across different asset categories. And so 0:15:24.840 --> 0:15:29.000 I think that's the important aspect of factor models. So 0:15:29.040 --> 0:15:32.760 when you put them dimensional and its founding in context 0:15:32.840 --> 0:15:35.360 of kind of a burgeoning field in the eighties and 0:15:35.400 --> 0:15:38.360 in the nineties, when more and more factor models were 0:15:38.400 --> 0:15:43.480 being developed and tested and so on. The founding was 0:15:44.560 --> 0:15:48.600 two I would say, address an institutional need that David 0:15:48.720 --> 0:15:53.440 had identified, which was there weren't many systematic strategies that 0:15:53.480 --> 0:15:58.040 targeted the returns of small cap stocks, and he found 0:15:58.040 --> 0:16:01.840 that that that was a hole in many institutional investor portfolios. 0:16:02.360 --> 0:16:05.160 And along the around the same time, because David had 0:16:05.200 --> 0:16:08.200 done his MBA at the University Chicago now Both School 0:16:08.240 --> 0:16:12.160 of Business, around that same time, there was evidence coming 0:16:12.200 --> 0:16:15.480 out that smaller cap stocks also had higher average returns 0:16:15.520 --> 0:16:18.800 historically and reasons you know, promoted about why that would 0:16:18.840 --> 0:16:22.120 be higher expected return is going forward, and so around 0:16:22.160 --> 0:16:24.720 that time was kind of when those factor models were developing. 0:16:24.720 --> 0:16:27.320 So I started with the client need, and then it 0:16:27.440 --> 0:16:30.840 was well, let me go to the academics and understand, 0:16:31.600 --> 0:16:35.080 what are the research around this client need. Am I 0:16:35.160 --> 0:16:37.160 going to do something here that makes sense or not 0:16:37.200 --> 0:16:40.320 makes sense from an academic perspective? And then how do 0:16:40.360 --> 0:16:45.040 I build a good robust solution to address that client need. 0:16:45.400 --> 0:16:47.040 And then, of course, in the nineties you had the 0:16:47.080 --> 0:16:49.240 three factor model come along, and then in the mid 0:16:49.320 --> 0:16:51.520 nineties you had momentum come along, and in the two 0:16:51.520 --> 0:16:54.800 thousands you had things like profitability and investment come along. 0:16:55.160 --> 0:16:57.800 So we had lots of different factors uncovered over time. 0:16:58.240 --> 0:16:59.480 But the way that we look on each one of 0:16:59.520 --> 0:17:02.680 those is their models. They give us insights from the data. 0:17:03.000 --> 0:17:05.840 How do you use that to build robust portfolios? And 0:17:05.880 --> 0:17:08.000 I would say that's been kind of part of our 0:17:08.040 --> 0:17:11.560 heritage for forty years. How do we build portfolios that 0:17:11.640 --> 0:17:16.280 can target these premiums but be robust regardless of the 0:17:16.320 --> 0:17:19.280 market environment. And we've been through many different market crises 0:17:19.640 --> 0:17:22.080 with a broad range of investment strategies that have come 0:17:22.080 --> 0:17:26.000 out quite well the other side. So we're we're pretty 0:17:26.080 --> 0:17:30.520 familiar in modern times with small cap indices like the 0:17:30.600 --> 0:17:33.919 Russell two thousand, or the S and P six hundred 0:17:34.000 --> 0:17:37.520 or whatever it happens to be. But when Sinkfeld and 0:17:37.760 --> 0:17:41.760 Booth were forming um D f A in the early eighties, 0:17:42.560 --> 0:17:46.920 these weren't really household names, if they even existed at all. 0:17:48.240 --> 0:17:51.320 It's amazing to think that there was a period where 0:17:52.040 --> 0:17:55.280 small caps weren't their own category. Tell us a little 0:17:55.320 --> 0:17:58.400 bit about how that evolved. Yeah, if you go back 0:17:58.400 --> 0:18:00.640 even further, so dimension was found in than eighty one, 0:18:00.680 --> 0:18:04.080 But if you go back a decade earlier, and I'll 0:18:04.119 --> 0:18:06.399 focus on David a little bit and his work with 0:18:06.480 --> 0:18:09.880 mc McCown, who was at Wells Fargo at the time, 0:18:09.920 --> 0:18:14.000 