WEBVTT - How the Hottest Hedge Funds on Wall Street Really Manage Risk 0:00:02.640 --> 0:00:16.239 Bloomberg Audio Studios, Podcasts, Radio News. 0:00:18.360 --> 0:00:22.000 Hello and welcome to another episode of the Odd Lots podcast. 0:00:22.079 --> 0:00:24.480 I'm Joe Wisenthal and I'm Tracy Alloway. 0:00:24.720 --> 0:00:28.240 Tracy, I still want to learn more about how multi 0:00:28.240 --> 0:00:29.640 strategy hedge funds work. 0:00:30.080 --> 0:00:31.600 I thought you were going to say, I still don't 0:00:31.600 --> 0:00:35.360 know anything about multi strategy. I feel like we're slowly 0:00:35.440 --> 0:00:39.640 getting there, and hopefully our listeners don't mind coming along 0:00:39.800 --> 0:00:42.800 with us for the ride. I feel like every time 0:00:42.960 --> 0:00:45.960 we have an episode on multi strategy hedge funds, or 0:00:45.960 --> 0:00:48.960 on the pod shops as they are sometimes called, we 0:00:49.040 --> 0:00:52.800 are deepening our understanding and we're sort of getting into 0:00:52.920 --> 0:00:56.400 more and more detail. And I feel confident that one day, 0:00:56.640 --> 0:00:59.320 after we've done like fifty episodes on this topic, we 0:00:59.360 --> 0:00:59.920 will get there. 0:01:00.280 --> 0:01:02.760 I do think it would take about fifty I think 0:01:02.760 --> 0:01:05.000 that's like an accurate number of what it would actually 0:01:05.040 --> 0:01:08.240 take to get there. But of course, most recently we 0:01:08.400 --> 0:01:13.120 had that episode with Giuseppe Pallioligo Gappy talking about some 0:01:13.240 --> 0:01:15.720 of the big ideas and sort of from a high 0:01:15.840 --> 0:01:19.080 level of how some of these funds actually work. They're 0:01:19.160 --> 0:01:22.000 very popular. They've done some of the big ones that 0:01:22.040 --> 0:01:25.480 people know, like the millenniums, like the Citadels have just 0:01:25.520 --> 0:01:29.479 had incredible runs. Really seems to be displacing a lot 0:01:29.520 --> 0:01:33.040 of the old style quant disrupting the sort of fund 0:01:33.040 --> 0:01:36.440 of funds idea that was popular. I have some sense, 0:01:36.520 --> 0:01:38.520 you know, you have all these managers and you give 0:01:38.520 --> 0:01:41.640 them very specific mandates and they have to really focus 0:01:41.680 --> 0:01:44.040 on that, and then if they're not too correlated with 0:01:44.080 --> 0:01:47.480 each other, you can get above market returns in theory 0:01:47.520 --> 0:01:52.360 and apparently in practice, but like how that actually works, 0:01:52.440 --> 0:01:53.360 I still really don't know. 0:01:53.680 --> 0:01:57.640 Well, Okay, so two things. Number One, everyone should definitely 0:01:57.880 --> 0:02:00.960 go and check out Gappy's book if you haven't already, 0:02:01.080 --> 0:02:04.640 Advanced Portfolio Management. A lot of the references that I'm 0:02:04.800 --> 0:02:07.080 about to throw out on this episode, anything that I 0:02:07.160 --> 0:02:11.160 say that might sound even remotely impressive or like I 0:02:11.200 --> 0:02:13.920 know what I'm talking about, has come up Gappy's book. 0:02:14.120 --> 0:02:16.080 And also I will say I've read that book going 0:02:16.120 --> 0:02:19.639 to and from work on the subway. It's pretty short, 0:02:19.680 --> 0:02:21.040 so I think I did it in like a week. 0:02:21.560 --> 0:02:24.840 And I have never gotten so many people like talking 0:02:24.880 --> 0:02:27.480 to me on the subway when they saw me pull 0:02:27.520 --> 0:02:30.079 out Advance Portfolio Management and they're like, what is. 0:02:30.040 --> 0:02:31.560 That that is very New York. 0:02:31.760 --> 0:02:35.000 Yes. And then secondly, the other thing I will say 0:02:35.160 --> 0:02:37.920 is we've been talking about multi strategy hedge funds. We 0:02:37.919 --> 0:02:40.280 want to learn more about them because they're this new 0:02:40.280 --> 0:02:42.920 thing on Wall Street that everyone seems very excited and 0:02:43.000 --> 0:02:46.960 interested in. But beyond that, there are recent