WEBVTT - Re-Run: Cliff Asness 0:00:02.720 --> 0:00:07.200 Bloomberg Audio Studios, podcasts, radio news. 0:00:08.160 --> 0:00:09.680 I was hoping to be like one of those like 0:00:09.680 --> 0:00:11.160 clips on TikTok you see. 0:00:11.320 --> 0:00:13.280 Fake that's the thing you look like you're on a 0:00:13.280 --> 0:00:14.760 podcast pads on home. 0:00:14.880 --> 0:00:16.680 Now it just looks like, you know, instead. 0:00:16.400 --> 0:00:20.200 Of we just randomly got together the chat and microphones 0:00:20.680 --> 0:00:23.720 part of it with Bloomberg Podcasts behind us. 0:00:24.079 --> 0:00:26.960 Yeah. Right, I guess we've conveyed podcasts efficiently with all 0:00:27.000 --> 0:00:28.520 of the Bloomberg podcasts. 0:00:28.600 --> 0:00:28.800 Joe. 0:00:29.040 --> 0:00:30.680 Part of the reason this has to be on video 0:00:31.000 --> 0:00:33.839 is because Matt shaved. Matt has had a beard for 0:00:33.960 --> 0:00:35.000 the past I don't know that. 0:00:35.120 --> 0:00:39.960 I've had a winter beard from like Christmas break through memorild. 0:00:39.600 --> 0:00:41.960 That I shaved over COVID for the first time in 0:00:41.960 --> 0:00:44.760 about thirty years and my kids freaked out. 0:00:45.080 --> 0:00:48.400 Yeah, they were like, you're an alien. Yeah, my kids 0:00:48.440 --> 0:00:50.400 didn't care that much. But one of my sons said, 0:00:50.760 --> 0:00:51.800 it's a different daddy. 0:00:52.120 --> 0:00:55.280 Well, you also have hair on the top of your head. 0:00:55.760 --> 0:00:57.880 When you shave the beard and you don't have any hair, 0:00:58.120 --> 0:00:59.320 Suddenly you're mister clean. 0:00:59.480 --> 0:01:01.000 You just does mister clean? 0:01:02.000 --> 0:01:05.639 No, I don't think so, mister clean. You know clean? 0:01:06.400 --> 0:01:09.600 Come on, I think that's a fair point. 0:01:13.280 --> 0:01:16.520 It's the Money Stuff podcast. We have a guest, Cliff 0:01:16.560 --> 0:01:21.520 has runs AQR. Thanks for coming in, Thanks for having me. 0:01:22.200 --> 0:01:23.920 I always like to ask how to my managers, like 0:01:25.200 --> 0:01:26.760 what do you do for a living? Like what is 0:01:26.760 --> 0:01:30.199 it like economic function of like the business that you run? 0:01:30.600 --> 0:01:34.080 Okay, those are to me slightly different questions. Right. 0:01:34.080 --> 0:01:35.680 One sounds like it's about you. One sounds like it's 0:01:35.680 --> 0:01:36.240 about AQR. 0:01:36.280 --> 0:01:39.520 I'm more interested in even if it's about a q R. 0:01:40.160 --> 0:01:43.600 What you do for a living is And people might 0:01:43.600 --> 0:01:47.200 not like this phraseology, but you're trying to predict what 0:01:47.240 --> 0:01:50.000 happens to securities. You're trying to buy ones that go 0:01:50.240 --> 0:01:52.840 up either in the absolute or more than some benchmark, 0:01:52.880 --> 0:01:57.320 and and sell ones that do the opposite. The broader question, 0:01:57.320 --> 0:02:00.800 which I think is behind what you're saying, is what 0:02:00.840 --> 0:02:03.360 do you do for the world by doing that? And 0:02:03.440 --> 0:02:06.520 they overlap, but they're not exactly the same thing. Something 0:02:06.520 --> 0:02:08.640 can have a net positive effect on the world even 0:02:08.639 --> 0:02:11.440 if you're not waking up. And I'll admit this, I 0:02:11.480 --> 0:02:13.840 don't think most people in their jobs waked up and 0:02:13.880 --> 0:02:17.360 always think that way, and certainly not active managers just 0:02:17.440 --> 0:02:19.840 go I am just making the world a better place. 