and he's a director of the firm, and so David 0:18:14.040 --> 0:18:18.600 and Mac were working on indexes. So in the very 0:18:18.600 --> 0:18:23.440 early seventies, the Max team with David created the first 0:18:23.480 --> 0:18:26.119 index fund. It wasn't for retail, it was for an 0:18:26.119 --> 0:18:29.480 institutional client, and it was based on US large cap stocks, 0:18:29.520 --> 0:18:33.000 So he's very familiar with index based approaches. Then David 0:18:33.040 --> 0:18:36.600 subsequently left and worked a gibecker for a while, understood 0:18:36.640 --> 0:18:40.320 more about what clients were interested in looking for required 0:18:41.080 --> 0:18:44.480 and so there wasn't a Russell two thousand available when 0:18:44.600 --> 0:18:47.000 he was building the firm, so there wasn't an index 0:18:47.040 --> 0:18:50.560 to attach the strategy to. The Other thing that was 0:18:51.240 --> 0:18:54.520 kind of feedback from academia is yes, small cap investing 0:18:54.560 --> 0:18:56.760 makes sense, but you're going to get killed on trading costs. 0:18:57.440 --> 0:19:00.280 And so then you have this kind of environment where 0:19:00.600 --> 0:19:03.280 there wasn't an index, it wasn't a household name. To 0:19:03.359 --> 0:19:06.240 your point, you know small cap stocks as an asset category, 0:19:06.600 --> 0:19:09.800 so you kind of have a blank canvas. If I know, 0:19:09.800 --> 0:19:12.280 knowing everything that I know, what's the right way to 0:19:12.320 --> 0:19:16.399 build a small cap strategy that hopefully then will be 0:19:16.440 --> 0:19:20.159 efficient and won't suffer too greatly from trading costs and 0:19:20.200 --> 0:19:24.199 implementing and investing client flows. So I think that it 0:19:24.280 --> 0:19:29.159 was in some respects a very big advantage starting with 0:19:29.200 --> 0:19:32.359 that blank canvas of how do you design the best portfolio, 0:19:32.400 --> 0:19:36.240 you know how, with as few constraints as possible, because 0:19:36.400 --> 0:19:39.200 you weren't worried about an index. And then subsequently Russell 0:19:39.920 --> 0:19:42.159 had the Russell two thousand, and then of course in 0:19:42.200 --> 0:19:46.080 the nineties, value versus growth became, you know, well established 0:19:46.080 --> 0:19:49.200 asset categories, and so asset categories have been added over time. 0:19:49.920 --> 0:19:53.120 So so let's talk a little bit about Gene Parma 0:19:53.359 --> 0:19:56.520 and Ken French is what started out as a three 0:19:56.520 --> 0:19:59.440 factor model, it eventually became five and seven. Now they're 0:19:59.480 --> 0:20:05.000 a hundred of factors, many of which um don't really 0:20:05.040 --> 0:20:07.880 add a whole lot of alpha or not consistent enough 0:20:07.920 --> 0:20:12.320 alpha to justify their complications and costs. Tell us a 0:20:12.359 --> 0:20:15.760 little bit about the Farmer French factor model. Yeah, so 0:20:16.160 --> 0:20:17.800 you know, when you when you go to the eighties, 0:20:17.840 --> 0:20:21.320 there was a lot of empirical evidence being uncovered that 0:20:21.400 --> 0:20:24.040 the prevailing model from the sixties and the seventies, the 0:20:24.080 --> 0:20:28.800 capital asset pricing model, didn't explain the data very well, 0:20:29.280 --> 0:20:30.639 so when you look at it, it was it was 0:20:30.680 --> 0:20:34.480 a beautiful model. It was very you know, intuitive, but 0:20:34.560 --> 0:20:37.400 it didn't explain the data all that well. And so 0:20:37.840 --> 0:20:41.040 Ken and Gene in the early nineties started to organize 0:20:41.040 --> 0:20:43.000 all the data to say, can we put some of 0:20:43.000 --> 0:20:47.240 these observations in one kind of unified viewpoint of the 0:20:47.320 --> 0:20:52.720 historical data. And from that, you know, exercise came a 0:20:52.880 --> 0:20:57.000 better model in the sense that it could explain the 0:20:57.080 --> 0:21:00.560 returns that you saw among stocks are better than the 0:21:00.600 --> 0:21:03.080 capital as sur pricing models, so explain more of the