events that 0:02:47.080 --> 0:02:50.600 make this an even more pressing topic. So we've seen 0:02:50.800 --> 0:02:53.359 some of the big winners in the market in recent 0:02:53.440 --> 0:02:56.360 months start to come down, So the big tech names 0:02:56.480 --> 0:03:00.400 things like Nvidia, we've seen small caps shoot up. A 0:03:00.440 --> 0:03:02.200 lot of people are talking about whether or not this 0:03:02.360 --> 0:03:06.639 is a factor rotation, and we'll get into what factors 0:03:06.680 --> 0:03:09.919 actually are. But I think the discussion that we're seeing 0:03:10.000 --> 0:03:12.360 right now, and I should caveat this with it is 0:03:12.480 --> 0:03:15.160 July eighteenth. So we've seen those big moves in the 0:03:15.200 --> 0:03:19.840 market very recently. The discussion that's happening now is how 0:03:19.960 --> 0:03:24.640 much does the I guess growth in factor investing feed 0:03:24.720 --> 0:03:28.000 into some of these moves, and also how does the 0:03:28.520 --> 0:03:33.000 risk models that go alongside this actually impact investor behavior 0:03:33.080 --> 0:03:35.720 and then also feed into these market moves. So is 0:03:35.720 --> 0:03:37.920 it the case that everyone's getting out of big tech 0:03:38.240 --> 0:03:39.920 because their risk models are telling them. 0:03:39.840 --> 0:03:43.840 Too totally and this is like a really important element 0:03:43.960 --> 0:03:48.240 for sort of understanding both how these investment vehicles work 0:03:48.280 --> 0:03:49.960 and the impact that they have on the market. Which 0:03:50.000 --> 0:03:52.040 is one of the things we know is that the 0:03:52.120 --> 0:03:56.440 various portfolio managers within these funds have very tight remits. 0:03:56.480 --> 0:04:00.680 It's like, your team is responsible for trading chip stocks, 0:04:00.720 --> 0:04:03.520 and your team is responsible for trading the short end 0:04:03.560 --> 0:04:07.320 of the Brazilian yield curve, and your team is responsible 0:04:07.480 --> 0:04:10.600 for international oil plays. And then we know that like 0:04:10.680 --> 0:04:12.880 and then you're not allowed to take any sector beta, 0:04:12.880 --> 0:04:14.920 and you're not allowed to take any market beta and 0:04:14.960 --> 0:04:17.359 all these things, and so you see fact you're neutral 0:04:17.560 --> 0:04:20.719 factor neutral, and then you know tight risk limits. So 0:04:20.720 --> 0:04:22.840 if something starts to go down, you don't want to 0:04:22.880 --> 0:04:25.479 lose your job, and you like get out of positions, 0:04:25.480 --> 0:04:28.120 and that can create interesting moves for the market. Anyway, 0:04:28.200 --> 0:04:29.880 suffice to say, there is much more to learn. 0:04:30.000 --> 0:04:32.720 Yes, well, the other thing, just one more thing, Yeah, 0:04:32.960 --> 0:04:34.839 the other other thing. The other other thing that I 0:04:34.839 --> 0:04:37.359 think is kind of funny now is remember whenever you 0:04:37.440 --> 0:04:40.680 had weird market moves. Yeah, like I guess it would 0:04:40.680 --> 0:04:42.960 have been fifteen years ago or something like that, it 0:04:43.000 --> 0:04:46.480 was always quant funds like the quant quake before two 0:04:46.520 --> 0:04:49.880 thousand and eight, and then it became CTA's and then 0:04:49.920 --> 0:04:53.200 it was risk parody, and now it's very much the 0:04:53.240 --> 0:04:56.040 pod shops that people point to when we start to 0:04:56.040 --> 0:04:58.239 see sketchiness in the market. So I think we should 0:04:58.240 --> 0:05:00.280 talk about, you know, what are the technicalities that are 0:05:00.360 --> 0:05:01.960 driving that pod shop behavior. 