1 00:00:02,360 --> 00:00:06,720 Speaker 1: Bloomberg Audio Studios, Podcasts, radio News. 2 00:00:06,920 --> 00:00:09,600 Speaker 2: I'm really pleased to say that I'm here with AQR's 3 00:00:09,720 --> 00:00:10,440 Speaker 2: Cliff Asmas. 4 00:00:10,440 --> 00:00:12,280 Speaker 1: Cliff, thank you so much for joining this morning. Thanks 5 00:00:12,280 --> 00:00:12,760 Speaker 1: for having me. 6 00:00:12,840 --> 00:00:15,440 Speaker 2: We have to start with your paper that caused quite 7 00:00:15,440 --> 00:00:16,759 Speaker 2: a bit of a stir. Let me just read the 8 00:00:16,760 --> 00:00:20,520 Speaker 2: title for our audience, the less efficient market hypothesis. Quite 9 00:00:20,560 --> 00:00:24,280 Speaker 2: the claim for someone who studied under arguably the godfather 10 00:00:24,360 --> 00:00:27,280 Speaker 2: Eugene Fama of the EMH. So what did you see 11 00:00:27,280 --> 00:00:29,800 Speaker 2: that led you to believe this market is less efficient? 12 00:00:30,120 --> 00:00:34,239 Speaker 3: Sure, well, efficient markets in general, this idea that prices 13 00:00:34,560 --> 00:00:37,960 Speaker 3: in some rational sense reflect all information. 14 00:00:38,560 --> 00:00:40,400 Speaker 1: That's the extreme version of that toothesis. 15 00:00:41,280 --> 00:00:45,040 Speaker 3: Gene tells the class, like third week of class, markets 16 00:00:45,040 --> 00:00:48,440 Speaker 3: are assuredly not perfectly efficient, and you get a gasp. 17 00:00:48,520 --> 00:00:50,479 Speaker 3: Because it's the University of Chicago. Everywhere else in the 18 00:00:50,520 --> 00:00:52,600 Speaker 3: world they just go, yeah, we know that. But in 19 00:00:52,680 --> 00:00:56,240 Speaker 3: Gene's class, that's a gas. So it begs the question, 20 00:00:56,320 --> 00:00:59,440 Speaker 3: once you accept they're not perfect, how imperfect and does 21 00:00:59,440 --> 00:01:01,120 Speaker 3: that change through times? 22 00:01:01,480 --> 00:01:03,680 Speaker 1: And I'm very clear about this in the paper. A 23 00:01:03,680 --> 00:01:05,000 Speaker 1: lot of this is op ed. 24 00:01:05,360 --> 00:01:10,240 Speaker 3: It's my life experience in this business, but over my career. 25 00:01:10,280 --> 00:01:13,520 Speaker 3: At the very beginning of my career, when we started AQR, 26 00:01:14,160 --> 00:01:17,360 Speaker 3: eighteen of our first nineteen months were the blowoff top 27 00:01:17,440 --> 00:01:21,080 Speaker 3: of the dot com bubbles, and we did a lot 28 00:01:21,120 --> 00:01:21,720 Speaker 3: of things back then. 29 00:01:21,720 --> 00:01:23,160 Speaker 1: We were losing a lot of money. That was a 30 00:01:23,200 --> 00:01:23,720 Speaker 1: bad part. 31 00:01:23,880 --> 00:01:26,880 Speaker 3: It wasn't statistically like beyond shocking, but it's not pleasant 32 00:01:26,880 --> 00:01:30,000 Speaker 3: to happen in your first year and a half. And 33 00:01:30,080 --> 00:01:32,880 Speaker 3: it turned out that the prices between cheap and expensive 34 00:01:32,920 --> 00:01:36,720 Speaker 3: stocks in general, using a bunch of valuation measures, maybe 35 00:01:36,760 --> 00:01:39,319 Speaker 3: they vary between six times more expensive to three times 36 00:01:39,319 --> 00:01:39,960 Speaker 3: more expensive. 37 00:01:40,240 --> 00:01:41,679 Speaker 1: The expensive stocks. 