0:02:19.880 --> 0:02:23.720 They're thinking, is in Nvidia undervalued? Is it overvalued? The 0:02:23.800 --> 0:02:26.440 things that I think you do for the world is 0:02:27.000 --> 0:02:32.560 first take the other side of positions other people disagree 0:02:32.600 --> 0:02:35.399 with you on or don't want to bear. That can 0:02:35.440 --> 0:02:39.640 take two forms. That can mean one of you is 0:02:40.080 --> 0:02:43.800 biased and wrong. You hope it's not you. But in 0:02:43.840 --> 0:02:45.520 that case, what you do for the world is you 0:02:45.560 --> 0:02:48.280 move prices back towards not necessarily all the way too 0:02:49.000 --> 0:02:52.400 in the abstract, the correct price. That's hard to define, actually, 0:02:52.480 --> 0:02:54.960 but something is mispriced and you take the other side 0:02:54.960 --> 0:02:57.119 of that. It could be a risk premium. Other people 0:02:57.160 --> 0:03:00.600 don't want to bear in a lot of strategies necessarily 0:03:00.600 --> 0:03:03.400 that they're making an error. If a merger's announced and 0:03:03.639 --> 0:03:05.880 three quarters of the pop you should get if it 0:03:05.919 --> 0:03:08.200 closes happens, a lot of people might not want to 0:03:08.200 --> 0:03:10.919 stick around for that last quarter. And if you're willing 0:03:11.000 --> 0:03:12.400 to take the other side of that, maybe you get 0:03:12.440 --> 0:03:14.639 paid a little for doing that, and that is a 0:03:14.720 --> 0:03:16.680 service to the market. People want to get out and 0:03:16.720 --> 0:03:22.640 you're helping. That's kind of the positive side. But again 0:03:22.760 --> 0:03:24.760 it's not what you're thinking about when you do the trade. 0:03:25.040 --> 0:03:29.600 There are positions active managers will take that are not 0:03:29.720 --> 0:03:31.880 about that. Let me put it this way. If what 0:03:31.919 --> 0:03:35.440 you're trying to do is predict returns, you can predict 0:03:35.440 --> 0:03:40.320 returns because the price is moving towards truth. But you 0:03:40.320 --> 0:03:42.760 can also make money if you predict the price moves 0:03:42.760 --> 0:03:46.440 further away from truth. You know, if you're a momentum 0:03:46.480 --> 0:03:49.440 memestock investor. And doesn't mean you can't get that right, 0:03:49.520 --> 0:03:50.720 you know. I think of that a little more as 0:03:50.760 --> 0:03:54.000 trading than investing. But they all come together, and it 0:03:54.120 --> 0:03:58.160 even gets complicated within some famous quantitative factors. One famous 0:03:58.200 --> 0:03:59.840 quant factor is the momentum factor. 0:04:00.480 --> 0:04:03.520 Asked the finance professor, I should I have asked? And 0:04:03.560 --> 0:04:07.240 he said, you should ask him if momentum trading makes 0:04:07.240 --> 0:04:08.560 markets more or less efficient. 0:04:08.720 --> 0:04:11.000 We don't fully know, but I can tell you the framework. 0:04:11.640 --> 0:04:16.600 There are two competing explanations in academia and a general 0:04:16.760 --> 0:04:19.680 world that cares about these things. For why momentum on 0:04:19.720 --> 0:04:22.760 average works, there's always a third that it will never 0:04:22.800 --> 0:04:24.400 work again and it was just random and it was 0:04:24.480 --> 0:04:26.920 just luck. But if it's true, why does it work? 0:04:27.240 --> 0:04:29.360 I actually think these can both coexist, so it's not 0:04:29.400 --> 0:04:32.960 truly embarrassing, but it sounds embarrassing that the two major 0:04:33.000 --> 0:04:37.960 explanations one is under reaction and the other is over reaction. 0:04:38.360 --> 0:04:40.120 When you've narrowed it down to two things that at 0:04:40.200 --> 0:04:42.479 least feel like the opposites, you should feel a little shame. 0:04:42.520 --> 0:04:45.279 For a second, there's the inituation, just like there's a 0:04:45.279 --> 0:04:48.640 correct price, and momentum is trading from below to above the. 0:04:48.560 --> 0:04:52.000 Credit friends, Well, I'll give you two scenarios. Information comes 0:04:52.040 --> 0:04:56.359 out and people have a behavioral bias that behavioral psychologists 0:04:56.400 --> 0:05:00.320 would call anchoring an adjustment. They move towards all the 0:05:00.360 --> 0:05:02.840 way to the new information, and I think that fits 0:05:02.880 --> 0:05:05.800 a lot of intuition in the short run that can 0:05:05.880 --> 0:05:10.240 make momentum work. If you're following fundamentals or prices, good 0:05:10.240 --> 0:05:13.680 news comes out or the price moves. If good news 0:05:13.680 --> 0:05:17.159 comes out, on average, the price goes up, but not enough. 