0:05:02.120 --> 0:05:04.800 Every time there's some big move in the market, someone 0:05:04.880 --> 0:05:06.200 tweets like, I hear a. 0:05:06.120 --> 0:05:08.440 Pod is blowing up. Yeah, Oh I hear some pods. 0:05:08.120 --> 0:05:10.400 Are blowing up. That's like, that's how to sound like 0:05:10.440 --> 0:05:12.520 an in guy on a finance twit. 0:05:13.440 --> 0:05:15.760 Little do they know the pod that's blowing up is off? 0:05:16.080 --> 0:05:18.520 If you don't a good one. If you don't know 0:05:18.560 --> 0:05:22.440 the pod that's blowing up, you're Anyway, we have the 0:05:22.440 --> 0:05:24.680 perfect guest. I'm very excited. We are going to be 0:05:24.720 --> 0:05:28.760 speaking with rich falk Wallace, who was previously a portfolio manager, 0:05:28.800 --> 0:05:32.040 who's at Citadel, who's at Viking, and now he is 0:05:32.040 --> 0:05:35.760 the CEO and co founder of Arcana, which builds models 0:05:35.760 --> 0:05:39.600 and software to help investors and hedge funds, et cetera 0:05:40.040 --> 0:05:43.159 actually track all of this stuff and actually track what 0:05:43.240 --> 0:05:46.960 kind of risks managers are taking and how they're actually 0:05:47.000 --> 0:05:50.960 performing relative to their benchmark or expectations. So we're gonna 0:05:50.960 --> 0:05:53.600 maybe understand a bit more of the technical aspects of 0:05:53.640 --> 0:05:55.880 all this stuff. So Rich, thank you so much for 0:05:55.960 --> 0:05:56.360 coming on. 0:05:56.279 --> 0:05:58.360 As thanks so much for having me. I appreciate it. 0:05:58.880 --> 0:06:00.720 Why don't we start with your best and obviously we're 0:06:00.720 --> 0:06:03.120 going to talk about your software company or Kenna and 0:06:03.200 --> 0:06:06.520 all that. But you were previously at a couple of 0:06:06.520 --> 0:06:08.720 these big funds. What did you do? 0:06:09.000 --> 0:06:11.080 Yeah, that's right. So I started my career on the 0:06:11.120 --> 0:06:14.120 buy side, started originally in investment banking out of college 0:06:14.279 --> 0:06:16.560 JP Morgan, and then worked at silver Point, which is 0:06:16.600 --> 0:06:19.720 like a large credit distressed hedge fund, very value oriented, 0:06:20.279 --> 0:06:22.240 none of that sort of risk model framework that gets 0:06:22.240 --> 0:06:24.279 deployed at the pods. And then after that was at 0:06:24.320 --> 0:06:27.120 Viking Global, which is I always describe the Tiger Cubs 0:06:27.120 --> 0:06:29.479 in some ways as like a hybrid between the sort 0:06:29.520 --> 0:06:33.920 of equity long short value orientation sort of philosophically and 0:06:33.960 --> 0:06:36.960 the multi manager systems. And then finally, most recently, was 0:06:37.000 --> 0:06:41.159 a portfolio manager at Citadel managed a global materials, natural 0:06:41.160 --> 0:06:42.920 resources and materials portfolio. 0:06:43.360 --> 0:06:46.320 I love this because when I think about silver Point, 0:06:46.440 --> 0:06:50.880 I think more sort of traditional value investing, and then 0:06:50.960 --> 0:06:54.680 you wind up doing metals at Citadel, which is a 0:06:54.760 --> 0:06:58.240 hedge fund that's known for being very quantitatively driven to 0:06:58.320 --> 0:07:01.880 better understand the pods. Now, talk to us about the 0:07:01.880 --> 0:07:06.000 differences between what you were doing at silver Point versus Citadel. 0:07:06.279 --> 0:07:09.400 Totally. Yeah, it's a great question. So the way that 0:07:09.560 --> 0:07:13.080 any value, super deep value oriented kind of fund works 0:07:13.160 --> 0:07:15.520 like a silver Point is that in the end, you 0:07:15.560 --> 0:07:18.160 do a ton of very deep research on the company, 0:07:18.240 --> 0:07:21.800 So you focus on what are the underlying fundamentals, what's 0:07:21.840 --> 0:07:25.240 the contract structure out many years in the future, what 0:07:25.240 --> 0:07:27.360 do the earnings look like, of course in the short term, 0:07:27.360 --> 0:07:30.920 but also in the long term, what's structurally happening competitively? 