1
00:00:02,720 --> 00:00:07,200
Speaker 1: Bloomberg Audio Studios, podcasts, radio news.

2
00:00:08,160 --> 00:00:09,680
Speaker 2: I was hoping to be like one of those like

3
00:00:09,680 --> 00:00:11,160
Speaker 2: clips on TikTok you see.

4
00:00:11,320 --> 00:00:13,280
Speaker 1: Fake that's the thing you look like you're on a

5
00:00:13,280 --> 00:00:14,760
Speaker 1: podcast pads on home.

6
00:00:14,880 --> 00:00:16,079
Speaker 2: Now it just looks like, you know.

7
00:00:16,040 --> 00:00:18,360
Speaker 3: Instead of we just randomly got together the.

8
00:00:18,440 --> 00:00:20,840
Speaker 2: Chat and microphonest.

9
00:00:20,960 --> 00:00:23,720
Speaker 3: It is like with Bloomberg podcasts behind us.

10
00:00:24,079 --> 00:00:26,960
Speaker 1: Yeah right, I guess we've conveyed podcasts efficiently with all

11
00:00:27,000 --> 00:00:28,520
Speaker 1: of the Bloomberg podcasts.

12
00:00:28,600 --> 00:00:28,800
Speaker 2: Joe.

13
00:00:29,040 --> 00:00:30,680
Speaker 4: Part of the reason this has to be on video

14
00:00:31,000 --> 00:00:33,839
Speaker 4: is because Matt shaved. Matt has had a beard for

15
00:00:33,960 --> 00:00:35,000
Speaker 4: the past I don't know that.

16
00:00:35,120 --> 00:00:39,960
Speaker 1: I've had a winter beard from like Christmas break through memorild.

17
00:00:39,600 --> 00:00:41,960
Speaker 5: That I shaved over COVID for the first time in

18
00:00:41,960 --> 00:00:45,320
Speaker 5: about thirty years. Okay, and my kids freaked out. Yeah,

19
00:00:45,400 --> 00:00:46,760
Speaker 5: they were like, you're an alien.

20
00:00:47,640 --> 00:00:49,720
Speaker 1: Yeah, my kids didn't care that much. But one of

21
00:00:49,720 --> 00:00:51,800
Speaker 1: my sons said, it's a different daddy.

22
00:00:52,120 --> 00:00:55,040
Speaker 5: Well, you also have hair on the top of your

23
00:00:55,080 --> 00:00:57,400
Speaker 5: head when you shave the beard and you don't have

24
00:00:57,440 --> 00:00:59,320
Speaker 5: any hair, Suddenly you're mister clean.

25
00:00:59,480 --> 00:01:01,000
Speaker 1: You just does mister clean.

26
00:01:02,000 --> 00:01:05,639
Speaker 2: No, I don't think so, mister clean. You know clean,

27
00:01:06,400 --> 00:01:08,680
Speaker 2: Come on, I think that's a fair point.

28
00:01:13,280 --> 00:01:16,520
Speaker 1: It's the Money Stuff podcast. We have a guest, Cliff

29
00:01:16,560 --> 00:01:20,880
Speaker 1: has runs a q R. Thanks for coming in, Thanks

30
00:01:20,920 --> 00:01:23,120
Speaker 1: for having me. I always like to ask how to

31
00:01:23,160 --> 00:01:26,240
Speaker 1: my managers, like what do you do for a living?

32
00:01:26,319 --> 00:01:29,160
Speaker 1: Like what is it like economic function of like the

33
00:01:29,200 --> 00:01:30,199
Speaker 1: business that you run.

34
00:01:30,600 --> 00:01:33,720
Speaker 3: Okay, those are to me slightly different questions.

35
00:01:33,920 --> 00:01:35,560
Speaker 1: Right. One sounds like it's about you. One sounds like

36
00:01:35,560 --> 00:01:36,240
Speaker 1: it's about AQR.

37
00:01:36,280 --> 00:01:39,520
Speaker 5: I'm more interested in even if it's about a q R.

38
00:01:40,160 --> 00:01:43,600
Speaker 5: What you do for a living is And people might

39
00:01:43,600 --> 00:01:47,200
Speaker 5: not like this phraseology, but you're trying to predict what

40
00:01:47,240 --> 00:01:50,000
Speaker 5: happens to securities. You're trying to buy ones that go

41
00:01:50,240 --> 00:01:52,840
Speaker 5: up either in the absolute or more than some benchmark,

42
00:01:52,880 --> 00:01:57,320
Speaker 5: and and sell ones that do the opposite. The broader question,

43
00:01:57,320 --> 00:02:00,800
Speaker 5: which I think is behind what you're saying, is what

44
00:02:00,840 --> 00:02:03,360
Speaker 5: do you do for the world by doing that? And

45
00:02:03,400 --> 00:02:06,520
Speaker 5: they overlap, but they're not exactly the same thing. Something

46
00:02:06,520 --> 00:02:08,640
Speaker 5: can have a net positive effect on the world even

47
00:02:08,639 --> 00:02:11,440
Speaker 5: if you're not waking up. And I'll admit this, I

48
00:02:11,480 --> 00:02:13,840
Speaker 5: don't think most people in their jobs waked up and

49
00:02:13,880 --> 00:02:17,360
Speaker 5: always think that way, and certainly not active managers just

50
00:02:17,440 --> 00:02:19,840
Speaker 5: go I am just making the world a better place.

51
00:02:19,840 --> 00:02:23,720
Speaker 3: They're thinking, is in Nvidia undervalued? Is it overvalued? The

52
00:02:23,800 --> 00:02:25,959
Speaker 3: things that I think you do for the world.

53
00:02:26,360 --> 00:02:31,840
Speaker 5: Is first, take the other side of positions other people

54
00:02:32,080 --> 00:02:35,160
Speaker 5: disagree with you on or don't want to bear. That

55
00:02:35,200 --> 00:02:38,800
Speaker 5: can take two forms. That can mean one of you

56
00:02:39,520 --> 00:02:41,000
Speaker 5: is biased and wrong.

57
00:02:41,400 --> 00:02:42,320
Speaker 3: You hope it's not you.

58
00:02:43,120 --> 00:02:45,240
Speaker 5: But in that case, what you do for the world

59
00:02:45,320 --> 00:02:47,840
Speaker 5: is you move prices back towards not necessarily all the

60
00:02:47,880 --> 00:02:51,440
Speaker 5: way too in the abstract, the correct price. That's hard

61
00:02:51,480 --> 00:02:54,400
Speaker 5: to define, actually, but something is mispriced and you take

62
00:02:54,440 --> 00:02:55,160
Speaker 5: the other side of that.

63
00:02:55,639 --> 00:02:57,520
Speaker 3: It could be a risk premium. Other people don't want

64
00:02:57,520 --> 00:03:00,919
Speaker 3: to bear in a lot of strategies necessarily that they're

65
00:03:00,919 --> 00:03:01,720
Speaker 3: making an error.

66
00:03:01,880 --> 00:03:05,120
Speaker 5: If a merger's announced and three quarters of the pop

67
00:03:05,160 --> 00:03:07,400
Speaker 5: you should get if it closes happens, a lot of

68
00:03:07,440 --> 00:03:09,600
Speaker 5: people might not want to stick around for that last quarter.

69
00:03:09,680 --> 00:03:11,160
Speaker 3: And if you're willing to.

70
00:03:11,080 --> 00:03:12,639
Speaker 5: Take the other side of that, maybe you get paid

71
00:03:12,639 --> 00:03:15,080
Speaker 5: a little for doing that, and that is a service

72
00:03:15,080 --> 00:03:17,600
Speaker 5: to the market. People want to get out and you're helping.

73
00:03:18,360 --> 00:03:23,000
Speaker 5: That's kind of the positive side, But again it's not

74
00:03:23,040 --> 00:03:24,000
Speaker 5: what you're thinking about.

75
00:03:23,800 --> 00:03:24,720
Speaker 3: When you do the trade.

76
00:03:25,040 --> 00:03:29,600
Speaker 5: There are positions active managers will take that are not

77
00:03:29,720 --> 00:03:30,200
Speaker 5: about that.

78
00:03:30,480 --> 00:03:31,440
Speaker 3: Let me put it this way.

79
00:03:31,639 --> 00:03:34,800
Speaker 5: If what you're trying to do is predict returns, you

80
00:03:34,840 --> 00:03:37,480
Speaker 5: can predict returns because the price.

81
00:03:37,280 --> 00:03:39,720
Speaker 3: Is moving towards truth.

82
00:03:40,120 --> 00:03:42,200
Speaker 5: But you can also make money if you predict the

83
00:03:42,240 --> 00:03:44,480
Speaker 5: price moves further away from truth.

84
00:03:44,640 --> 00:03:45,560
Speaker 3: You know, if you're a.

85
00:03:45,880 --> 00:03:49,440
Speaker 5: Momentum memestock investor. And doesn't mean you can't get that right,

86
00:03:49,520 --> 00:03:50,720
Speaker 5: you know. I think of that a little more as

87
00:03:50,760 --> 00:03:54,000
Speaker 5: trading than investing. But they all come together, and it

88
00:03:54,120 --> 00:03:58,160
Speaker 5: even gets complicated within some famous quantitative factors. One famous

89
00:03:58,200 --> 00:03:59,840
Speaker 5: quant factor is the momentum factor.

90
00:04:00,400 --> 00:04:02,360
Speaker 1: I asked the finance professor I should I have asked?

91
00:04:03,400 --> 00:04:06,840
Speaker 1: And he said, you should ask him if momentum trading

92
00:04:07,000 --> 00:04:08,560
Speaker 1: makes markets more or less efficient.

93
00:04:08,720 --> 00:04:11,000
Speaker 3: We don't fully know, but I can tell you the framework.

94
00:04:11,640 --> 00:04:12,920
Speaker 3: There are two.

95
00:04:12,800 --> 00:04:17,360
Speaker 5: Competing explanations in academia and a general world that cares

96
00:04:17,400 --> 00:04:20,440
Speaker 5: about these things. For why momentum on average.