38 00:01:41,840 --> 00:01:44,880 Speaker 3: They had shot up to like thirteen and they'd never 39 00:01:44,920 --> 00:01:47,120 Speaker 3: been closed. It was fifty years of this and then this. 40 00:01:47,960 --> 00:01:49,520 Speaker 3: So then you spend a lot of time saying, is 41 00:01:49,560 --> 00:01:50,760 Speaker 3: there some way we're wrong? 42 00:01:51,560 --> 00:01:53,640 Speaker 1: Is there reason this is rational? And if you include 43 00:01:53,680 --> 00:01:54,840 Speaker 1: know you stick with it. 44 00:01:55,760 --> 00:01:57,960 Speaker 3: Then if someone had said to me at that point, 45 00:01:58,120 --> 00:02:01,880 Speaker 3: is this going to happen again in your career, I hopefully. 46 00:02:01,560 --> 00:02:02,440 Speaker 1: Would never say never. 47 00:02:03,160 --> 00:02:05,160 Speaker 3: No one in our field, your field, my field should 48 00:02:05,160 --> 00:02:07,880 Speaker 3: ever say never about anything financial. But I think I 49 00:02:07,920 --> 00:02:10,960 Speaker 3: would have said, oh yeah, probably not, almost definitely not, 50 00:02:11,560 --> 00:02:15,160 Speaker 3: because it hadn't happened in fifty years my career is 51 00:02:15,200 --> 00:02:16,280 Speaker 3: not going to be another fifty years. 52 00:02:16,280 --> 00:02:17,840 Speaker 1: I wish it were, but it's not going to be. 53 00:02:18,320 --> 00:02:22,040 Speaker 3: And if the question in your career presupposes you're still. 54 00:02:21,800 --> 00:02:23,880 Speaker 1: Here and you're not the only one, and. 55 00:02:23,840 --> 00:02:26,200 Speaker 3: You guys are probably in charge, So how's this going 56 00:02:26,240 --> 00:02:29,040 Speaker 3: to happen again? And then even before COVID it was 57 00:02:29,080 --> 00:02:33,360 Speaker 3: getting there, but COVID took that same differential and blew 58 00:02:33,400 --> 00:02:34,440 Speaker 3: it out of the water the cliff. 59 00:02:34,440 --> 00:02:36,240 Speaker 2: I kind of argue, it'll play out the same as 60 00:02:36,320 --> 00:02:39,079 Speaker 2: last time, that spreads will come back down, value will 61 00:02:39,120 --> 00:02:41,040 Speaker 2: work again, and it's just we're in another one of 62 00:02:41,040 --> 00:02:41,560 Speaker 2: these period. 63 00:02:41,680 --> 00:02:44,840 Speaker 3: That's exactly what we do, argue, So you are totally 64 00:02:44,840 --> 00:02:50,040 Speaker 3: permitted to argue that. But the bigger question I'm trying 65 00:02:50,040 --> 00:02:52,160 Speaker 3: to answer is I see this as a sign of 66 00:02:52,200 --> 00:02:57,640 Speaker 3: the market getting less efficient about pricing. And in principle, 67 00:02:57,720 --> 00:03:00,000 Speaker 3: we all think things should get more efficient over time. 68 00:03:00,000 --> 00:03:04,280 Speaker 3: Technology marches on, data is more available, things that were 69 00:03:04,360 --> 00:03:07,560 Speaker 3: available at a leg or available instantly now, but I 70 00:03:07,560 --> 00:03:09,920 Speaker 3: think most of those things go to speed, not to 71 00:03:10,000 --> 00:03:13,680 Speaker 3: accurate price. It. Yeah, things get in prices faster, but 72 00:03:14,040 --> 00:03:15,520 Speaker 3: I think there's strong evidence so. 73 00:03:15,600 --> 00:03:17,639 Speaker 1: Not just these two things, but these are the two 74 00:03:17,680 --> 00:03:18,119 Speaker 1: major ones. 75 00:03:18,160 --> 00:03:21,320 Speaker 3: I experience that the crazy periods are getting crazier and 76 