0:07:31.360 --> 0:07:33.040 You kind of go way down the rabbit hole. There's 0:07:33.040 --> 0:07:35.280 a lot more and we can go into that. And 0:07:35.320 --> 0:07:37.720 then as you kind of migrate sort of down the 0:07:38.040 --> 0:07:41.080 time horizon spectrum, at least from what a thesis looks 0:07:41.120 --> 0:07:43.960 like on a single stock, what you're kind of doing 0:07:44.080 --> 0:07:47.160 is thinking about where are the catalysts that change the 0:07:47.160 --> 0:07:50.920 market's perception of that long term. So, like I remember 0:07:50.920 --> 0:07:54.320 when I joined Viking, I remember asking the question just generally, 0:07:54.320 --> 0:07:56.240 like how much do you care about earnings. I think 0:07:56.240 --> 0:07:59.320 for anybody who's very value oriented, you kind of are 0:07:59.520 --> 0:08:01.880 concerned and about like, am I just going to be 0:08:01.880 --> 0:08:04.640 focused on the next data point, the next earnings and 0:08:04.640 --> 0:08:06.760 not sort of able to you know, see the forest 0:08:06.760 --> 0:08:08.480 for the trees and sort of care about, you know, 0:08:08.520 --> 0:08:10.880 what does this data mean for the long term? But 0:08:10.920 --> 0:08:14.160 the answer I got back in general, not specifically there, 0:08:14.160 --> 0:08:17.000 but is in that kind of framework it's or the 0:08:17.240 --> 0:08:19.960 way the question was answered to me was, Hey, the 0:08:20.040 --> 0:08:22.960 long term is a function a DCF is a function 0:08:23.000 --> 0:08:25.640 of years. Years are a function of quarters, and so 0:08:25.640 --> 0:08:27.920 therefore we care about the quarters. But what that tells 0:08:27.920 --> 0:08:30.680 you is that, like the answer is what about the 0:08:30.720 --> 0:08:33.839 short term, catalyst changes the perspective about the long term 0:08:33.920 --> 0:08:36.520 valuation of the company. And so I think what people 0:08:36.800 --> 0:08:39.120 sometimes looking from a far don't appreciate is the extent 0:08:39.160 --> 0:08:41.480 to which there's actually a little bit more of a 0:08:41.520 --> 0:08:46.080 convergence across styles from the underlying analyst workflow that like 0:08:46.200 --> 0:08:48.640 even a very long term investor to some extent is saying, 0:08:49.000 --> 0:08:51.760 even if I'm betting on the long term, the interim 0:08:51.840 --> 0:08:55.240 proof points illustrate the view of that long term and 0:08:55.280 --> 0:08:57.760 the short term guy says, well, I may get the 0:08:57.840 --> 0:08:59.599 number right in the short term, but that only is 0:08:59.679 --> 0:09:02.320 meaning to the change in the markets price if it 0:09:02.360 --> 0:09:04.120 tells you something about that long term. And so there's 0:09:04.120 --> 0:09:05.880 a little bit of like a yeah, and I think 0:09:05.920 --> 0:09:08.360 that convergence is happening more and more where people are 0:09:08.400 --> 0:09:12.520 kind of pushing towards that center actually, where everybody both 0:09:12.559 --> 0:09:15.560 cares about the short term data point and is looking 0:09:15.600 --> 0:09:18.440 to what that means about the long term. So but anyway, 0:09:18.480 --> 0:09:20.880 at the beginning of that process, at the silver point 0:09:20.920 --> 0:09:24.240 or any deep value type place, you're just really focused 0:09:24.240 --> 0:09:27.160 on that longer term story. You're less focused on the 0:09:27.320 --> 0:09:30.200 quarter or the catalyst than trying to understand sometimes things 0:09:30.200 --> 0:09:31.880 that And I was a junior analyst when I kind 0:09:31.880 --> 0:09:34.240 of started there. There was a first job out of banking, 0:09:34.440 --> 0:09:36.160 and you know, but you can be looking at like 0:09:36.200 --> 0:09:38.560 what does the rail contract look like in twenty twenty 0:09:38.559 --> 0:09:40.280 four and how does that step up? And you're like, man, 0:09:40.280 --> 0:09:43.080 does this matter to the stock. It's great training. It's 0:09:43.120 --> 0:09:44.719 a perfect place to kind of get that. 0:09:44.920 --> 0:09:45.040 You know. 0:09:45.040 --> 0:09:46.959 It's almost like private equity like where you're just sort 0:09:47.000 --> 0:09:49.880 of you know, looking through everything. But that's kind of 0:09:49.920 --> 0:09:51.319 how that started, all right. 