97
00:04:20,120 --> 00:04:22,920
Speaker 3: Works, there's always a third that it will never.

98
00:04:22,800 --> 00:04:24,400
Speaker 5: Work again and it was just random and it was

99
00:04:24,480 --> 00:04:26,920
Speaker 5: just luck. But if it's true, why does it work?

100
00:04:27,240 --> 00:04:29,360
Speaker 5: I actually think these can both coexist, so it's not

101
00:04:29,400 --> 00:04:32,960
Speaker 5: truly embarrassing. But it sounds embarrassing that the two major

102
00:04:33,000 --> 00:04:37,960
Speaker 5: explanations one is under reaction and the other is over reaction.

103
00:04:38,360 --> 00:04:40,200
Speaker 5: When you've narrowed it down to two things that at.

104
00:04:40,200 --> 00:04:42,480
Speaker 3: Least feel like the opposites, you should feel a little shame.

105
00:04:42,520 --> 00:04:45,279
Speaker 1: For a second, there's the situation just like there's a

106
00:04:45,279 --> 00:04:48,640
Speaker 1: correct price and momentum is trading from below to above the.

107
00:04:48,560 --> 00:04:50,680
Speaker 3: Credit friends, Well, I'll give you two scenarios.

108
00:04:51,120 --> 00:04:55,000
Speaker 5: Information comes out and people have a behavioral bias that

109
00:04:55,080 --> 00:04:57,880
Speaker 5: behavioral psychologists would call anchoring an adjustment.

110
00:04:58,360 --> 00:05:00,920
Speaker 3: They move towards all the way to the.

111
00:05:00,880 --> 00:05:03,159
Speaker 5: New information, and I think that fits a lot of

112
00:05:03,160 --> 00:05:06,920
Speaker 5: intuition in the short run that can make momentum work.

113
00:05:07,200 --> 00:05:11,320
Speaker 5: If you're following fundamentals or prices, good news comes out

114
00:05:11,600 --> 00:05:14,719
Speaker 5: or the price moves. If good news comes out, on average,

115
00:05:14,760 --> 00:05:18,880
Speaker 5: the price goes up but not enough. If the price moves,

116
00:05:19,400 --> 00:05:21,680
Speaker 5: it may on average be responding to good news, and

117
00:05:21,720 --> 00:05:25,479
Speaker 5: simply by observing the price move, you can say, Okay,

118
00:05:25,560 --> 00:05:28,800
Speaker 5: sometimes it's wrong, but on average it doesn't quite move enough.

119
00:05:29,400 --> 00:05:31,520
Speaker 5: If that's the reason, And I think the weight of

120
00:05:31,560 --> 00:05:36,479
Speaker 5: the opinion in academia I believe is towards this underreaction explanation. Then,

121
00:05:36,600 --> 00:05:38,960
Speaker 5: even though you're trading on momentum, you're still moving the

122
00:05:39,000 --> 00:05:42,839
Speaker 5: price towards kind of truth or equilibrium or something. But

123
00:05:42,920 --> 00:05:45,760
Speaker 5: the flip side is overreaction. Do you think of that

124
00:05:45,920 --> 00:05:49,880
Speaker 5: more as just your classic positive feedback loop. Someone's buying

125
00:05:49,920 --> 00:05:53,039
Speaker 5: something just for you know, fomo. It's been going up

126
00:05:53,120 --> 00:05:55,080
Speaker 5: in there, and well that could be a negative reason,

127
00:05:55,160 --> 00:05:57,960
Speaker 5: or they're just predicting more people buy it because of Fomo.

128
00:05:58,480 --> 00:06:01,400
Speaker 5: In that sense, if you buy some weird meme coin,

129
00:06:01,920 --> 00:06:04,039
Speaker 5: you could do that for a rational reason, not that

130
00:06:04,120 --> 00:06:06,160
Speaker 5: you are a long term holder, but you just believe

131
00:06:06,600 --> 00:06:07,559
Speaker 5: it's going to keep going.

132
00:06:07,839 --> 00:06:09,440
Speaker 1: So did you make money on GameStop?

133
00:06:10,680 --> 00:06:11,680
Speaker 3: This the god's honest truth.

134
00:06:11,720 --> 00:06:13,520
Speaker 5: You won't believe me. I have no idea if we

135
00:06:13,520 --> 00:06:15,520
Speaker 5: were long or short game stop during the whole thing.

136
00:06:15,680 --> 00:06:16,920
Speaker 2: You never went back and check.

137
00:06:17,080 --> 00:06:19,520
Speaker 5: No, I never did. I did have an episode where

138
00:06:19,560 --> 00:06:22,520
Speaker 5: I mentioned on perhaps a different TV network that we

139
00:06:22,520 --> 00:06:23,479
Speaker 5: were short, AMC.

140
00:06:23,960 --> 00:06:25,520
Speaker 2: What does your Twitter look like after that?

141
00:06:25,600 --> 00:06:28,080
Speaker 5: It was very ugly. Yeah, I certainly knew of that world,

142
00:06:28,120 --> 00:06:30,880
Speaker 5: even though we're quants. I watch the markets all day,

143
00:06:30,920 --> 00:06:33,279
Speaker 5: even even if I don't do anything about it. It's

144
00:06:33,320 --> 00:06:35,120
Speaker 5: like the old joke about the weather. Everyone talks about it,

145
00:06:35,160 --> 00:06:36,280
Speaker 5: but nobody does anything about it.

146
00:06:36,360 --> 00:06:37,400
Speaker 2: Best thing to talk about.

147
00:06:37,720 --> 00:06:40,640
Speaker 5: So it's not like everything that went on with GameStop

148
00:06:40,720 --> 00:06:44,000
Speaker 5: Melvin Capitol. You know, I'm watching it every day, but

149
00:06:44,279 --> 00:06:47,839
Speaker 5: we take relevant tiny positions in every stock. There was

150
00:06:47,839 --> 00:06:50,160
Speaker 5: nothing weird in our P and L. And yes, I

151
00:06:50,200 --> 00:06:53,160
Speaker 5: was not even curious. It probably wasn't even in our

152
00:06:53,279 --> 00:06:54,560
Speaker 5: universe at that point of things.

153
00:06:54,560 --> 00:06:55,200
Speaker 3: We trade.

154
00:06:55,200 --> 00:06:59,600
Speaker 5: But then I'm going on this other network it's allowed

155
00:07:01,720 --> 00:07:04,240
Speaker 5: and in kind of a pre call of what are

156
00:07:04,240 --> 00:07:04,960
Speaker 5: we going to talk about?

157
00:07:05,000 --> 00:07:05,680
Speaker 3: You guys know, you.

158
00:07:05,600 --> 00:07:07,640
Speaker 5: Don't want to get on there and have absolutely nothing

159
00:07:07,680 --> 00:07:09,880
Speaker 5: to talk about. You want to have some not necessarily

160
00:07:09,880 --> 00:07:12,520
Speaker 5: the answers worked out, but agreed upon topics. They're like

161
00:07:12,560 --> 00:07:15,600
Speaker 5: everyone in this segment gives us some longs and shorts,

162
00:07:16,000 --> 00:07:18,200
Speaker 5: but I'm saying, as a quant that's kind of silly.

163
00:07:18,680 --> 00:07:21,000
Speaker 5: They're not indicative, and we kind of made a deal

164
00:07:21,680 --> 00:07:26,200
Speaker 5: where they'd let me briefly explain that doesn't make a

165
00:07:26,240 --> 00:07:28,160
Speaker 5: whole lot of sense for quantum, but it might be fun.

166
00:07:28,720 --> 00:07:30,600
Speaker 5: And my way of saying it was, if I give

167
00:07:30,600 --> 00:07:32,200
Speaker 5: you a few names long and a few names short,

168
00:07:33,040 --> 00:07:34,840
Speaker 5: you could look in six months later and think we

169
00:07:34,880 --> 00:07:37,240
Speaker 5: had a fantastic year or a terrible year and be

170
00:07:37,480 --> 00:07:40,920
Speaker 5: terribly wrong in either direction because they're tiny. So I

171
00:07:40,960 --> 00:07:44,200
Speaker 5: went through an AMC was I think I'm accidentally doing

172
00:07:44,240 --> 00:07:44,640
Speaker 5: it again.

173
00:07:45,520 --> 00:07:46,960
Speaker 2: We'll keep going hopefully.

174
00:07:46,600 --> 00:07:47,680
Speaker 3: They don't listen to you, guys.

175
00:07:49,600 --> 00:07:53,600
Speaker 5: But it was bad on every single thing in our

176
00:07:53,680 --> 00:07:57,440
Speaker 5: model practically, which is hard to do. It was expensive, unprofitable,

177
00:07:57,520 --> 00:08:00,800
Speaker 5: high beta. They were issuing shares, not buying backshit. There

178
00:08:00,800 --> 00:08:03,080
Speaker 5: are more examples, and so I said that, but then

179
00:08:03,120 --> 00:08:07,880
Speaker 5: I added that were only short twelve basis points, so

180
00:08:08,160 --> 00:08:12,000
Speaker 5: the crazy people could be right and it doesn't really matter.

181
00:08:12,280 --> 00:08:14,440
Speaker 5: I discovered two things. They're not going to like a

182
00:08:14,440 --> 00:08:19,000
Speaker 5: short period, and crazy people don't always like being called crazy. Yeah,

183
00:08:19,040 --> 00:08:21,239
Speaker 5: I had to discover that from myself. So my Twitter

184
00:08:21,280 --> 00:08:23,600
Speaker 5: got ugly for a while. You may have noticed. I've

185
00:08:23,600 --> 00:08:26,120
Speaker 5: gotten I think a fair amount better at this, But

186
00:08:26,240 --> 00:08:28,760
Speaker 5: I used to be pretty bad about responding to ugly,

187
00:08:29,400 --> 00:08:30,920
Speaker 5: which you learn your lesson on that.

188
00:08:30,920 --> 00:08:32,720
Speaker 2: You always feed the trolls.