00:03:21,400 --> 00:03:24,920 Speaker 3: lasting longer. So the next question to ask in the paper, 77 00:03:25,320 --> 00:03:29,200 Speaker 3: and this is very speculative, I admit, is why why? 78 00:03:29,639 --> 00:03:32,360 Speaker 3: And there are probably other great ones, but I focus 79 00:03:32,440 --> 00:03:35,000 Speaker 3: on three, and I'll say I'm really fast sure. The 80 00:03:35,200 --> 00:03:38,680 Speaker 3: first is the rise of passive. That's everyone's favorite. Passive 81 00:03:38,760 --> 00:03:41,280 Speaker 3: is broken the market. I think that is probably part 82 00:03:41,280 --> 00:03:44,680 Speaker 3: of it. No one knows how many people can actually 83 00:03:44,680 --> 00:03:46,880 Speaker 3: be passive without breaking the market, but we know it's 84 00:03:46,880 --> 00:03:49,440 Speaker 3: not one hundred percent. That's a weird world where nobody 85 00:03:49,480 --> 00:03:51,440 Speaker 3: is thinking about whether Nvidio is worth more than a 86 00:03:51,440 --> 00:03:55,160 Speaker 3: candy story. So the idea that there would be less 87 00:03:55,320 --> 00:03:57,520 Speaker 3: could weaken the tether to reality. 88 00:03:57,480 --> 00:03:59,040 Speaker 2: And I'll let you get to the other two because 89 00:03:59,200 --> 00:04:01,480 Speaker 2: I find them really but justin in terms of that 90 00:04:01,600 --> 00:04:04,960 Speaker 2: Quesially we're talking about Nvidia and this being a crazy period. 91 00:04:04,960 --> 00:04:06,720 Speaker 1: No quants know nothing about Indorce, so. 92 00:04:06,760 --> 00:04:08,840 Speaker 2: You know nothing about a single stock. I'm very aware 93 00:04:08,880 --> 00:04:11,520 Speaker 2: of that. How crazy is this period? 94 00:04:12,920 --> 00:04:16,160 Speaker 3: We're still and I again I won't talk about Nvidia 95 00:04:16,160 --> 00:04:18,679 Speaker 3: in particular, but when it comes to the spread between 96 00:04:18,760 --> 00:04:23,600 Speaker 3: cheap and expensive stocks, we were at a new hundredth 97 00:04:23,640 --> 00:04:26,680 Speaker 3: percentile in COVID. I jokingly called it one hundred and 98 00:04:26,680 --> 00:04:29,760 Speaker 3: twentieth percentile, which from my math friends listening to you, 99 00:04:29,960 --> 00:04:31,880 Speaker 3: there is no such thing. It's just a new hundredth, 100 00:04:32,040 --> 00:04:34,160 Speaker 3: but it's about twenty percent above the prior hundredth. 101 00:04:34,640 --> 00:04:38,000 Speaker 1: We're back down to about the eightieth percentile spread versus history. 102 00:04:38,320 --> 00:04:39,880 Speaker 3: We still have a little bit of a tilt on, 103 00:04:40,000 --> 00:04:42,800 Speaker 3: but far smaller than when it was one hundred and twenty. 104 00:04:43,760 --> 00:04:45,920 Speaker 3: So I would say things are still out of whack, 105 00:04:47,040 --> 00:04:49,839 Speaker 3: but it's not an arbitragy you can drive a truck 106 00:04:49,880 --> 00:04:52,960 Speaker 3: through right now. But so I think passive as part 107 00:04:52,960 --> 00:04:54,920 Speaker 3: of it. But it's very hard to know how much. 108 00:04:55,320 --> 00:04:59,400 Speaker 3: The second one I'll do beyond quick post GFC ten 109 00:04:59,520 --> 00:05:03,120 Speaker 3: fifteen years of super low interest rates, I don't think 110 00:05:03,240 --> 00:05:05,520 Speaker 3: justifies these spreads, and we've written a lot about that. 111 00:05:06,040 --> 00:05:09,480 Speaker 3: But in a very non technical sense, may that drive 112 00:05:09,560 --> 00:05:11,880 Speaker 3: investors mad and to do some crazy things? 