0:09:51.400 --> 0:09:56.319 So then at Citadel you mentioned you covered materials, commodities, 0:09:56.320 --> 0:10:00.240 stuff like that. I guess two questions. When you come 0:10:00.240 --> 0:10:02.480 in the door there and you're told like, okay, this 0:10:02.520 --> 0:10:04.679 is what you do. What are you're told as your 0:10:04.720 --> 0:10:09.199 constraints and your specific remit and then also like how 0:10:09.200 --> 0:10:10.000 do you pick a stock? 0:10:10.240 --> 0:10:13.959 Yeah? Yeah, And I'll talk about this in a general sense, 0:10:14.000 --> 0:10:17.120 not specific set it up, but to talk about multimanagers 0:10:17.120 --> 0:10:19.559 in general, and our client base today at Arkana is 0:10:19.600 --> 0:10:21.920 about fifty to fifty split. I would say between people 0:10:21.920 --> 0:10:23.920 who I call like natives who come from the risk 0:10:23.960 --> 0:10:27.040 model system either any of the major pods or related 0:10:27.240 --> 0:10:28.800 and the other half can be like a deep value 0:10:28.800 --> 0:10:30.520 fund that says, hey, I don't want to limit myself 0:10:30.559 --> 0:10:32.800 to this stuff, but I see just like you guys 0:10:32.840 --> 0:10:35.680 are saying, this is an increasingly important part of markets 0:10:35.880 --> 0:10:37.959 and I want to be deep on it, educated whatever. 0:10:38.000 --> 0:10:41.520 So to answer your question on how multimanagers kind of pickstocks, 0:10:41.559 --> 0:10:44.880 run processes and think, not specific to any one place, 0:10:44.920 --> 0:10:48.480 but that sort of natives group in general. So at 0:10:48.480 --> 0:10:51.080 any of these places, the core contract is to say, 0:10:51.400 --> 0:10:53.160 the core difference, I guess is the other way to 0:10:53.160 --> 0:10:56.600 say it is versus a deep value place, it's about turnover. 0:10:56.960 --> 0:10:59.160 In the end, it's two things. It's risk limits and 0:10:59.200 --> 0:11:02.720 it's about how freak your book turns over. So at 0:11:02.800 --> 0:11:05.640 a deep value fund, the goal might be in sort 0:11:05.679 --> 0:11:07.560 of theory to have more than a year long average 0:11:07.559 --> 0:11:10.720 hold period. In practice it'll be often shorter than that, 0:11:10.760 --> 0:11:13.280 you know, nine months or whatever as bad idea cycloud 0:11:13.360 --> 0:11:16.000 or whatever. But at a multi manager those numbers can 0:11:16.000 --> 0:11:18.400 be anywhere from like ten to fifteen to even higher, 0:11:18.720 --> 0:11:21.360 meaning the entire book turns over ten to fifteen times 0:11:21.400 --> 0:11:22.439 in a year. 0:11:22.520 --> 0:11:22.720 Wow. 0:11:22.920 --> 0:11:25.520 Think about it simply like the average idea stays on 0:11:25.559 --> 0:11:27.120 for a month is the way to put it in 0:11:27.160 --> 0:11:29.719 the book. And so as you step into any of 0:11:29.760 --> 0:11:32.120 these places to your point A, you have a you know, 0:11:32.120 --> 0:11:34.640 there's a structure. There's an analyst, and then a portfolio manager. 0:11:34.880 --> 0:11:37.720 And the analyst generally has a single industry focused so 0:11:37.720 --> 0:11:39.520 it's like, hey, i am, as you said, chip stocks, 0:11:39.600 --> 0:11:42.559 or you know, somebody else might have software, or it's 0:11:42.559 --> 0:11:45.240 a sort of defined single universe. And then a portfolio 0:11:45.240 --> 0:11:47.600 manager will have a set of analysts below them who 0:11:47.679 --> 0:11:50.960 have typically very related coverage universes and will feed up 0:11:50.960 --> 0:11:53.400 into the portfolio manager. So that's like kind of the structure. 0:11:53.960 --> 0:11:57.080 Stockpicking kind of ends up being what we were talking 0:11:57.080 --> 0:11:59.719 about earlier. In the end, it's the analyst job to 0:12:00.640 --> 0:12:03.960 have a detailed model, of course, to have a view 0:12:04.000 --> 0:12:07.400 on earnings across their coverage universe, and that coverage universe 0:12:07.400 --> 0:12:09.120 for that analyst, by the way, can be and it 0:12:09.160 --> 0:12:11.040 varies by a multi manager, but it can be anywhere 0:12:11.080 --> 0:12:13.320 from like thirty names at the low end to like 0:12:13.480 --> 0:12:16.280 eighty names at the high end buy analyst. So there's 0:12:16.280 --> 0:12:17.199 a lot of process there. 