189
00:08:32,640 --> 00:08:34,840
Speaker 3: Less so than I used to. At least I think

190
00:08:34,920 --> 00:08:35,360
Speaker 3: I'm better.

191
00:08:35,440 --> 00:08:39,000
Speaker 5: Maybe I'm wrong, but I became public enemy number three

192
00:08:39,080 --> 00:08:39,760
Speaker 5: to the meme.

193
00:08:39,600 --> 00:08:41,560
Speaker 3: Stock crowd for a while, I did.

194
00:08:41,520 --> 00:08:45,800
Speaker 5: Not really Yes, Kenny Ken Griffin was number one. I

195
00:08:45,840 --> 00:08:48,080
Speaker 5: don't believe Ken did this, but it's the whole poll,

196
00:08:48,240 --> 00:08:53,320
Speaker 5: the Bible. And oddly enough, Gary Gensler was clearly number two. Yeah,

197
00:08:53,360 --> 00:08:57,280
Speaker 5: because they thought he was covering for the manipulators and

198
00:08:57,320 --> 00:08:59,840
Speaker 5: the naked shorts and whatnot. I've met both, I know,

199
00:09:00,040 --> 00:09:01,480
Speaker 5: and a little better than Gary. I don't think there

200
00:09:01,520 --> 00:09:04,160
Speaker 5: are any two people on Earth less likely to be

201
00:09:04,240 --> 00:09:07,079
Speaker 5: in cahoots than those two. I think they're on the

202
00:09:07,120 --> 00:09:10,240
Speaker 5: opposite side of most issues. But that was the theory.

203
00:09:10,280 --> 00:09:12,520
Speaker 5: But both Ken and Gary are too smart to respond

204
00:09:12,520 --> 00:09:15,400
Speaker 5: to them on Twitter, so I certainly became the most

205
00:09:15,440 --> 00:09:19,280
Speaker 5: actively engaged, and then I never did that again before today,

206
00:09:19,640 --> 00:09:21,000
Speaker 5: when I've accidentally done it.

207
00:09:21,120 --> 00:09:22,959
Speaker 4: Before we get back to stuff that matters, can I

208
00:09:23,000 --> 00:09:26,960
Speaker 4: just on AMC, I tweeted, when did June two come out?

209
00:09:27,080 --> 00:09:27,920
Speaker 2: It was like last summer.

210
00:09:28,120 --> 00:09:28,720
Speaker 3: Yeah, year ago.

211
00:09:28,800 --> 00:09:29,520
Speaker 2: I tweeted that I.

212
00:09:29,440 --> 00:09:33,200
Speaker 4: Fell asleep during Dune two, and it reawakened that crowd,

213
00:09:33,200 --> 00:09:35,839
Speaker 4: at least on my Twitter, because I've the same.

214
00:09:35,640 --> 00:09:39,400
Speaker 3: Crowds crowd because he's dissing a movie.

215
00:09:40,440 --> 00:09:43,360
Speaker 5: Don't understand you have to like every movie or else

216
00:09:43,400 --> 00:09:46,200
Speaker 5: here anti America, And then there were.

217
00:09:46,160 --> 00:09:49,840
Speaker 4: A lot of conspiracy theories about Bloomberg reporter lies about

218
00:09:50,160 --> 00:09:51,040
Speaker 4: falling asleep in.

219
00:09:51,080 --> 00:09:54,000
Speaker 2: Doom too because she hates AMC. It would be, but

220
00:09:54,040 --> 00:09:55,480
Speaker 2: it wasn't to.

221
00:09:55,760 --> 00:09:59,520
Speaker 5: Oh, yeah, you do innocuous that I was obnoxious, so

222
00:09:59,559 --> 00:10:00,560
Speaker 5: I'd kind of deserved it.

223
00:10:00,640 --> 00:10:03,120
Speaker 3: You didn't deserve Thank you for saying that you didn't

224
00:10:03,120 --> 00:10:03,600
Speaker 3: deserve it.

225
00:10:03,679 --> 00:10:05,520
Speaker 1: I haven't sent. That's not doing what I was good.

226
00:10:05,880 --> 00:10:07,520
Speaker 2: That's all too. It's pretty good.

227
00:10:07,760 --> 00:10:09,600
Speaker 3: No you didn't, actually I didn't.

228
00:10:09,760 --> 00:10:12,360
Speaker 5: You were asleep, well no, not the whole time, not

229
00:10:12,400 --> 00:10:13,000
Speaker 5: the whole.

230
00:10:12,760 --> 00:10:15,679
Speaker 1: Time, but also like you and and they were like,

231
00:10:15,800 --> 00:10:19,040
Speaker 1: give me some shorts, and you gave them some lungs

232
00:10:19,040 --> 00:10:21,720
Speaker 1: and shorts. Would you have known that or did you

233
00:10:21,760 --> 00:10:22,880
Speaker 1: have to be like I got it.

234
00:10:22,960 --> 00:10:25,079
Speaker 3: I would not have known that, right, So you don't.

235
00:10:24,880 --> 00:10:26,439
Speaker 5: Know, you're ar It would have been better for me

236
00:10:26,480 --> 00:10:28,520
Speaker 5: if we didn't have the call, because then I could

237
00:10:28,520 --> 00:10:32,720
Speaker 5: have just said I don't know. I rarely know individual stocks,

238
00:10:33,120 --> 00:10:35,400
Speaker 5: and if I do know, I'm probably not happy I know.

239
00:10:35,960 --> 00:10:38,120
Speaker 5: And even then it's like we lost twenty BIPs on

240
00:10:38,160 --> 00:10:40,000
Speaker 5: that today, which is a giant number for us to

241
00:10:40,040 --> 00:10:42,680
Speaker 5: lose on one stock, and even then I probably don't

242
00:10:42,720 --> 00:10:44,040
Speaker 5: notice twenty bit.

243
00:10:44,040 --> 00:10:46,200
Speaker 1: If someone comes to you and says that's not.

244
00:10:46,320 --> 00:10:48,959
Speaker 3: One, I might be told about it.

245
00:10:49,000 --> 00:10:51,480
Speaker 5: For us, it's whether these seven hundred and fifty stocks be

246
00:10:51,600 --> 00:10:54,840
Speaker 5: these seven hundred and fifty stocks. Yeah, I don't memorize

247
00:10:54,840 --> 00:10:58,120
Speaker 5: fifteen hundred stocks. Now, we take a fair amount of risks,

248
00:10:58,160 --> 00:11:01,040
Speaker 5: some funds more, some funds designed to be less. And

249
00:11:01,080 --> 00:11:04,840
Speaker 5: that's about the size of these two positions. So when

250
00:11:04,840 --> 00:11:07,320
Speaker 5: I say small, I'm not saying we're not both taking

251
00:11:07,400 --> 00:11:10,520
Speaker 5: risk and trying to generate pretty decent returns. But if

252
00:11:10,520 --> 00:11:12,640
Speaker 5: you just think about it, a quant is playing the odds.

253
00:11:12,920 --> 00:11:17,319
Speaker 5: They're saying, affirm a company with these characteristics. And this

254
00:11:17,360 --> 00:11:20,520
Speaker 5: can be old school factor quants from the nineteen nineties,

255
00:11:20,559 --> 00:11:23,760
Speaker 5: these can be modern machine learning. But with these characteristics

256
00:11:23,760 --> 00:11:26,880
Speaker 5: tend to beat these characteristics. If that's all you know,

257
00:11:26,960 --> 00:11:29,680
Speaker 5: and it is all we know, why on earth would

258
00:11:29,679 --> 00:11:31,320
Speaker 5: you take a lot of risk in any one company.

259
00:11:31,520 --> 00:11:33,080
Speaker 3: AMC really could have done well.

260
00:11:33,120 --> 00:11:35,600
Speaker 5: It could have been bad on every single thing that

261
00:11:35,720 --> 00:11:38,120
Speaker 5: on average doesn't work, and it could be a special

262
00:11:38,600 --> 00:11:41,240
Speaker 5: situation that we don't understand. Something could be good on

263
00:11:41,320 --> 00:11:44,920
Speaker 5: everything and the CEO can embezzle all the money. We

264
00:11:45,040 --> 00:11:46,920
Speaker 5: don't want to take a lot of risk on any

265
00:11:46,920 --> 00:11:48,880
Speaker 5: one thing because we have no insight in that it's

266
00:11:48,960 --> 00:11:49,920
Speaker 5: risk for no return.

267
00:11:50,320 --> 00:11:53,000
Speaker 1: One thing you've written is that like over time, the

268
00:11:53,400 --> 00:11:56,439
Speaker 1: quant like factor model has moved closer to being what

269
00:11:56,559 --> 00:11:59,520
Speaker 1: Gramm and dot investors do. Like are you like an

270
00:11:59,559 --> 00:12:02,520
Speaker 1: abstract like Metagram and Dada investors? I like the way

271
00:12:02,559 --> 00:12:03,520
Speaker 1: to think of what you do.

272
00:12:04,320 --> 00:12:05,760
Speaker 3: I think it's moved closer.

273
00:12:05,800 --> 00:12:09,200
Speaker 5: There are still differences and maybe some of the like

274
00:12:09,480 --> 00:12:13,199
Speaker 5: momentum if it's overreaction, if you're riding momentum. I don't

275
00:12:13,200 --> 00:12:14,959
Speaker 5: think a Graham and DoD manager does that. So I

276
00:12:14,960 --> 00:12:18,360
Speaker 5: don't want to push the analogy. But this came out

277
00:12:18,480 --> 00:12:20,720
Speaker 5: very very early in my career. This is like thirty

278
00:12:20,920 --> 00:12:23,480
Speaker 5: years ago. This is like my Goldman Sax days. I

279
00:12:23,559 --> 00:12:26,600
Speaker 5: started hearing a lot of active stock pickers, some I'm

280
00:12:26,640 --> 00:12:29,320
Speaker 5: still still friends with one guy in particular. I was

281
00:12:29,600 --> 00:12:31,679
Speaker 5: laughing at them, and I was telling my friend, you

282
00:12:31,720 --> 00:12:33,559
Speaker 5: all say the same thing. You all say you're looking

283
00:12:33,600 --> 00:12:36,800
Speaker 5: for valuation plus a catalyst. It's like a I don't

284
00:12:36,800 --> 00:12:38,959
Speaker 5: know if everyone says it, but I've heard it many

285
00:12:39,000 --> 00:12:41,880
Speaker 5: times and I'm making fun of him, and at some

286
00:12:41,880 --> 00:12:43,959
Speaker 5: point he looks at me and goes, you do value

287
00:12:43,960 --> 00:12:47,960
Speaker 5: on momentum, and it was a gotcha. He won that

288
00:12:48,040 --> 00:12:51,480
Speaker 5: round because I'm like, okay, I see your point. So

289
00:12:51,640 --> 00:12:54,160
Speaker 5: even back then, you can think of those two together.