113 00:05:12,120 --> 00:05:14,080 Speaker 1: Yes, that's all I'll say about that. 114 00:05:14,640 --> 00:05:18,320 Speaker 3: My third and favorite hypopnsis because it's so counterintuitive. Is 115 00:05:19,040 --> 00:05:21,919 Speaker 3: I started out saying technology might make things faster, but 116 00:05:22,000 --> 00:05:22,760 Speaker 3: not more accurate. 117 00:05:23,560 --> 00:05:25,720 Speaker 1: It might. In fact, I would. 118 00:05:25,560 --> 00:05:29,279 Speaker 3: Argue probably does contribute to the less accurate pricing, and 119 00:05:29,279 --> 00:05:32,919 Speaker 3: in particular I'm talking about the melu of instant information 120 00:05:33,200 --> 00:05:33,919 Speaker 3: free trading. 121 00:05:34,400 --> 00:05:35,720 Speaker 1: Certainly social media. 122 00:05:36,279 --> 00:05:38,480 Speaker 3: The dot com bubble was the first go round of 123 00:05:38,600 --> 00:05:41,599 Speaker 3: kinna proto social media where you're. 124 00:05:41,480 --> 00:05:44,520 Speaker 1: A little too young for this one, but message boards 125 00:05:45,080 --> 00:05:46,520 Speaker 1: were very hot. 126 00:05:46,560 --> 00:05:51,119 Speaker 3: People are all discussing their favorite e toys and doctorcoop. 127 00:05:50,520 --> 00:05:51,359 Speaker 1: Dot coms and. 128 00:05:54,200 --> 00:05:58,080 Speaker 3: If markets are efficient to some degree not perfect. The 129 00:05:58,080 --> 00:06:00,880 Speaker 3: wisdom of crowds, the famous idea that a crowd can 130 00:06:00,920 --> 00:06:03,159 Speaker 3: be more wise than the average component of the crowd, 131 00:06:03,440 --> 00:06:04,160 Speaker 3: is a huge. 132 00:06:03,960 --> 00:06:04,440 Speaker 1: Part of it. 133 00:06:04,600 --> 00:06:07,520 Speaker 2: How much of your Twitter interactions color this because you're 134 00:06:07,720 --> 00:06:09,400 Speaker 2: fair to talk to? I just you know, when you're 135 00:06:09,440 --> 00:06:11,240 Speaker 2: talking to people who you don't agree with online, do 136 00:06:11,320 --> 00:06:13,440 Speaker 2: some of that color? It's saying maybe this crowd isn't 137 00:06:13,480 --> 00:06:13,919 Speaker 2: so wise. 138 00:06:14,040 --> 00:06:19,359 Speaker 1: I hope not. But but may it may it some 139 00:06:20,000 --> 00:06:20,840 Speaker 1: may it sneak in. 140 00:06:21,040 --> 00:06:23,320 Speaker 3: I would have to admit it. Might it might it 141 00:06:23,400 --> 00:06:25,960 Speaker 3: might sneak in. But if a wise crowd is the 142 00:06:26,000 --> 00:06:28,960 Speaker 3: goal in a market, have we ever had a vehicle 143 00:06:29,040 --> 00:06:32,040 Speaker 3: better than social media for turning a wise crowd into 144 00:06:32,040 --> 00:06:32,920 Speaker 3: a dangerous mob? 145 00:06:33,480 --> 00:06:37,360 Speaker 1: And I'm not going to get political except I'm not. 146 00:06:37,279 --> 00:06:41,520 Speaker 3: Going to choose a political side. But if our politics 147 00:06:41,560 --> 00:06:43,640 Speaker 3: are not a great example of this, I mean there 148 00:06:43,680 --> 00:06:45,920 Speaker 3: was a point twenty years ago where we had utopian 149 00:06:46,000 --> 00:06:48,039 Speaker 3: views with social media, that social media would make us 150 00:06:48,080 --> 00:06:50,280 Speaker 3: all like each other more. I don't think I need 151 00:06:50,320 --> 00:06:53,680 Speaker 3: to convince any of your viewers that that has not happened, 152 00:06:53,680 --> 00:06:55,840 Speaker 3: and I think a lot of the same has creeped 153 00:06:55,839 --> 00:06:57,360 Speaker 3: into markets. 