0:12:32.480 --> 0:12:35.800 There are many differences between a retail investor and a 0:12:35.920 --> 0:12:38.959 multi strategy fund, but one of the key ones I 0:12:39.000 --> 0:12:43.760 think is maybe position sizing. So if you're a retail 0:12:43.760 --> 0:12:47.160 investor and you have a single stock thesis, I don't know, 0:12:47.200 --> 0:12:49.640 you want to buy in video or something, you buy 0:12:49.640 --> 0:12:52.880 in video, and you're probably making that decision based on 0:12:53.080 --> 0:12:56.520 how much cash you have in your Robinhood account or 0:12:56.559 --> 0:12:59.760 something like that. But if you're at a multi fund, 0:13:00.080 --> 0:13:03.560 it seems like a much more sophisticated process. So I 0:13:03.559 --> 0:13:06.680 guess I'm curious if you're at a podshop, how do 0:13:06.760 --> 0:13:09.760 you know how much to buy? How do you know 0:13:09.880 --> 0:13:13.960 how much to allocate to a single stock. And I 0:13:13.960 --> 0:13:17.280 guess another way of saying it is you're looking at 0:13:17.280 --> 0:13:20.120 that single stock on a risk adjusted basis, right, Like, 0:13:20.200 --> 0:13:23.079 that's what you want to get, right, the risk adjusted performance, 0:13:23.200 --> 0:13:25.040 not just the single stock performance. 0:13:25.120 --> 0:13:27.960 That's right. That's right. So there are two or three 0:13:28.000 --> 0:13:31.880 ways that gets implemented. So the first is constraints. So 0:13:32.280 --> 0:13:35.040 step one is dollar neutrality. I'm long as many dollars 0:13:35.080 --> 0:13:37.280 as i'm short. That's a simple limit, sort of. One 0:13:37.320 --> 0:13:40.000 level higher is beta neutrality relative to the overall market? 0:13:40.040 --> 0:13:42.760 Am I longer short? On a beta adjusted basis? Sort of? 0:13:42.760 --> 0:13:46.400 The third level is factor neutrality. I'm balanced against all 0:13:46.400 --> 0:13:49.000 of these sort of if you maybe simplified just slightly 0:13:49.040 --> 0:13:51.880 the subcomponents of beta, so instead of like, hey, I 0:13:51.920 --> 0:13:53.560 have a beta to the market, I actually have a 0:13:53.600 --> 0:13:56.599 beta to the basket of size large companies. I have 0:13:56.640 --> 0:13:59.000 a beta to the basket of companies with momentum. I 0:13:59.000 --> 0:14:00.640 have a basket to the beta. 0:14:00.960 --> 0:14:01.240 I see. 0:14:01.320 --> 0:14:03.000 So you decompose beta. 0:14:02.880 --> 0:14:07.120 Essentially, it's a decomposition. Essentially, when people talk about factors 0:14:07.120 --> 0:14:10.880 and factor neutrality, it's a decomposition of beta into its 0:14:10.880 --> 0:14:13.920 constituent parts. There's a lot of statistics that goes under 0:14:13.920 --> 0:14:17.199 the hood to make that orthogonal and precise and. 0:14:18.559 --> 0:14:19.160 Orthogonal. 0:14:19.720 --> 0:14:21.800 But at the sort of functional level, at the level 0:14:21.800 --> 0:14:24.680 that people at the stock picking level, at multi managers 0:14:24.720 --> 0:14:28.040 interact with the model, it's essentially just a decomposition of 0:14:28.320 --> 0:14:32.480 beta's and then you add up those exposures on each side, 0:14:32.560 --> 0:14:35.280 and you are limited essentially by the percent of your 0:14:35.360 --> 0:14:39.119 bets in a book in aggregate that are betting basically 0:14:39.280 --> 0:14:43.200 on factor type bets as compared to the percentage of 0:14:43.200 --> 0:14:45.760 your bets that are betting on the remainder term, the 0:14:45.840 --> 0:14:48.800 non factor component of any stock. So as you look 0:14:48.840 --> 0:14:51.600 at any stock, it fits within that broader portfolio that 0:14:51.640 --> 0:14:52.680 you're putting together. 