290
00:12:54,240 --> 00:12:55,280
Speaker 3: We literally add them up.

291
00:12:55,320 --> 00:12:57,240
Speaker 5: But if you think of him as this holistic system,

292
00:12:57,240 --> 00:12:58,960
Speaker 5: we're looking for cheap things that are starting to get

293
00:12:59,360 --> 00:13:03,600
Speaker 5: better in price or fundamentals over time. And now I'm

294
00:13:03,920 --> 00:13:07,360
Speaker 5: I'm not talking the really modern stuff, alternative data, machine learning.

295
00:13:07,640 --> 00:13:11,000
Speaker 5: I'm talking just classic quant stuff that's been academia and

296
00:13:11,000 --> 00:13:13,199
Speaker 5: then imports over to applied.

297
00:13:13,800 --> 00:13:15,200
Speaker 3: Profitability is a factor.

298
00:13:15,360 --> 00:13:17,440
Speaker 5: Robert nov Mark's wrote a great paper that we all

299
00:13:17,880 --> 00:13:22,160
Speaker 5: incorporated where also equal give evaluation and momentum.

300
00:13:22,400 --> 00:13:25,000
Speaker 3: If a company is more profitable on some.

301
00:13:25,080 --> 00:13:29,319
Speaker 5: Famous scale's gross profitability ROA Roe, that's some degree a

302
00:13:29,400 --> 00:13:33,240
Speaker 5: positive low beta investing. Two of my colleagues, Andrea Frazini

303
00:13:33,440 --> 00:13:36,760
Speaker 5: and los A. Peterson, resurrecting stuff Fisher Black did, and

304
00:13:36,800 --> 00:13:40,040
Speaker 5: they're very good about saying Fisher did it first. That

305
00:13:40,320 --> 00:13:44,360
Speaker 5: lower beta stocks. If you're famously a capital asset pricing

306
00:13:44,559 --> 00:13:48,160
Speaker 5: model person, they're supposed to sell on average underperform higher

307
00:13:48,200 --> 00:13:51,960
Speaker 5: beta stocks. That's the main output of that model. It

308
00:13:52,040 --> 00:13:54,480
Speaker 5: doesn't work. I mean, it's one of the largest empirical

309
00:13:54,480 --> 00:13:56,920
Speaker 5: failures ever. It doesn't work in any kind even outside

310
00:13:56,920 --> 00:13:59,880
Speaker 5: of stocks people tested in other places. Therefore, it kind

311
00:13:59,880 --> 00:14:02,720
Speaker 5: of makes low beta stocks a little bit of a

312
00:14:02,760 --> 00:14:05,920
Speaker 5: free lunch because they are lower risk and they keep up.

313
00:14:06,800 --> 00:14:09,360
Speaker 5: If you actually go read Graham and DoD, they're not

314
00:14:09,480 --> 00:14:13,400
Speaker 5: just buying low multiples. They're much more holistic than that.

315
00:14:13,640 --> 00:14:17,080
Speaker 5: You know, high quality companies that have a moat, that

316
00:14:17,120 --> 00:14:19,400
Speaker 5: have some kind of margin of safety I think was

317
00:14:19,440 --> 00:14:22,320
Speaker 5: the term they use. Margin of safety and looking for

318
00:14:22,400 --> 00:14:26,720
Speaker 5: low risk doesn't sound so so different. So over time

319
00:14:26,760 --> 00:14:30,320
Speaker 5: I've thought at least a core amount of what quants

320
00:14:30,440 --> 00:14:33,560
Speaker 5: and academics, if you take them as a whole, are finding,

321
00:14:34,240 --> 00:14:36,560
Speaker 5: is the full paneplyy of stuff that a Graham and

322
00:14:36,600 --> 00:14:39,560
Speaker 5: DoD investor. We do it very differently. Again, we're betting

323
00:14:39,560 --> 00:14:41,320
Speaker 5: on the concept working on average.

324
00:14:41,320 --> 00:14:42,280
Speaker 3: They are using it.

325
00:14:42,840 --> 00:14:44,880
Speaker 5: In a soft or a hard sense as a screen

326
00:14:45,000 --> 00:14:47,480
Speaker 5: to look for candidates, and then they're trying to learn

327
00:14:47,520 --> 00:14:50,680
Speaker 5: a lot about that situation. They're upside as if they

328
00:14:50,800 --> 00:14:52,680
Speaker 5: learn a lot about that situation, they could be more

329
00:14:52,720 --> 00:14:55,160
Speaker 5: reliable than my You know, hey, we could be wrong

330
00:14:55,200 --> 00:14:58,360
Speaker 5: about AMC, and their downside is they better be right.

331
00:14:58,840 --> 00:15:00,880
Speaker 5: The concept can work, and they can still lose if

332
00:15:00,920 --> 00:15:04,080
Speaker 5: they're wrong about the specific So over time I've gotten

333
00:15:04,120 --> 00:15:06,760
Speaker 5: a little less hubrious about this. I think quants caused

334
00:15:06,800 --> 00:15:08,920
Speaker 5: the problem, by the way. I think when Gene fam

335
00:15:09,040 --> 00:15:11,320
Speaker 5: and Ken French started looking at like price to book

336
00:15:11,400 --> 00:15:14,480
Speaker 5: in the late eighties and early nineties, and there were

337
00:15:14,480 --> 00:15:16,360
Speaker 5: other people who did it too, I'm just Chicago guy,

338
00:15:16,400 --> 00:15:18,200
Speaker 5: so I'm going to just go with Fama and French.

339
00:15:18,560 --> 00:15:22,160
Speaker 5: They did it best. In my very biased opinion. I

340
00:15:22,200 --> 00:15:23,840
Speaker 5: don't think i'd have to go back and check, but

341
00:15:23,880 --> 00:15:26,080
Speaker 5: I don't think the first few papers use the word

342
00:15:26,200 --> 00:15:31,000
Speaker 5: value investing. Over time, low multiple investing in the quant

343
00:15:31,040 --> 00:15:34,840
Speaker 5: world came to be called value investing, and in the

344
00:15:34,880 --> 00:15:36,760
Speaker 5: Gramm and Dodd world they get kind of mad at that,

345
00:15:37,560 --> 00:15:39,320
Speaker 5: and they'd be like, it's not value investing. There are

346
00:15:39,320 --> 00:15:42,600
Speaker 5: plenty of low multiple companies that deserve to be low multiple,

347
00:15:42,640 --> 00:15:45,200
Speaker 5: and there are plenty of high multiple companies that deserve it,

348
00:15:45,800 --> 00:15:51,000
Speaker 5: and I think over time, the quantitative process agreed with them.

349
00:15:51,000 --> 00:15:51,600
Speaker 3: More and more.

350
00:15:51,720 --> 00:15:54,040
Speaker 5: So I still think it's a communication problem because they'll

351
00:15:54,080 --> 00:15:56,960
Speaker 5: still talk about the value factor. In the quant world,

352
00:15:56,960 --> 00:15:59,680
Speaker 5: that's just low multiples. And if a Gramm and Dodd

353
00:16:00,360 --> 00:16:03,280
Speaker 5: gets mad or any old school active stock picker gets

354
00:16:03,320 --> 00:16:06,080
Speaker 5: mad at that, I'll just say you're right, because value

355
00:16:06,160 --> 00:16:08,600
Speaker 5: implies a more holistic thing.

356
00:16:08,640 --> 00:16:10,640
Speaker 1: But isn't like modern quantit and what you're doing now

357
00:16:10,760 --> 00:16:13,000
Speaker 1: kind of just that more holistic thing, like you're just

358
00:16:13,160 --> 00:16:15,760
Speaker 1: ingesting more data plans and you have a less linear model,

359
00:16:15,800 --> 00:16:17,760
Speaker 1: and it's like moving towards that.

360
00:16:17,760 --> 00:16:19,960
Speaker 5: Anyway, if you look at machine learning, where either to

361
00:16:20,000 --> 00:16:23,680
Speaker 5: construct factors. One of the best uses of mL we

362
00:16:23,840 --> 00:16:26,920
Speaker 5: found is a subset of mL called natural language processing,

363
00:16:27,320 --> 00:16:30,560
Speaker 5: where you take textual data and you try to say

364
00:16:30,600 --> 00:16:33,200
Speaker 5: is this good news or bad news? Quants have kind

365
00:16:33,200 --> 00:16:36,480
Speaker 5: of done this forever. You get transcript of an earnings call.

366
00:16:36,560 --> 00:16:38,040
Speaker 5: And the old school way to do this for a

367
00:16:38,040 --> 00:16:41,200
Speaker 5: long time was you count up good words and bad words,

368
00:16:41,640 --> 00:16:45,120
Speaker 5: good phrases and bad phrases, so increasing plus one and

369
00:16:45,160 --> 00:16:47,600
Speaker 5: you tech parse the whole thing, and sure, you guys

370
00:16:47,640 --> 00:16:50,400
Speaker 5: immediately see the problem. If the actual sentence was massive

371
00:16:50,400 --> 00:16:53,960
Speaker 5: embezzlement is increasing, then you were off on that plus one.