0:05:17.800 --> 0:05:21.039 If the price moves, it may on average be responding 0:05:21.040 --> 0:05:23.320 to good news, and simply by observing the price move, 0:05:24.120 --> 0:05:27.480 you can say, Okay, sometimes it's wrong, but on average 0:05:27.520 --> 0:05:30.479 it doesn't quite move enough. If that's the reason, and 0:05:30.520 --> 0:05:33.279 I think the weight of the opinion in academia I 0:05:33.320 --> 0:05:37.120 believe is towards this underreaction explanation. Then, even though you're 0:05:37.160 --> 0:05:40.400 trading on momentum, you're still moving the price towards kind 0:05:40.400 --> 0:05:43.640 of truth or equilibrium or something. But the flip side 0:05:43.680 --> 0:05:46.719 is overreaction. Do you think of that more as just 0:05:46.800 --> 0:05:50.840 your classic positive feedback loop. Someone's buying something just for 0:05:51.120 --> 0:05:53.720 you know, fomo. It's been going up in there, and 0:05:54.080 --> 0:05:55.480 well that could be a negative reason, or they're just 0:05:55.560 --> 0:05:59.159 predicting more people buy it because of fomo. In that sense, 0:05:59.360 --> 0:06:02.320 if you buy some weird meme coin, you could do 0:06:02.360 --> 0:06:04.440 that for a rational reason, not that you are a 0:06:04.480 --> 0:06:07.559 long term holder, but you just believe it's gonna keep going. 0:06:07.839 --> 0:06:09.440 So did you make money on GameStop? 0:06:10.680 --> 0:06:12.719 This the God's honest truth. You won't believe me. I 0:06:12.760 --> 0:06:14.400 have no idea if we were long or short game 0:06:14.440 --> 0:06:15.520 stop during the whole thing. 0:06:15.680 --> 0:06:16.920 You never went back and check. 0:06:17.080 --> 0:06:19.520 No, I never did. I did have an episode where 0:06:19.560 --> 0:06:22.520 I mentioned on perhaps a different TV network that we 0:06:22.520 --> 0:06:23.479 were short, AMC. 0:06:23.960 --> 0:06:25.520 What does your Twitter look like after that? 0:06:25.600 --> 0:06:28.080 It was very ugly. Yeah, I certainly knew of that world, 0:06:28.120 --> 0:06:30.880 even though we're quants. I watch the markets all day, 0:06:30.920 --> 0:06:33.279 even even if I don't do anything about it. It's 0:06:33.320 --> 0:06:35.120 like the old joke about the weather. Everyone talks about it, 0:06:35.160 --> 0:06:36.360 but nobody does anything about it. 0:06:36.400 --> 0:06:37.400 Best thing to talk about. 0:06:37.720 --> 0:06:40.640 So it's not like everything that went on with GameStop 0:06:40.720 --> 0:06:44.000 Melvin Capitol. You know, I'm watching it every day, but 0:06:44.279 --> 0:06:47.839 we take relevant tiny positions in every stock. There was 0:06:47.839 --> 0:06:50.160 nothing weird in our P and L. And yes, I 0:06:50.200 --> 0:06:53.160 was not even curious. It probably wasn't even in our 0:06:53.279 --> 0:06:56.000 universe at that point of things. We trade. But then 0:06:56.040 --> 0:07:02.720 I'm going on this other network. It's allowed, and in 0:07:02.960 --> 0:07:04.400 kind of a pre call of what are we going 0:07:04.480 --> 0:07:06.000 to talk about? You guys know, you don't want to 0:07:06.000 --> 0:07:08.159 get on there and have absolutely nothing to talk about. 