returns, 0:21:03.119 --> 0:21:05.560 more of the variation that you saw on the returns 0:21:05.600 --> 0:21:09.400 across stocks, and so that so subsequently came the three 0:21:09.440 --> 0:21:12.480 factor model. Then to your point, lots of factors have 0:21:12.520 --> 0:21:14.600 been added. If you look at family frenches are even 0:21:14.680 --> 0:21:18.360 Ken's website, now you'll see a profitability factor, you'll see 0:21:18.359 --> 0:21:22.560 an investment factor, you'll see momentum factors. You'll see all 0:21:22.560 --> 0:21:25.520 different types of factors. And as I mentioned earlier, factors 0:21:25.520 --> 0:21:28.479 are really great to help you organize the historical data. 0:21:29.400 --> 0:21:31.880 But you don't want to get kind of two stereoid 0:21:31.960 --> 0:21:35.320 about the latest factor model. I kind of view a 0:21:35.320 --> 0:21:37.840 lot of the academic research over the past thirty years 0:21:38.200 --> 0:21:41.719 as doing variance on a theme, and so it's not 0:21:41.840 --> 0:21:44.760 that kind of a have brand new discovery, but it 0:21:44.800 --> 0:21:48.760 refines your understanding of existing factors. So there's probably twenty 0:21:48.840 --> 0:21:51.359 or thirty or forty different value factors out there, but 0:21:51.440 --> 0:21:53.600 you don't need all twenty or thirty or forty when 0:21:53.600 --> 0:21:56.760 you're managing a strategy. But you can get insights from 0:21:56.760 --> 0:21:59.720 the different factors on how to manage a strategy effectively. 0:22:00.359 --> 0:22:03.000 And so what I mean by that is if you 0:22:03.240 --> 0:22:05.639 if if you think about what datas are are available. 0:22:06.080 --> 0:22:09.400 You have security prices, you have data from income statements, 0:22:09.480 --> 0:22:12.960 so things like income or profits or revenues or expenses, 0:22:13.359 --> 0:22:15.880 and you have data from balance sheets, assets and liabilities. 0:22:16.040 --> 0:22:19.320 They're the broadly the data that are available to go test. 0:22:20.160 --> 0:22:22.600 And when you look at all of those factor models 0:22:22.600 --> 0:22:24.679 their variants on the theme, the right are looking at 0:22:24.680 --> 0:22:28.760 current values of those variables, whether it's current income or 0:22:28.840 --> 0:22:31.639 current price to book ratios or price earnings ratios, are 0:22:31.640 --> 0:22:34.400 they're looking at how they've changed, How to have prices 0:22:34.480 --> 0:22:37.480 changed over the past number of months, How have assets 0:22:37.480 --> 0:22:40.359 grown over the past number of months, How is profitability 0:22:40.440 --> 0:22:42.360 changed over the past number of months. So there's three 0:22:42.440 --> 0:22:45.159 data sources and people do two things with them, so 0:22:45.200 --> 0:22:48.120 there's actually really kind of six that you can think 0:22:48.160 --> 0:22:51.679 about that kind of encompass most of the hundreds of 0:22:51.720 --> 0:22:54.080 factors that you see out there. And I think that 0:22:54.119 --> 0:22:57.560 if you have coverage of those six current prices, current 0:22:57.600 --> 0:23:00.359 balance sheet items, current income statement items, and then how 0:23:00.359 --> 0:23:03.240 each one of those have changed in recent past, you 0:23:03.320 --> 0:23:07.480 have pretty broad coverage of all the various different factor 0:23:07.560 --> 0:23:10.960 literature that's available. And that's what we do at Dimensional. 0:23:11.320 --> 0:23:14.440 So so let's for the lay person get a little 0:23:14.440 --> 0:23:19.439 more granular with some of the more popular and effective 0:23:19.960 --> 0:23:26.560 factors um. The four biggest ones I think are size, value, quality, 0:23:26.600 --> 0:23:29.080 and momentum. Is there anything you would add to that 0:23:29.280 --> 0:23:32.600 beyond beta which is just a given? So there's five? 0:23:33.000 --> 0:23:35.720 What else would you add to that list? I would 0:23:35.720 --> 0:23:39.240 add probably investment and proxy for investment is how a 0:23:39.320 --> 0:23:42.879