372
00:16:54,160 --> 00:16:58,000
Speaker 5: Quantitative stuff can survive doing some horrifically stupid things in isolation,

373
00:16:58,520 --> 00:17:00,720
Speaker 5: If fifty three percent of the time increasing is a

374
00:17:00,720 --> 00:17:04,440
Speaker 5: good word, than forty percent of the time you were stupid,

375
00:17:04,840 --> 00:17:07,840
Speaker 5: turns out natural language processing or NLP, if we want

376
00:17:07,880 --> 00:17:10,720
Speaker 5: to sound like the cool kids. That is, taking that

377
00:17:10,800 --> 00:17:16,480
Speaker 5: same data and training a mL model to say what

378
00:17:16,560 --> 00:17:18,840
Speaker 5: predicts and what doesn't predict, and of course never gets

379
00:17:18,840 --> 00:17:21,320
Speaker 5: near perfect at the end of the day, though we

380
00:17:21,440 --> 00:17:24,240
Speaker 5: believe it does a lot better than the word count

381
00:17:24,280 --> 00:17:27,600
Speaker 5: methods and is additive to a model. But importantly, and

382
00:17:27,640 --> 00:17:31,920
Speaker 5: I think this was your point, Matt, it's not qualitatively different.

383
00:17:32,200 --> 00:17:36,760
Speaker 5: We've looked at both price and fundamental momentum forever. Fundamental

384
00:17:36,760 --> 00:17:39,360
Speaker 5: momentum the classic measures of things like our earnings being

385
00:17:39,440 --> 00:17:41,040
Speaker 5: revised and you want the revision.

386
00:17:41,080 --> 00:17:42,280
Speaker 3: You want the new news.

387
00:17:42,720 --> 00:17:46,520
Speaker 5: Up faster or slower. Our earning surprise is coming in

388
00:17:46,600 --> 00:17:49,200
Speaker 5: positive or negative. And this is this anchoring and adjustment

389
00:17:49,240 --> 00:17:52,560
Speaker 5: idea that if that's good or bad news, you can

390
00:17:52,560 --> 00:17:54,800
Speaker 5: make money trading on it. Because it's not fully incorporated

391
00:17:54,800 --> 00:17:58,880
Speaker 5: in the stock price parsing and earnings call poorly as

392
00:17:58,920 --> 00:18:01,159
Speaker 5: in the past or better now, we think of it

393
00:18:01,200 --> 00:18:03,879
Speaker 5: as just another form of our good things happening in

394
00:18:03,920 --> 00:18:07,200
Speaker 5: the new year term, and if so, they're probably under appreciated.

395
00:18:07,280 --> 00:18:10,560
Speaker 5: So yes, I don't think it's changed dramatically. It's much

396
00:18:10,560 --> 00:18:13,520
Speaker 5: more of an evolution rather than a spirit change.

397
00:18:13,760 --> 00:18:16,280
Speaker 3: But we got bigger tools in the tool chest now.

398
00:18:16,840 --> 00:18:18,399
Speaker 1: I mean, like, you know, you ask, like what a

399
00:18:18,560 --> 00:18:20,879
Speaker 1: sort of like traditional old school fundamental manager would do.

400
00:18:20,880 --> 00:18:22,480
Speaker 1: One thing they do is listen to their next call

401
00:18:22,720 --> 00:18:25,280
Speaker 1: and like talk to management. So it's like your machines

402
00:18:25,280 --> 00:18:27,240
Speaker 1: are moving in the direction of being an old school analyst.

403
00:18:27,440 --> 00:18:32,080
Speaker 5: Yeah, as someone with four kids in or around college age,

404
00:18:32,600 --> 00:18:34,199
Speaker 5: my wife and I and she's done a lot more

405
00:18:34,200 --> 00:18:35,680
Speaker 5: of this, has spent a lot of time trying to

406
00:18:35,720 --> 00:18:41,520
Speaker 5: figure out what careers will not be utterly destroyed by MLO,

407
00:18:41,800 --> 00:18:44,359
Speaker 5: and it's not an easy question.

408
00:18:46,560 --> 00:18:48,240
Speaker 3: A podcaster is probably a good one.

409
00:18:48,080 --> 00:18:48,400
Speaker 5: For a while.

410
00:18:48,480 --> 00:18:53,520
Speaker 3: They have those fakes, so it will all be podcasters.

411
00:18:53,640 --> 00:18:56,879
Speaker 5: But yeah, it's another example. I always say, these useless

412
00:18:56,880 --> 00:18:59,639
Speaker 5: statements like them sufficiently long horizon, we're all replaced by

413
00:18:59,680 --> 00:19:00,879
Speaker 5: machine learning.

414
00:19:00,880 --> 00:19:01,679
Speaker 3: It's a question of what.

415
00:19:03,240 --> 00:19:06,200
Speaker 5: We're getting way too philosophical, but the transition to that

416
00:19:06,280 --> 00:19:09,600
Speaker 5: can be very painful and weird. But a world of

417
00:19:09,680 --> 00:19:13,880
Speaker 5: abundance and leisure. Maybe our horrible fate in the long term,

418
00:19:13,920 --> 00:19:14,520
Speaker 5: I'll take it.

419
00:19:14,640 --> 00:19:28,520
Speaker 6: Yeah.

420
00:19:30,119 --> 00:19:32,280
Speaker 1: I want to talk about market timing because I was

421
00:19:32,320 --> 00:19:35,280
Speaker 1: reading the Virtue of Complexity paper from Brian Kelly at All.

422
00:19:35,200 --> 00:19:36,800
Speaker 5: Which is I will say he may be one of

423
00:19:36,840 --> 00:19:39,679
Speaker 5: the only hosts of a podcast to read that paper.

424
00:19:39,720 --> 00:19:42,560
Speaker 3: This is not a simple paper. It's like he writes,

425
00:19:42,640 --> 00:19:43,000
Speaker 3: very clue.

426
00:19:43,040 --> 00:19:44,880
Speaker 1: It's the sort of like it's the notion of taking

427
00:19:44,920 --> 00:19:46,760
Speaker 1: like a sort of simple factor model and blowing it

428
00:19:46,840 --> 00:19:51,400
Speaker 1: up into like a nonlinear AI model, and one almost

429
00:19:51,440 --> 00:19:55,920
Speaker 1: throwaway sentence in the pieces like this, like simplified AI

430
00:19:56,040 --> 00:19:59,359
Speaker 1: model that he built for like illustrative purposes, lowered its

431
00:19:59,600 --> 00:20:03,200
Speaker 1: risk before fourteen of the last fifteen recessions. And I

432
00:20:03,320 --> 00:20:06,720
Speaker 1: always thought the naive like the best way to invest

433
00:20:06,760 --> 00:20:09,320
Speaker 1: would be just market timing, just like you know, have

434
00:20:09,400 --> 00:20:11,160
Speaker 1: all your money in the stock market before the market

435
00:20:11,160 --> 00:20:12,800
Speaker 1: goes up and not before it goes down, if you

436
00:20:12,880 --> 00:20:16,360
Speaker 1: can do it. My impression is that like respectable headgeplot managers,

437
00:20:16,359 --> 00:20:19,600
Speaker 1: respectable clime managers, respectable academics say the hardest thing in

438
00:20:19,600 --> 00:20:22,320
Speaker 1: the world is market timing, and like no one claims

439
00:20:22,320 --> 00:20:23,960
Speaker 1: to get alpha from at and it's not a thing.

440
00:20:25,480 --> 00:20:27,639
Speaker 1: Is that changed, and is that changed due to like

441
00:20:27,720 --> 00:20:28,399
Speaker 1: machine learning?

442
00:20:28,840 --> 00:20:33,240
Speaker 5: It has changed a big Do we do trend following

443
00:20:33,520 --> 00:20:36,720
Speaker 5: on macro assets old school CTA stuff. We think we've

444
00:20:36,720 --> 00:20:39,400
Speaker 5: made a new school by incorporating fundamental momentum by doing

445
00:20:39,440 --> 00:20:42,720
Speaker 5: a lot of more esoteric market so we think even

446
00:20:42,720 --> 00:20:45,360
Speaker 5: that's had a march of progress to it, But all

447
00:20:45,359 --> 00:20:48,440
Speaker 5: else equal. I wrote my dissertation on momentum in individual

448
00:20:48,480 --> 00:20:51,720
Speaker 5: stocks for some reason that I cannot explain. If you're

449
00:20:51,800 --> 00:20:54,920
Speaker 5: using past returns to predict the future, just what's going

450
00:20:55,000 --> 00:20:57,560
Speaker 5: up will keep going up, and vice versa to pick

451
00:20:57,600 --> 00:21:01,520
Speaker 5: individual stocks. The whole industry calls momentum. And if you

452
00:21:01,560 --> 00:21:04,040
Speaker 5: think markets tend to keep going in the same direction,

453
00:21:04,119 --> 00:21:05,320
Speaker 5: everyone calls it trend falling.

454
00:21:05,840 --> 00:21:08,560
Speaker 3: Same thing. Starting in I think two thousand and eight,

455
00:21:08,640 --> 00:21:09,159
Speaker 3: we started.

456
00:21:09,240 --> 00:21:11,560
Speaker 5: We always had it in our macro models, but we

457
00:21:11,640 --> 00:21:15,119
Speaker 5: started formally offering separate trend falling products.

458
00:21:15,400 --> 00:21:18,200
Speaker 3: It doesn't take hero bets on anyone market.

459
00:21:18,480 --> 00:21:21,800
Speaker 5: It's not Gazarelli selling all the stocks a minute before

460
00:21:21,840 --> 00:21:25,280
Speaker 5: October nineteenth of eighty seven. I'm dating myself. My wife's

461
00:21:25,280 --> 00:21:28,080
Speaker 5: birthday is October nineteenth, which always always gets a little amused.

462
00:21:28,600 --> 00:21:30,040
Speaker 3: Crash eighty seven meant to.

463
00:21:30,000 --> 00:21:31,560
Speaker 1: Be Did you really crashed them together?

464
00:21:31,640 --> 00:21:34,880
Speaker 5: Yeah, there have been some jokes over over time. So

465
00:21:35,000 --> 00:21:39,120
Speaker 5: trend following it's essentially market timing, but it's highly diverse,

466
00:21:39,200 --> 00:21:42,399
Speaker 5: many small bets. Whatever's been happening tends to keep happening.