0:07:08.240 --> 0:07:10.720 You want to have some not necessarily the answers worked out, 0:07:10.720 --> 0:07:13.560 but agreed upon topics. They're like everyone in this segment 0:07:13.640 --> 0:07:16.680 gives us some lungs and shorts. But I'm saying, as 0:07:16.720 --> 0:07:19.880 a quant that's kind of silly. They're not indicative, and 0:07:19.920 --> 0:07:22.880 we kind of made a deal where they'd let me 0:07:23.000 --> 0:07:26.880 briefly explain that doesn't make a whole lot of sense 0:07:26.880 --> 0:07:29.440 for quantum, but it might be fun. And my way 0:07:29.480 --> 0:07:30.880 of saying it was, if I give you a few 0:07:30.960 --> 0:07:33.560 names long and a few names short, you could look 0:07:33.560 --> 0:07:35.640 in six months later and think we had a fantastic 0:07:35.720 --> 0:07:38.200 year or a terrible year and be terribly wrong in 0:07:38.240 --> 0:07:41.440 either direction because they're tiny. So I went through an 0:07:41.480 --> 0:07:45.680 AMC was I think I'm accidentally doing it again. We'll 0:07:45.720 --> 0:07:49.640 keep going hopefully they don't listen to you, guys. But 0:07:51.080 --> 0:07:54.440 it was bad on every single thing in our model practically, 0:07:54.440 --> 0:07:58.080 which is hard to do. It was expensive, unprofitable, high beta. 0:07:58.120 --> 0:08:01.880 They were issuing shares, not buying backshit. There are more examples, 0:08:02.120 --> 0:08:04.240 and so I said that, but then I added that 0:08:04.600 --> 0:08:09.000 were only short twelve basis points, so the crazy people 0:08:09.000 --> 0:08:13.000 could be right and it doesn't really matter. I discovered 0:08:13.000 --> 0:08:15.160 two things. They're not going to like a short period, 0:08:15.480 --> 0:08:19.000 and crazy people don't always like being called crazy. Yeah, 0:08:19.040 --> 0:08:21.239 I had to discover that from myself. So my Twitter 0:08:21.280 --> 0:08:23.600 got ugly for a while. You may have noticed. I've 0:08:23.600 --> 0:08:26.120 gotten I think a fair amount better at this, but 0:08:26.240 --> 0:08:28.760 I used to be pretty bad about responding to ugly, 0:08:29.400 --> 0:08:30.920 which you learn your lesson on that. 0:08:30.920 --> 0:08:33.240 You always feed the trolls less so than. 0:08:33.160 --> 0:08:33.480 I used to. 0:08:33.840 --> 0:08:36.600 At least I think I'm better. Maybe I'm wrong, but 0:08:37.120 --> 0:08:39.880 I became public enemy number three to the meme stock 0:08:39.960 --> 0:08:44.000 crowd for a while, I did not really. Yes, Kenny 0:08:44.040 --> 0:08:46.880 Ken Griffin was number one. I don't believe Ken did this, 0:08:47.000 --> 0:08:50.920 but it's the whole poll, the Bible thing. And oddly enough, 0:08:50.920 --> 0:08:54.040 Gary Gensler was clearly number two, Yeah, because they thought 0:08:54.080 --> 0:08:58.160 he was covering for the manipulators and the naked shorts 0:08:58.200 --> 0:09:00.400 and whatnot. I've met both, I know, and a little 0:09:00.400 --> 0:09:01.959 better than Gary. I don't think there are any two 0:09:02.000 --> 0:09:05.720 people on Earth less likely to be in cahoots than 0:09:05.960 --> 0:09:07.960 those two. I think they're on the opposite side of 0:09:08.000 --> 0:09:10.960 most issues. But that was the theory. But both Ken 0:09:11.000 --> 0:09:13.760 and Gary are too smart to respond to them on Twitter, 0:09:13.800 --> 0:09:17.120 so I certainly became the most actively engaged. And then 0:09:17.240 --> 0:09:20.640 I never did that again before today, when I've accidentally 0:09:20.640 --> 0:09:21.000 done it. 0:09:21.120 --> 0:09:22.959 Before we get back to stuff that matters, can I 0:09:23.000 --> 0:09:26.520 just say on AMC, I tweeted, when did June two 0:09:26.559 --> 0:09:26.960 come out? 0:09:27.080 --> 0:09:27.920 It was like last summer. 0:09:28.120 --> 0:09:28.720 Yeah, year ago. 0:09:28.800 --> 0:09:31.439 I tweeted that I fell asleep during Dune two, and 0:09:31.600 --> 0:09:34.560 it reawakened that crowd, at least on my Twitter, because 0:09:34.600 --> 0:09:35.839 I've the same. 0:09:35.640 --> 0:09:41.080 Crowds crowd because he's dissing a movie, don't you understand 0:09:41.120 --> 0:09:45.000 you have to like every movie or else. You're anti America. 