154 00:06:57,560 --> 00:07:00,360 Speaker 1: It's unfair to choose the most crazy example. 155 00:07:01,560 --> 00:07:04,719 Speaker 3: But a poster child, not not not not a proof 156 00:07:04,920 --> 00:07:06,200 Speaker 3: are the US meme stocks. 157 00:07:06,480 --> 00:07:08,240 Speaker 1: They have the most extreme example of. 158 00:07:08,160 --> 00:07:12,360 Speaker 3: What I'm talking about, crazy evaluations driven by social media. 159 00:07:12,760 --> 00:07:15,520 Speaker 3: But life is not just them and everyone else. It's 160 00:07:15,560 --> 00:07:18,160 Speaker 3: a point on a spectrum. So I do think markets 161 00:07:18,160 --> 00:07:21,320 Speaker 3: have gotten someone less efficient. I think the final interesting 162 00:07:21,400 --> 00:07:24,040 Speaker 3: question is what does that mean for a rational style 163 00:07:24,080 --> 00:07:24,680 Speaker 3: of investing? 164 00:07:24,800 --> 00:07:27,120 Speaker 2: Can I ask you point out in your paper that 165 00:07:27,200 --> 00:07:29,640 Speaker 2: it's a problem for society? 166 00:07:29,800 --> 00:07:31,120 Speaker 1: Why is it a problem for society? 167 00:07:31,120 --> 00:07:32,760 Speaker 3: Well, if you're actually going to read the paper, it's 168 00:07:32,760 --> 00:07:34,800 Speaker 3: going to make this all take longer, you know that, right. 169 00:07:36,320 --> 00:07:38,640 Speaker 3: It's a it's a problem for society, and I think 170 00:07:38,960 --> 00:07:42,800 Speaker 3: I'll be a cheerleader for our industry. Quick seconds I 171 00:07:42,800 --> 00:07:45,520 Speaker 3: don't know of enough people who think about it this way, 172 00:07:45,560 --> 00:07:48,640 Speaker 3: but rational prices are very important to a free enterprise 173 00:07:48,720 --> 00:07:52,400 Speaker 3: society when price is a whilely irrational prices are how we. 174 00:07:52,360 --> 00:07:56,280 Speaker 1: Allocate resources, and we do it in a very unproductive way. 175 00:07:56,560 --> 00:07:59,080 Speaker 3: If companies that are not very productive and are speculative 176 00:07:59,120 --> 00:08:01,560 Speaker 3: and pie in the sky are are getting all the money, yeah, 177 00:08:01,800 --> 00:08:02,880 Speaker 3: so it matters. 178 00:08:03,120 --> 00:08:06,160 Speaker 1: And let me finish with one thought, what does it mean? 179 00:08:06,840 --> 00:08:10,000 Speaker 3: Well, if markets are somewhat I don't say they're terrible 180 00:08:10,080 --> 00:08:10,720 Speaker 3: less efficient. 181 00:08:11,400 --> 00:08:14,120 Speaker 1: Deviations from fair value We got ended there. 182 00:08:14,160 --> 00:08:15,000 Speaker 2: I'm so sorry. 183 00:08:15,240 --> 00:08:18,120 Speaker 3: Deviations from fair value will be bigger, last longer, but you'll. 184 00:08:18,000 --> 00:08:18,920 Speaker 1: Make more money from them. 185 00:08:20,200 --> 00:08:20,680 Speaker 2: You nailed it. 186 00:08:20,760 --> 00:08:21,400 Speaker 1: Thanks so much. 187 00:08:21,640 --> 00:08:25,200 Speaker 2: That was cliff astness of AQR over here at the 188 00:08:25,280 --> 00:08:26,080 Speaker 2: Capitlar conference.