467
00:21:42,440 --> 00:21:44,520
Speaker 5: That's the main way we'll do market timing.

468
00:21:44,600 --> 00:21:47,639
Speaker 1: You want to like your main equity fund is like sometimes.

469
00:21:47,320 --> 00:21:49,760
Speaker 5: That that was kind of a toy model to illustraate

470
00:21:49,840 --> 00:21:52,320
Speaker 5: a point. I know we're not taking a lot of

471
00:21:52,359 --> 00:21:54,840
Speaker 5: risks on that model. Market timing I still think is

472
00:21:54,880 --> 00:21:59,680
Speaker 5: quite hard. Might there be advances in it in the future, absolutely,

473
00:22:00,080 --> 00:22:04,879
Speaker 5: but are we taking significant risk in it now aside

474
00:22:04,880 --> 00:22:06,840
Speaker 5: from trend following, not really.

475
00:22:07,160 --> 00:22:09,800
Speaker 1: You probably sim in more papers to financial journals than

476
00:22:09,800 --> 00:22:11,600
Speaker 1: the average like asset.

477
00:22:11,280 --> 00:22:12,640
Speaker 3: Manager, by like one hundred percent.

478
00:22:12,720 --> 00:22:14,520
Speaker 1: I want to ask two questions. One is like, I

479
00:22:14,520 --> 00:22:16,960
Speaker 1: want to learn about your relationship with academia, because I

480
00:22:16,960 --> 00:22:20,119
Speaker 1: think it is fascinating that like you employ half of

481
00:22:20,160 --> 00:22:24,000
Speaker 1: the l faculty the finance PhD pipeline runs mainly to

482
00:22:24,000 --> 00:22:26,399
Speaker 1: AQR and then too because I think of like what

483
00:22:26,480 --> 00:22:30,560
Speaker 1: HEGEMA manager does as sort of like finding anomalies, finding

484
00:22:30,600 --> 00:22:34,480
Speaker 1: like market inefficiencies, finding factors that are predictive every turns,

485
00:22:34,520 --> 00:22:36,439
Speaker 1: and I think of what a finance academic does is

486
00:22:36,480 --> 00:22:38,960
Speaker 1: mainly also that when do you publish and when do

487
00:22:39,000 --> 00:22:39,719
Speaker 1: you just trade on it?

488
00:22:39,720 --> 00:22:42,199
Speaker 5: Oh, there's so much to talk about here. First, there

489
00:22:42,240 --> 00:22:44,840
Speaker 5: are a lot of reasons we do it. One is

490
00:22:44,920 --> 00:22:48,159
Speaker 5: just personal consumption. We grew up in this world. We

491
00:22:48,359 --> 00:22:50,600
Speaker 5: liked being part of it. We were interested in this

492
00:22:50,600 --> 00:22:54,880
Speaker 5: stuff in a purely academic sense before we got seduced

493
00:22:54,920 --> 00:22:55,760
Speaker 5: into making money.

494
00:22:56,160 --> 00:22:56,600
Speaker 3: I mean, I.

495
00:22:56,520 --> 00:22:59,399
Speaker 1: Would like pause it that you think there is a

496
00:22:59,480 --> 00:23:04,040
Speaker 1: value to employing the finance PhDs to like build your models,

497
00:23:04,280 --> 00:23:07,679
Speaker 1: and the way to attract fancy finance PhDs is to

498
00:23:07,720 --> 00:23:12,679
Speaker 1: offer them the most academia like possible working.

499
00:23:11,680 --> 00:23:14,200
Speaker 5: And to letting them publish. There are people, including our

500
00:23:14,240 --> 00:23:17,159
Speaker 5: two Yale professors. I don't don't know if would have

501
00:23:17,359 --> 00:23:19,120
Speaker 5: come take you are if we said you can never

502
00:23:19,160 --> 00:23:23,520
Speaker 5: write about this. We do a crass calculation, though, if

503
00:23:23,560 --> 00:23:27,360
Speaker 5: we think there is something that we are relatively unique

504
00:23:27,840 --> 00:23:30,639
Speaker 5: on entirely unique, or a very small handful of people

505
00:23:31,200 --> 00:23:36,240
Speaker 5: know this, we won't publish it. The optimal the optimal

506
00:23:36,280 --> 00:23:39,320
Speaker 5: time to publish a paper is after you've made money

507
00:23:39,359 --> 00:23:42,440
Speaker 5: from something for eleven years and an hour and a

508
00:23:42,480 --> 00:23:44,480
Speaker 5: half before someone else is going to publish the paper,

509
00:23:45,200 --> 00:23:47,800
Speaker 5: and we cannot get that right. I can think of

510
00:23:47,840 --> 00:23:50,399
Speaker 5: one example of momentum in factor. It's the fact that

511
00:23:50,560 --> 00:23:55,359
Speaker 5: factors themselves, like value, profitability, also exhibit momentum as something

512
00:23:55,440 --> 00:23:57,919
Speaker 5: that we've traded on for many years that we've always

513
00:23:57,920 --> 00:23:59,919
Speaker 5: refrained from writing a paper on because no one else has.

514
00:24:00,280 --> 00:24:02,200
Speaker 5: And again, we knew we couldn't be the only people

515
00:24:02,240 --> 00:24:03,639
Speaker 5: who do this, but we didn't think the cat was

516
00:24:03,680 --> 00:24:06,360
Speaker 5: out of the bag. And then someone else wrote the paper.

517
00:24:06,600 --> 00:24:08,560
Speaker 5: And I'm sure we've done this to other people too,

518
00:24:08,600 --> 00:24:11,800
Speaker 5: because you never know what they're doing internally. So life,

519
00:24:11,920 --> 00:24:13,479
Speaker 5: you know, you do it long enough, life works out

520
00:24:13,560 --> 00:24:14,560
Speaker 5: kind of fair, but that.

521
00:24:14,640 --> 00:24:15,200
Speaker 3: Is the goal.

522
00:24:15,320 --> 00:24:18,520
Speaker 1: But you were mad there because, like as academics, you

523
00:24:18,640 --> 00:24:19,920
Speaker 1: wanted credit for that paper.

524
00:24:20,040 --> 00:24:20,359
Speaker 3: It's fun.

525
00:24:21,440 --> 00:24:23,560
Speaker 5: You could call it childish, but it's human. It's it's

526
00:24:23,600 --> 00:24:24,680
Speaker 5: fun to discover something.

527
00:24:24,680 --> 00:24:27,639
Speaker 1: Well, you also mad because publishing that reduces the value

528
00:24:27,640 --> 00:24:28,679
Speaker 1: of the signal or maybe a.

529
00:24:28,680 --> 00:24:30,080
Speaker 3: Little bit, maybe a little bit.

530
00:24:30,240 --> 00:24:32,840
Speaker 5: So far, it's still worked wonderfully. The trend is still

531
00:24:32,880 --> 00:24:35,840
Speaker 5: your friend when it comes to two factors, but it

532
00:24:35,840 --> 00:24:37,840
Speaker 5: would be a fireable offense for make you are to

533
00:24:37,880 --> 00:24:40,680
Speaker 5: publish something that we thought was truly proprietary.

534
00:24:40,920 --> 00:24:42,680
Speaker 3: Let me give you my favorite example of this, because

535
00:24:42,680 --> 00:24:43,520
Speaker 3: it came up recently.

536
00:24:43,840 --> 00:24:46,080
Speaker 5: One of the things that we've gotten into in a

537
00:24:46,119 --> 00:24:48,560
Speaker 5: fairly big way, as have some other quants, is what's

538
00:24:48,600 --> 00:24:52,160
Speaker 5: called alternative data. Those are new data sets that put

539
00:24:52,200 --> 00:24:55,639
Speaker 5: people put together with sweat equity. The classic example and

540
00:24:55,680 --> 00:24:57,520
Speaker 5: the only one. And this is the point that I'm

541
00:24:57,560 --> 00:25:00,560
Speaker 5: allowed to talk about our credit card receiver data were

542
00:25:00,600 --> 00:25:02,520
Speaker 5: a credit cards because that's been discussed.

543
00:25:02,080 --> 00:25:03,520
Speaker 3: A million times.

544
00:25:04,080 --> 00:25:04,639
Speaker 1: Anative there.

545
00:25:04,720 --> 00:25:08,400
Speaker 5: Yeah, yeah, Well that's the point is this stuff is arbitrageble.

546
00:25:08,560 --> 00:25:09,600
Speaker 3: I mean, it can go away.

547
00:25:10,320 --> 00:25:13,720
Speaker 5: Value, be it the narrow Quan sense of value or

548
00:25:13,720 --> 00:25:16,159
Speaker 5: the more broad gram and Odd sense of value, is

549
00:25:16,200 --> 00:25:19,159
Speaker 5: trying to take advantage of I think basic human nature.

550
00:25:19,680 --> 00:25:22,440
Speaker 5: I think spreads being cheap and expensive are wider still

551
00:25:22,480 --> 00:25:24,840
Speaker 5: than the historical average, not tighter. I don't think there's

552
00:25:24,840 --> 00:25:26,719
Speaker 5: a lot of evidence that so much capital is in

553
00:25:26,720 --> 00:25:27,320
Speaker 5: that that.

554
00:25:27,280 --> 00:25:28,160
Speaker 3: It's making you go away.

555
00:25:28,600 --> 00:25:32,240
Speaker 5: If you get a short term information advantage because you

556
00:25:32,320 --> 00:25:35,080
Speaker 5: have put the time in or paid someone who's put

557
00:25:35,119 --> 00:25:38,119
Speaker 5: the time in to create a new data set that

558
00:25:38,200 --> 00:25:40,960
Speaker 5: other people don't have. They're going to have it eventually.