0:09:45.240 --> 0:09:48.000 And then there were a lot of conspiracy theories about 0:09:48.040 --> 0:09:52.000 Bloomberg reporter lies about falling asleep in doom too because 0:09:52.000 --> 0:09:52.920 she hates AMC. 0:09:53.280 --> 0:09:54.800 It would be, but it wasn't. 0:09:55.760 --> 0:09:59.520 Oh yeah, you do innocuous that I was obnoxious, so 0:09:59.559 --> 0:10:02.040 I'd kind of deserved it. You didn't deserve Thank you 0:10:02.120 --> 0:10:03.600 for saying that you didn't deserve it. 0:10:03.679 --> 0:10:05.520 I haven't sent. That's not doing what I was good. 0:10:05.880 --> 0:10:07.520 That's all too. It's pretty good. 0:10:07.760 --> 0:10:10.560 No you didn't, Actually I didn't. You were asleep. 0:10:10.960 --> 0:10:13.000 Well no, not the whole time, not the whole. 0:10:12.760 --> 0:10:15.679 Time, but also like you and and they were like, 0:10:15.800 --> 0:10:19.000 give me some shorts, and you gave them some lungs 0:10:19.040 --> 0:10:21.720 and shorts. Would you have known that or did you 0:10:21.760 --> 0:10:22.880 have to be like I got it. 0:10:22.960 --> 0:10:24.960 I would not have known that, right, So you don't know, 0:10:25.000 --> 0:10:26.560 you're ar It would have been better for me if 0:10:26.600 --> 0:10:28.559 we didn't have the call, because then I could have 0:10:28.679 --> 0:10:32.720 just said I don't know. I rarely know individual stocks, 0:10:33.120 --> 0:10:35.400 and if I do know, I'm probably not happy I know. 0:10:35.960 --> 0:10:38.120 And even then, it's like we lost twenty BIPs on 0:10:38.160 --> 0:10:40.000 that today, which is a giant number for us to 0:10:40.040 --> 0:10:44.000 lose on one stock, and even then I probably don't notice. 0:10:43.520 --> 0:10:45.240 Verse twenty bit. If someone comes to you and says 0:10:45.880 --> 0:10:46.200 that's not. 0:10:46.280 --> 0:10:49.520 One, I might be told about it. For us, it's 0:10:49.559 --> 0:10:52.280 whether these seven hundred fifty stocks be these seven hundred 0:10:52.280 --> 0:10:56.120 and fifty stocks. Yeah, I don't memorize fifteen hundred stocks. Now, 0:10:56.160 --> 0:10:58.880 we take a fair amount of risks, some funds more, 0:10:58.960 --> 0:11:01.640 some funds designed to be less. And that's about the 0:11:01.679 --> 0:11:05.520 size of these two positions. So when I say small, 0:11:05.559 --> 0:11:07.920 I'm not saying we're not both taking risk and trying 0:11:07.960 --> 0:11:10.920 to generate pretty decent returns. But if you just think 0:11:10.920 --> 0:11:13.439 about it, a quant is playing the odds. They're saying, 0:11:13.440 --> 0:11:17.679 affirm a company with these characteristics. And this can be 0:11:17.960 --> 0:11:20.840 old school factor quants from the nineteen nineties, these can 0:11:20.880 --> 0:11:24.040 be modern machine learning. But with these characteristics tend to 0:11:24.040 --> 0:11:27.200 beat these characteristics. If that's all you know, and it 0:11:27.240 --> 0:11:29.880 is all we know, why on earth would you take 0:11:29.880 --> 0:11:32.360 a lot of risk in any one company. AMC really 0:11:32.360 --> 0:11:33.960 could have done well. It could have been bad on 0:11:34.040 --> 0:11:37.280 every single thing that on average doesn't work, and it 0:11:37.280 --> 0:11:40.400 could be a special situation that we don't understand. Something 0:11:40.400 --> 0:11:43.560 could be good on everything and the CEO can embezzle 0:11:43.600 --> 0:11:46.120 all the money. We don't want to take a lot 0:11:46.160 --> 0:11:48.160 of risk on any one thing because we have no 0:11:48.200 --> 0:11:49.920 insight in that it's risk for no return. 0:11:50.320 --> 0:11:53.000 What thing you've written is that like over time, the 0:11:53.400 --> 0:11:56.439 quant like factor model has moved closer to being what 0:11:56.559 --> 0:11:59.520 Gramm and dot investors do. Like are you like an 0:11:59.559 --> 0:12:02.520 abstract like Metagram and Dada investors? I like the way 0:12:02.559 --> 0:12:03.520 to think of what you do. 0:12:04.320 --> 0:12:07.560 I think it's moved closer. There are still differences and 0:12:07.640 --> 0:12:12.120 maybe some of the like momentum if it's overreaction, if 0:12:12.120 --> 0:12:13.960 you're riding momentum. I don't think a Graham and DoD 0:12:14.000 --> 0:12:16.439 manager does that. So I don't want to push the analogy. 