0:14:53.560 --> 0:14:56.720 Okay, and then the second thing that you talked about 0:14:56.760 --> 0:15:01.600 earlier is this idea of turnover. So just to press 0:15:01.600 --> 0:15:05.560 on this point, how much do trading costs factor into 0:15:05.800 --> 0:15:10.440 investment decisions? And also position sizing, because as you just stated, 0:15:10.880 --> 0:15:15.080 you could theoretically size or arrange all of your positions 0:15:15.120 --> 0:15:19.080 to be factor neutral or neutral in terms of systematic risk. 0:15:19.160 --> 0:15:22.840 I guess, but I imagine in order to do that, you 0:15:22.880 --> 0:15:25.960 would have to be trading pretty much like constantly right, 0:15:26.000 --> 0:15:30.280 which would add to your execution costs. So does that 0:15:30.360 --> 0:15:31.400 come into play as well? 0:15:31.520 --> 0:15:36.000 Yeah, it does. In practice, the stock picker, portfolio manager 0:15:36.000 --> 0:15:40.239 and analyst doesn't flow in a complicated set of formulas 0:15:40.280 --> 0:15:43.200 to their decision around sort of how do I optimize 0:15:43.200 --> 0:15:47.040 trading costs? The engines operating at the multi managers do 0:15:47.120 --> 0:15:50.320 think a ton about how do I take the stock 0:15:50.360 --> 0:15:53.280 picks that a single portfolio does and then execute them 0:15:53.720 --> 0:15:56.600 in a optimal way a crossing some firms doing some 0:15:56.680 --> 0:15:59.240 prints don't cross each other's orders within the pod level 0:16:00.080 --> 0:16:02.600 about all of that, and then so the first level 0:16:02.640 --> 0:16:04.840 of how do people get limited is the constraints on 0:16:04.840 --> 0:16:06.640 what percent of my bets are in factor type bets 0:16:06.680 --> 0:16:09.160 versus non factor type bets. There are all also a 0:16:09.160 --> 0:16:11.880 bunch of like single position limits, So that's like one 0:16:12.080 --> 0:16:15.200 version is basically limiting the portfolio manager to have to 0:16:15.240 --> 0:16:19.280 sort of live pickstocks under this constraint. The other framework 0:16:19.320 --> 0:16:21.120 of how do you size positions to your sort of 0:16:21.160 --> 0:16:23.440 earlier question which comes around to this trading cost question 0:16:23.560 --> 0:16:26.600 is there are tools that are called optimizers that basically 0:16:27.280 --> 0:16:30.120 look at the expected return that each portfolio manager thinks 0:16:30.120 --> 0:16:32.400 they have in their book of stocks and tries to 0:16:32.520 --> 0:16:36.400 solve for the optimal balance of the expected return against 0:16:36.440 --> 0:16:39.160 the volatility of those stocks and the volatility the factor 0:16:39.200 --> 0:16:41.320 bets in the book, and it'll spit out an answer 0:16:41.360 --> 0:16:43.040 for you. That answer may not be exactly where you 0:16:43.080 --> 0:16:46.600 want to land, but in the most sophisticated places, that 0:16:46.800 --> 0:16:50.160 answer that the optimizer spitting out is including how much 0:16:50.280 --> 0:16:52.920 trading cost impacts the book. So it's sort of flowing 0:16:52.960 --> 0:16:57.680 that mathematically into a machine driven optimal book. But again 0:16:57.800 --> 0:16:59.720 that's sort of in the more science bucket. Of course 0:16:59.760 --> 0:17:02.360 there's art even underneath that statistics, but basically that's in 0:17:02.400 --> 0:17:05.200 the more sort of science bucket. Then the PORTFOLI manaddressed 0:17:05.200 --> 0:17:07.439 to say, Okay, the machine sort of took my expected returns, 0:17:07.480 --> 0:17:09.560 took the variance of those pieces and the trading costs 0:17:09.600 --> 0:17:11.639 into account. I gave me an answer. Does that actually 0:17:11.640 --> 0:17:14.400 still fit with my you know, fundamental bottoms at work? 0:17:14.440 --> 0:17:16.560 Back to hey, the contract of this company changes in 0:17:16.560 --> 0:17:18.760 twenty twy six, the earnings going to be this. Here's 0:17:18.760 --> 0:17:21.040 the positioning and set up and crowding of other pods, 0:17:21.040 --> 0:17:23.560 you know, playing the game you mentioned earlier. So there's 0:17:23.600 --> 0:17:25.800 then the sort of second