559
00:25:41,240 --> 00:25:45,080
Speaker 5: So I am talking to the Australian Financial Review. I

560
00:25:45,119 --> 00:25:47,920
Speaker 5: don't even know if we had discussed alternative data before him,

561
00:25:47,960 --> 00:25:50,240
Speaker 5: but he asked me about it, and I said to

562
00:25:50,320 --> 00:25:54,640
Speaker 5: him that our head of stock selection, a fellow named

563
00:25:54,800 --> 00:25:59,280
Speaker 5: Andrea Ferzini, has asked me, I might have said, told me,

564
00:25:59,800 --> 00:26:01,760
Speaker 5: but has asked me not to talk.

565
00:26:01,600 --> 00:26:02,280
Speaker 3: About these things.

566
00:26:02,320 --> 00:26:04,000
Speaker 5: There are things I'll talk about, but there are things

567
00:26:04,000 --> 00:26:06,920
Speaker 5: we think are you know, overused word, but our true

568
00:26:07,000 --> 00:26:10,560
Speaker 5: alpha that the world doesn't know, and I think it's reasonable,

569
00:26:10,600 --> 00:26:13,159
Speaker 5: So I really can't talk about him. He writes the

570
00:26:13,280 --> 00:26:17,560
Speaker 5: article and what it says is mostly that, but instead

571
00:26:17,600 --> 00:26:20,080
Speaker 5: of saying he's asked me not to talk about it,

572
00:26:20,080 --> 00:26:23,680
Speaker 5: it says Andrea Ferzini won't tell me what we're doing

573
00:26:23,960 --> 00:26:28,760
Speaker 5: in alternative data, which it has a slightly different connotation,

574
00:26:28,880 --> 00:26:31,200
Speaker 5: and that's a connotation of adult old man, don't worry

575
00:26:31,240 --> 00:26:31,600
Speaker 5: about it.

576
00:26:31,640 --> 00:26:32,800
Speaker 3: We're doing some stuff here.

577
00:26:33,160 --> 00:26:36,400
Speaker 5: So again it's an example where there certainly are things

578
00:26:36,440 --> 00:26:37,320
Speaker 5: we won't talk about.

579
00:26:37,640 --> 00:26:39,960
Speaker 1: Is that ever true? Are you like most cutting edge

580
00:26:40,000 --> 00:26:42,439
Speaker 1: machine learning people like that? Or are that or do

581
00:26:42,160 --> 00:26:43,359
Speaker 1: you read all the papers.

582
00:26:43,480 --> 00:26:45,480
Speaker 5: I read most of the papers. I don't read all

583
00:26:45,520 --> 00:26:48,479
Speaker 5: of the papers. I at least skim all the papers.

584
00:26:48,480 --> 00:26:51,200
Speaker 5: I know the gist. We do a lot of different

585
00:26:51,200 --> 00:26:53,280
Speaker 5: things at this point. Once you're a quant you want

586
00:26:53,280 --> 00:26:55,240
Speaker 5: more factors and you want to trade it in more places.

587
00:26:55,880 --> 00:26:59,560
Speaker 5: And I will admit that the week doesn't go by

588
00:26:59,560 --> 00:27:01,520
Speaker 5: where I don't and ask someone for a review of

589
00:27:01,520 --> 00:27:04,200
Speaker 5: what we're doing. Somewhere at some point I approved it,

590
00:27:05,000 --> 00:27:09,240
Speaker 5: so Pastcliff had some understanding. Yeah, but also just you know,

591
00:27:09,320 --> 00:27:11,640
Speaker 5: some of this is pretty decent math, and the old

592
00:27:11,680 --> 00:27:14,720
Speaker 5: mathematicians who their best work in their twenties, I can

593
00:27:14,760 --> 00:27:17,200
Speaker 5: tell you I'm probably still decent in math, but I'm

594
00:27:17,200 --> 00:27:19,800
Speaker 5: not what I was when I was twenty two. Wisdom

595
00:27:19,840 --> 00:27:24,360
Speaker 5: has hopefully replaced some of mathematical ability because.

596
00:27:24,080 --> 00:27:27,480
Speaker 1: The personnel at AQR different from a like I don't

597
00:27:27,480 --> 00:27:30,119
Speaker 1: know who you think of as like quantity competitors, but like,

598
00:27:31,520 --> 00:27:33,919
Speaker 1: my impression is that like you have many more finance

599
00:27:34,000 --> 00:27:38,080
Speaker 1: PhDs than like other places that might have more like

600
00:27:38,560 --> 00:27:41,199
Speaker 1: pure math people or math undergrads or something like.

601
00:27:41,440 --> 00:27:43,760
Speaker 3: Yeah, I think there's some truth to that.

602
00:27:44,000 --> 00:27:45,200
Speaker 1: What makes them better or worse?

603
00:27:45,560 --> 00:27:46,359
Speaker 3: First, I think.

604
00:27:46,240 --> 00:27:48,119
Speaker 5: There's going to be a correlation that the closer you

605
00:27:48,119 --> 00:27:50,479
Speaker 5: get to the high frequency world, the more you're going

606
00:27:50,520 --> 00:27:54,000
Speaker 5: to be more in the pure math realm. My very

607
00:27:54,040 --> 00:27:58,480
Speaker 5: tortured analogy is quantum mechanics versus Newtonian physics. When you

608
00:27:58,520 --> 00:28:00,960
Speaker 5: get into the high frequency world, first you have a

609
00:28:01,000 --> 00:28:04,120
Speaker 5: ton of data. By nature, you just have a lot

610
00:28:04,160 --> 00:28:05,560
Speaker 5: more instances.

611
00:28:05,560 --> 00:28:06,840
Speaker 3: And a lot less theory.

612
00:28:07,080 --> 00:28:10,159
Speaker 5: So turning yourself over to the data more so than

613
00:28:10,200 --> 00:28:14,119
Speaker 5: having an economic rationale is far more rational, and that

614
00:28:14,200 --> 00:28:16,080
Speaker 5: becomes a more mathematical exercise.

615
00:28:16,320 --> 00:28:18,639
Speaker 3: We tend to think in our long.

616
00:28:18,520 --> 00:28:22,160
Speaker 5: Seven hundred short seven hundred stock portfolios, average holding period

617
00:28:22,280 --> 00:28:24,440
Speaker 5: is maybe nine months, maybe closer a year.

618
00:28:24,680 --> 00:28:28,280
Speaker 3: Times. That's nowhere near high frequency. Yeah, high frequent. Two.

619
00:28:28,280 --> 00:28:30,080
Speaker 5: When I wrote my dissertation and I did have this

620
00:28:30,160 --> 00:28:33,479
Speaker 5: in there was a monthly contrarian strategy. Now we're talking,

621
00:28:33,680 --> 00:28:36,280
Speaker 5: you know, sub seconds kind of thing. I think when

622
00:28:36,320 --> 00:28:40,240
Speaker 5: you're gonna have a medium holding period strategy, as I

623
00:28:40,280 --> 00:28:42,440
Speaker 5: think of us, we're not Warren Buffett, but we're not HFT.

624
00:28:43,520 --> 00:28:48,080
Speaker 5: Even the machine learning stuff needs some economics too. You

625
00:28:48,120 --> 00:28:50,360
Speaker 5: simply don't have enough data and there's too much of

626
00:28:50,400 --> 00:28:54,000
Speaker 5: a dimensionality even with Brian's virtue of complexity, you need

627
00:28:54,040 --> 00:28:57,000
Speaker 5: to give it some structure or else it's going to

628
00:28:57,080 --> 00:29:00,880
Speaker 5: overfit and go mad. So I think think that's the reason,

629
00:29:01,120 --> 00:29:03,440
Speaker 5: not saying we'll never do something in a higher frequency world,

630
00:29:03,440 --> 00:29:05,920
Speaker 5: in which case we'd probably have to shift more. But

631
00:29:06,080 --> 00:29:11,320
Speaker 5: in our world being a mathematician, being an excellent programmer,

632
00:29:11,480 --> 00:29:14,680
Speaker 5: but also having the economics behind it, it's kind of

633
00:29:14,680 --> 00:29:15,520
Speaker 5: what we're looking for.

634
00:29:16,240 --> 00:29:21,360
Speaker 1: I think of like renaissances famously employing exclusively people who

635
00:29:21,440 --> 00:29:26,000
Speaker 1: have never thought about markets or financer economics, and like

636
00:29:26,080 --> 00:29:28,160
Speaker 1: they come to it pure, and I feel that you

637
00:29:29,200 --> 00:29:31,480
Speaker 1: are very much like people who have math shops, but

638
00:29:31,680 --> 00:29:35,720
Speaker 1: like have economic intuition and think about it as an economy.

639
00:29:36,040 --> 00:29:39,239
Speaker 5: I think that's accurate. I do have a couple of

640
00:29:39,400 --> 00:29:40,680
Speaker 5: Renaissance observations.

641
00:29:40,720 --> 00:29:41,000
Speaker 1: Okay.

642
00:29:41,360 --> 00:29:44,960
Speaker 5: One of my favorite questions people asked me was howd

643
00:29:45,000 --> 00:29:47,040
Speaker 5: they do it? And I love that question because I

644
00:29:47,040 --> 00:29:50,760
Speaker 5: get to respond. So your hypothesis is, I know how

645
00:29:50,840 --> 00:29:53,520
Speaker 5: they took a few billion dollars and still take a

646
00:29:53,520 --> 00:29:55,480
Speaker 5: few billion dollars out of the market to share among

647
00:29:55,800 --> 00:29:58,800
Speaker 5: a relevant tiny group of people every year would apparently

648
00:29:58,920 --> 00:30:01,719
Speaker 5: very low risk, and I choose not to I have

649
00:30:01,880 --> 00:30:04,640
Speaker 5: some inklings of the general whether they've worked on. My

650
00:30:04,960 --> 00:30:07,360
Speaker 5: biggest guess is that they were ten fifteen years ahead

651
00:30:07,360 --> 00:30:09,320
Speaker 5: of everyone else on most of this stuff and are

652
00:30:09,400 --> 00:30:12,640
Speaker 5: just have developed more sophisticated systems over time. I think

653
00:30:12,720 --> 00:30:14,760
Speaker 5: natural language processing they were very early on.