0:12:17.080 --> 0:12:19.760 But this came out very very early in my career. 0:12:19.800 --> 0:12:21.840 This is like thirty years ago. This is like my 0:12:21.880 --> 0:12:25.400 Goldman Sax days. I started hearing a lot of active 0:12:25.400 --> 0:12:28.000 stock pickers, some I'm still still friends with one guy 0:12:28.040 --> 0:12:31.000 in particular. I was laughing at them, and I was 0:12:31.000 --> 0:12:32.720 telling my friend, you all say the same thing. You 0:12:32.760 --> 0:12:35.960 all say you're looking for valuation plus a catalyst. It's 0:12:36.000 --> 0:12:37.840 like a I don't know if everyone says it, but 0:12:37.840 --> 0:12:41.360 I've heard it many times and I'm making fun of him, 0:12:41.480 --> 0:12:42.960 and at some point he looks at me and goes, 0:12:43.320 --> 0:12:47.319 you do value on momentum, and it was a gotcha. 0:12:47.440 --> 0:12:50.360 He won that round because I'm like, okay, I see 0:12:50.400 --> 0:12:53.400 your point. So even back then, you can think of 0:12:53.400 --> 0:12:55.600 those two together. We literally add them up. But if 0:12:55.600 --> 0:12:57.640 you think of him as this holistic system, we're looking 0:12:57.640 --> 0:12:59.719 for cheap things that are starting to get better in 0:12:59.760 --> 0:13:04.160 price or fundamentals over time. And now I'm I'm not 0:13:04.200 --> 0:13:07.840 talking the really modern stuff, alternative data, machine learning. I'm 0:13:07.840 --> 0:13:11.120 talking just classic quant stuff that's been academia and then 0:13:11.200 --> 0:13:15.880 imports over to applied. Profitability is a factor Robert nov 0:13:16.000 --> 0:13:18.880 Mark's fort a great paper that we all incorporated, where 0:13:19.600 --> 0:13:23.240 also equal give invaluation and momentum. If a company is 0:13:23.280 --> 0:13:28.119 more profitable on some famous scale's gross profitability ROA Roe, 0:13:28.480 --> 0:13:31.720 that's some degree a positive low beta investing. Two of 0:13:31.720 --> 0:13:35.839 my colleagues, Andrea Ferzini and los A. Peterson, resurrecting stuff 0:13:35.840 --> 0:13:38.640 Fisher Black did, and they're very good about saying Fisher 0:13:38.640 --> 0:13:42.960 did it first. That lower beta stocks. If you're famously 0:13:43.000 --> 0:13:46.360 a capital asset pricing model person, they're supposed to sell 0:13:46.360 --> 0:13:50.520 on average underperform higher beta stocks. That's the main output 0:13:50.559 --> 0:13:53.400 of that model. It doesn't work. I mean, it's one 0:13:53.400 --> 0:13:56.000 of the largest empirical failures ever. It doesn't work in 0:13:56.000 --> 0:13:59.480 any kind even outside of stocks people tested in other places. Therefore, 0:13:59.480 --> 0:14:02.559 it kind of makes low beta stocks a little bit 0:14:02.559 --> 0:14:05.079 of a free lunch because they are lower risk and 0:14:05.160 --> 0:14:08.800 they keep up. If you actually go read Graham and DoD, 0:14:09.000 --> 0:14:13.000 they're not just buying low multiples. They're much more holistic 0:14:13.040 --> 0:14:16.800 than that. You know, high quality companies that have a moat, 0:14:16.960 --> 0:14:19.240 that have some kind of margin of safety I think 0:14:19.360 --> 0:14:22.240 was the term they use. Margin of safety and looking 0:14:22.240 --> 0:14:26.400 for low risk doesn't sound so so different. So over 0:14:26.480 --> 0:14:29.640 time I've thought at least a core amount of what 0:14:29.880 --> 0:14:32.400 quants and academics, if you take them as a whole, 0:14:32.880 --> 0:14:36.200 are finding, is the full paneplyy of stuff that a 0:14:36.240 --> 0:14:39.080 Graham and DoD investor. We do it very differently. Again, 0:14:39.080 --> 0:14:41.600 we're betting on the concept working on average. They are 0:14:41.680 --> 0:14:44.400 using it in a soft or a hard sense as 0:14:44.440 --> 0:14:46.720 a screen to look for candidates, and then they're trying 0:14:46.760 --> 0:14:50.320 to learn a lot about that situation. They're upside as 0:14:50.360 --> 0:14:52.440 if they learn a lot about that situation, they could 0:14:52.440 --> 0:14:54.840 be more reliable than my You know, hey, we could 0:14:54.880 --> 0:14:57.960 be wrong about AMC, and their downside is they better 0:14:57.960 --> 0:15:00.560 be right. The concept can work, and they can still 0:15:00.560 --> 0:15:03.440 lose if they're wrong about the specific So over time 0:15:03.480 --> 0:15:06.000 I've gotten a little less hubrious about this. I think 0:15:06.120 --> 0:15:08.400 quants caused the problem, by the way. I think when 0:15:08.480 --> 0:15:10.880 Gene fam and Ken French started looking at like price 0:15:10.920 --> 0:15:14.200 to book in the late eighties and early nineties, and 0:15:14.240 --> 0:15:15.560 there were other people who did it too, I'm just 0:15:15.680 --> 0:15:17.800 Chicago guy, so I'm going to just go with Fama 0:15:17.840 --> 0:15:21.640 and French. They did it best. In my very biased opinion. 0:15:22.040 --> 0:15:23.720 I don't think i'd have to go back and check, 0:15:23.760 --> 0:15:25.800 but I don't think the first few papers use the 0:15:25.840 --> 0:15:30.600 word value investing. Over time, low multiple investing in the 0:15:30.680 --> 0:15:34.720 quant world came to be called value investing, and in 0:15:34.760 --> 0:15:36.480 the Gramm and Dodd world they get kind of mad 0:15:36.520 --> 0:15:39.080 at that, and they'd be like, it's not value investing. 0:15:39.120 --> 0:15:41.280 There are plenty of low multiple companies that deserve to 0:15:41.280 --> 0:15:43.880 be low multiple, and there are plenty of high multiple 0:15:43.880 --> 0:15:47.800 companies that deserve it, and I think over time, the 0:15:48.000 --> 0:15:51.920 quantitative process agreed with them more and more. So I 0:15:51.920 --> 0:15:54.440 still think it's a communication problem because they'll still talk 0:15:54.440 --> 0:15:57.320 about the value factor. In the quant world, that's just 0:15:57.400 --> 0:16:00.840 low multiples. And if a Gramm and Dodd gets mad 0:16:00.920 --> 0:16:03.760 or any old school active stock picker gets mad at that, 0:16:04.000 --> 0:16:06.960 I'll just say, you're right, because value implies a more 0:16:06.960 --> 0:16:08.600 holistic thing. 0:16:08.640 --> 0:16:10.640 But isn't like modern quantit and what you're doing now 0:16:10.760 --> 0:16:13.000 kind of just that more holistic thing, like you're just 0:16:13.160 --> 0:16:15.760 ingesting more data plants and you have a less linear model, 0:16:15.800 --> 0:16:17.760 and it's like moving towards that. 0:16:17.760 --> 0:16:19.960 Anyway, if you look at machine learning, where either to 0:16:20.000 --> 0:16:23.680 construct factors. One of the best uses of mL we 0:16:23.840 --> 0:16:26.920 found is a subset of mL called natural language processing, 0:16:27.320 --> 0:16:30.560 where you take textual data and you try to say, 0:16:30.600 --> 0:16:33.200 is this good news or bad news? Quants have kind 0:16:33.200 --> 0:16:36.440 of done this forever. You get transcript of an earnings call. 0:16:36.560 --> 0:16:38.040 And the old school way to do this for a 0:16:38.040 --> 0:16:41.200 long time was you count up good words and bad words, 0:16:41.640 --> 0:16:45.120 good phrases and bad phrases, so increasing plus one and 0:16:45.160 --> 0:16:47.600 you tech parse the whole thing, and sure, you guys 0:16:47.640 --> 0:16:50.400 immediately see the problem. If the actual sentence was massive 0:16:50.400 --> 0:16:53.960 embezzlement is increasing, then you were off on that plus one. 0:16:54.160 --> 0:16:58.000 Quantitative stuff can survive doing some horrifically stupid things in isolation, 0:16:58.520 --> 0:17:00.720 If fifty three percent of the time increasing is a 0:17:00.720 --> 0:17:04.440 good word, than forty percent of the time