level of art that goes 0:17:25.800 --> 0:17:26.439 to the top of that. 0:17:26.600 --> 0:17:29.680 Yeah, So I want to talk more about the speed 0:17:29.880 --> 0:17:33.440 of turnover because Okay, let's say you're like bullish on 0:17:33.520 --> 0:17:36.320 and video and videos had this big run and you're like, 0:17:36.359 --> 0:17:39.119 all right, but I don't want to have size exposure 0:17:39.240 --> 0:17:41.359 because it's going to be correlated to big caps. I 0:17:41.359 --> 0:17:44.760 don't want to have general market beta because probably if 0:17:44.760 --> 0:17:47.040 the market goes up and video is going to go up, 0:17:47.040 --> 0:17:49.040 and I don't have chip beta and all this stuff. 0:17:49.200 --> 0:17:52.240 So what you're trying to identify is just the Invidia 0:17:52.359 --> 0:17:56.199 specific idiosyncredit. That's exactly right, right, But why does that 0:17:56.320 --> 0:18:00.440 inherently lend itself when you're thinking about I mean, I 0:18:00.440 --> 0:18:02.520 feel like there must be some connection, But you're trying 0:18:02.560 --> 0:18:04.760 to strip out all of these different factors that you 0:18:04.800 --> 0:18:06.760 don't want to have exposure to. You're trying to find 0:18:06.800 --> 0:18:10.560 the idiosyncratic drivers of a specific name. What is it 0:18:10.640 --> 0:18:14.520 about that process that sort of inherently lends itself to 0:18:14.640 --> 0:18:15.640 shorthold periods. 0:18:15.720 --> 0:18:17.720 That's a great question, and as sort of deep one, 0:18:18.000 --> 0:18:19.840 and you might get different answers to that question from 0:18:19.840 --> 0:18:23.000 a few different people. I'll give you mine. The essential 0:18:23.240 --> 0:18:27.000 reality is that In order for this entire model to work, 0:18:27.560 --> 0:18:31.560 you have to have a great deal of diversification across 0:18:31.640 --> 0:18:36.080 idiosyncratic bets non factor bets. And the way to think 0:18:36.119 --> 0:18:38.720 about that is the core reason a lot of these 0:18:38.720 --> 0:18:41.840 models work, is that the residual return or idiosyncratic return 0:18:42.520 --> 0:18:47.360 is approximately normally distributed across a certain window. Meaning it's 0:18:47.400 --> 0:18:50.199 sort of, you know, like flipping a coin basically, And 0:18:50.240 --> 0:18:53.119 the intuition is, if you flip a thousand coins, obviously 0:18:53.280 --> 0:18:56.760 you'll center around whatever your hit rate is on that coin. 0:18:56.800 --> 0:18:59.280 If the coin is loaded fifty two percent, yeah, versus 0:18:59.320 --> 0:19:01.879 fifty As you flip three coins, it could be you know, 0:19:01.920 --> 0:19:04.080 the mean, the expected value of that is going to 0:19:04.080 --> 0:19:06.000 be you know, who knows, right, But as you flip 0:19:06.040 --> 0:19:08.919 a thousand coins or ten thousand coins, you will center 0:19:08.960 --> 0:19:12.640 around that mean. And so and that variance is effectively, 0:19:13.080 --> 0:19:15.200 if you think about things from a return standpoint, the 0:19:15.240 --> 0:19:17.200 sharp ratio, right is they're returned about it about the 0:19:17.280 --> 0:19:20.880 variance of the volatility of that return. And so as 0:19:20.920 --> 0:19:23.639 you have more and more bets, you shrink the variants 0:19:23.800 --> 0:19:25.800 relative to the return you're generating. And the more and 0:19:25.800 --> 0:19:29.359 more your bets are in idiosyncratic bets which are normally 0:19:29.359 --> 0:19:32.239 distributed unlike market bets you know, which are you know 0:19:32.240 --> 0:19:32.919 can be wild. 0:19:33.040 --> 0:19:35.760 Right, Wait, can you actually just explain that point, because 0:19:35.800 --> 0:19:39.360 that's a great answer. You're basically you have some assumption 0:19:39.520 --> 0:19:42.120 about returns, but there's going to be a lot of variants, 0:19:42.280 --> 0:19:44.080 so you want to make a lot of bets. That's 0:19:44.160 --> 0:19:47.800 exactly in order to achieve that. Why is it that 0:19:47.960 --> 0:19:53.719