you were stupid, 0:17:04.840 --> 0:17:07.840 turns out natural language processing or NLP, if we want 0:17:07.880 --> 0:17:10.720 to sound like the cool kids. That is, taking that 0:17:10.800 --> 0:17:16.480 same data and training a mL model to say what 0:17:16.560 --> 0:17:18.840 predicts and what doesn't predict, and of course never gets 0:17:18.840 --> 0:17:21.320 near perfect at the end of the day, though we 0:17:21.440 --> 0:17:24.240 believe it does a lot better than the word count 0:17:24.280 --> 0:17:27.600 methods and is additive to a model. But importantly, and 0:17:27.640 --> 0:17:31.920 I think this was your point, Matt, it's not qualitatively different. 0:17:32.200 --> 0:17:36.760 We've looked at both price and fundamental momentum forever. Fundamental 0:17:36.760 --> 0:17:39.360 momentum the classic measures of things like our earnings being 0:17:39.440 --> 0:17:41.600 revised and you want the revision. You want the new 0:17:41.640 --> 0:17:46.440 news up faster or slower. Our earning surprise is coming 0:17:46.440 --> 0:17:48.639 in positive or negative. And this is this anchoring and 0:17:48.720 --> 0:17:52.320 adjustment idea that if that's good or bad news, you 0:17:52.400 --> 0:17:54.240 can make money trading on it. Because it's not fully 0:17:54.280 --> 0:17:58.440 incorporated in the stock price parsing and earnings call poorly 0:17:58.680 --> 0:18:01.080 as in the past or better now, we think of 0:18:01.119 --> 0:18:03.440 it as just another form of our good things happening 0:18:03.800 --> 0:18:05.560 in the new year term, and if so, they're probably 0:18:05.680 --> 0:18:10.119 under appreciated. So yes, I don't think it's changed dramatically. 0:18:10.119 --> 0:18:13.520 It's much more of an evolution rather than a spirit change. 0:18:13.760 --> 0:18:16.280 But we got bigger tools in the tool chest now. 0:18:16.840 --> 0:18:18.399 I mean, like, you know, you ask, like what a 0:18:18.560 --> 0:18:20.879 sort of like traditional old school fundamental manager would do. 0:18:20.880 --> 0:18:22.480 One thing they do is listen to their next call 0:18:22.720 --> 0:18:25.280 and like talk to management. So it's like your machines 0:18:25.280 --> 0:18:27.240 are moving in the direction of being an old school analyst. 0:18:27.440 --> 0:18:32.080 Yeah, as someone with four kids in or around college age, 0:18:32.600 --> 0:18:34.199 my wife and I and she's done a lot more 0:18:34.200 --> 0:18:35.680 of this, has spent a lot of time trying to 0:18:35.720 --> 0:18:41.600 figure out what careers will not be utterly destroyed by MLO, 0:18:41.800 --> 0:18:47.680 and it's not an easy question. A podcaster is probably 0:18:47.680 --> 0:18:49.240 a good one for a while. They have those fakes, 0:18:52.000 --> 0:18:55.399 so it will all be podcasters. But yeah, it's another example. 0:18:55.440 --> 0:18:58.800 I always say, these useless statements like them sufficiently long horizon, 0:18:58.840 --> 0:19:01.480 we're all replaced by machine learning. It's a question of 0:19:01.640 --> 0:19:05.960 what we're getting way too philosophical, but the transition to 0:19:06.040 --> 0:19:09.280 that can be very painful and weird. But a world 0:19:09.359 --> 0:19:13.359 of abundance and leisure. Maybe our horrible fate in the 0:19:13.359 --> 0:19:14.520 long term, I'll take it. 0:19:14.640 --> 0:19:27.640 Yeah. 0:19:30.119 --> 0:19:32.280 I want to talk about market timing because I was 0:19:32.320 --> 0:19:35.280 reading the Virtue of Complexity paper from Brian Kelly at All. 0:19:35.200 --> 0:19:36.800 Which is I will say he may be one of 0:19:36.840 --> 0:19:39.679 the only hosts of a podcast to read that paper. 0:19:39.720 --> 0:19:42.560 This is not a simple paper. It's like he writes, 0:19:42.640 --> 0:19:43.000 very clue. 0:19:43.040 --> 0:19:44.880 It's the sort of like it's the notion of taking 0:19:44.920 --> 0:19:46.760 like a sort of simple factor model and blowing it 0:19:46.840 --> 0:19:50.760