654
00:30:15,160 --> 00:30:17,640
Speaker 1: It came from like from.

655
00:30:18,480 --> 00:30:19,400
Speaker 3: A few other things.

656
00:30:19,600 --> 00:30:21,600
Speaker 5: John Leu and I, my one of my founding partners,

657
00:30:21,640 --> 00:30:24,440
Speaker 5: wrote when Fama and Schiller shared the Nobel Prize, we

658
00:30:24,440 --> 00:30:27,360
Speaker 5: wrote a whole overview of market efficiency and the debate

659
00:30:27,400 --> 00:30:30,040
Speaker 5: about it, and I brought them up as an example.

660
00:30:30,320 --> 00:30:34,280
Speaker 5: The Medallion Fund has almost nothing to say about market efficiency.

661
00:30:34,480 --> 00:30:37,080
Speaker 5: It says, these guys can extract a toll on the

662
00:30:37,120 --> 00:30:42,200
Speaker 5: market with reliable consistency, But in terms of market size

663
00:30:42,440 --> 00:30:44,680
Speaker 5: it's a giant number for a few people to make

664
00:30:44,760 --> 00:30:47,160
Speaker 5: each year. It's a tiny number in terms of whether

665
00:30:47,200 --> 00:30:49,760
Speaker 5: the markets are efficient, and apparently the rest of us

666
00:30:49,840 --> 00:30:51,600
Speaker 5: can't do it quite as well as them. They're the

667
00:30:51,640 --> 00:30:52,560
Speaker 5: goat when it comes to this.

668
00:30:52,800 --> 00:30:56,200
Speaker 1: Yeah, I think of like, you know, like Jane Street

669
00:30:56,200 --> 00:30:58,239
Speaker 1: can rely excite somebody from the market, and I think

670
00:30:58,280 --> 00:31:00,280
Speaker 1: of that as like a few first service ye, like

671
00:31:00,280 --> 00:31:02,600
Speaker 1: a market maker, rather than being like a predictor of

672
00:31:02,640 --> 00:31:03,240
Speaker 1: asset precesce.

673
00:31:03,320 --> 00:31:05,440
Speaker 5: I'm gonna guess, and again it's a purely guess that

674
00:31:05,560 --> 00:31:07,959
Speaker 5: not all of it, but a fair amount of Renaissance

675
00:31:07,960 --> 00:31:09,840
Speaker 5: had a similar similar flavor.

676
00:31:09,920 --> 00:31:11,800
Speaker 1: I think when you talk to people who run pod shops, like,

677
00:31:11,840 --> 00:31:14,520
Speaker 1: some of what they do they conceive of as it's

678
00:31:14,520 --> 00:31:17,280
Speaker 1: not literally market making, but it has that sort of flavor.

679
00:31:17,480 --> 00:31:20,880
Speaker 5: And the constraint on those things is typically capacity. Now

680
00:31:20,880 --> 00:31:23,200
Speaker 5: it's a ton of money for those firms. They're amazing firms,

681
00:31:23,960 --> 00:31:26,800
Speaker 5: but give you an example of Renaissance. I've complimented them

682
00:31:26,840 --> 00:31:28,560
Speaker 5: like crazy, so hopefully they won't be mad at me

683
00:31:28,640 --> 00:31:28,880
Speaker 5: for this.

684
00:31:28,880 --> 00:31:30,240
Speaker 3: One last comment.

685
00:31:30,880 --> 00:31:33,600
Speaker 5: Everyone talks about the Medallion Fund, but no one's allowed

686
00:31:33,600 --> 00:31:36,040
Speaker 5: to invest in that. They keep that for themselves. Another

687
00:31:36,120 --> 00:31:39,320
Speaker 5: favorite question I get is this doesn't happen much anymore,

688
00:31:39,320 --> 00:31:40,840
Speaker 5: but every once in a while I would get an

689
00:31:40,880 --> 00:31:43,840
Speaker 5: investor saying, so, Medallion Fund better than you guys. I'd

690
00:31:43,840 --> 00:31:46,640
Speaker 5: be like, oh, hell yes, but they won't take your

691
00:31:46,680 --> 00:31:49,120
Speaker 5: money or my money, and I think we're pretty darn

692
00:31:49,160 --> 00:31:51,800
Speaker 5: good and we will. So why is that relevant? But

693
00:31:51,840 --> 00:31:54,960
Speaker 5: Renaissance also runs a fair amount of money in more

694
00:31:55,360 --> 00:31:59,640
Speaker 5: open institutional products where they look very good. I'm not

695
00:31:59,680 --> 00:32:02,080
Speaker 5: gonna to ride them at all, but they don't look

696
00:32:02,080 --> 00:32:07,080
Speaker 5: better than us. They look like really good, solid, regular quants.

697
00:32:07,520 --> 00:32:10,720
Speaker 5: So what they cannot do, I think it's kind of obvious,

698
00:32:11,080 --> 00:32:14,280
Speaker 5: is take the medallion process and scale it up many,

699
00:32:14,320 --> 00:32:17,280
Speaker 5: many times bigger. They've discovered a way, as you put it,

700
00:32:17,360 --> 00:32:20,040
Speaker 5: to just take a certain amount out to provide a service.

701
00:32:20,080 --> 00:32:23,400
Speaker 5: I called it taking a toll, and that's amazing. But

702
00:32:23,520 --> 00:32:27,120
Speaker 5: they've not discovered a way to do that at institutional scale,

703
00:32:27,120 --> 00:32:29,160
Speaker 5: which thank god, because that leaves something for the rest

704
00:32:29,200 --> 00:32:29,719
Speaker 5: of us to do.

705
00:32:30,040 --> 00:32:31,720
Speaker 1: I'll get back to something you talked about the very beginning,

706
00:32:31,720 --> 00:32:33,440
Speaker 1: because you mentioned that paper you wrote with John Leu,

707
00:32:33,520 --> 00:32:38,040
Speaker 1: which you talking about like the two possible explanations for

708
00:32:38,400 --> 00:32:41,560
Speaker 1: making money for anomolies for whatever, which are sort of

709
00:32:41,560 --> 00:32:46,360
Speaker 1: behavioral irrationality or whatever, and you were bearing a risk

710
00:32:46,440 --> 00:32:49,120
Speaker 1: for someone. Everything you write, and like what you said

711
00:32:49,160 --> 00:32:51,880
Speaker 1: here you have, you seem sort of like agnostic about

712
00:32:52,200 --> 00:32:55,840
Speaker 1: which one or what combination there's like, do you prefer one,

713
00:32:56,280 --> 00:32:58,280
Speaker 1: is it more reliable to get paid for taking your

714
00:32:58,320 --> 00:32:58,920
Speaker 1: risk or.

715
00:32:59,520 --> 00:33:02,000
Speaker 5: Well forgetting which one you think is really going on?

716
00:33:02,120 --> 00:33:05,240
Speaker 5: They do have different characteristics The positive behind a risk

717
00:33:05,240 --> 00:33:08,400
Speaker 5: premium is you may be able to get it forever

718
00:33:08,560 --> 00:33:11,880
Speaker 5: because it's rational that you get it. The negative behind

719
00:33:11,920 --> 00:33:14,280
Speaker 5: it is it's a risk premium. You have to figure

720
00:33:14,320 --> 00:33:17,600
Speaker 5: out why. If the risk premium is most of the

721
00:33:17,640 --> 00:33:19,880
Speaker 5: time this is fine, but it has depression risk. It's

722
00:33:19,880 --> 00:33:22,880
Speaker 5: going to do particularly bad in a depression. Well, that's

723
00:33:22,920 --> 00:33:25,320
Speaker 5: not a very pleasant risk to have. You may get

724
00:33:25,320 --> 00:33:30,520
Speaker 5: paid for doing that. The positive of behavioral is it's

725
00:33:30,640 --> 00:33:33,600
Speaker 5: essentially over the long term. Over the short term it

726
00:33:33,600 --> 00:33:36,280
Speaker 5: can be very very painful lunch because the noise work.

727
00:33:36,680 --> 00:33:38,400
Speaker 5: But if on average it works long term, it's a

728
00:33:38,440 --> 00:33:42,040
Speaker 5: free lunch and that you're not taking additional systematic risk.

729
00:33:42,600 --> 00:33:43,880
Speaker 3: The negative is it can go away.

730
00:33:44,240 --> 00:33:46,760
Speaker 5: It can be arbitraged away if that ra stops being made,

731
00:33:46,760 --> 00:33:49,760
Speaker 5: If too much capital chases it, it can be arbitraged away.

732
00:33:49,840 --> 00:33:52,800
Speaker 5: It's hard to arbitrage some of them away, like basic valuation.

733
00:33:52,960 --> 00:33:56,200
Speaker 5: It's very easy to arbitrage some others, like alternative data,

734
00:33:56,240 --> 00:33:58,520
Speaker 5: where the point is to be quicker to getting a

735
00:33:58,600 --> 00:34:01,160
Speaker 5: data set and to trading the day to set. I

736
00:34:01,200 --> 00:34:03,800
Speaker 5: will say this, and I hope teen Fama isn't listening.

737
00:34:04,480 --> 00:34:07,440
Speaker 5: I was probably seventy five twenty five. For risk guy

738
00:34:07,720 --> 00:34:10,600
Speaker 5: thirty years ago, I'm probably seventy five twenty five.

739
00:34:10,640 --> 00:34:13,359
Speaker 1: A behavioral is that GameStop or something not?

740
00:34:13,840 --> 00:34:18,880
Speaker 5: Well, the GameStop maybe the unfairly extreme example. It's not

741
00:34:18,920 --> 00:34:21,440
Speaker 5: fair to pick the most extreme. Craziness is to make

742
00:34:21,480 --> 00:34:25,640
Speaker 5: your point, But yeah, it's real life experience. It's watching

743
00:34:26,280 --> 00:34:29,359
Speaker 5: the spread between cheap and expensive stocks in ninety nine

744
00:34:29,400 --> 00:34:29,960
Speaker 5: two thousand.

745
00:34:30,160 --> 00:34:32,120
Speaker 3: We started our firm in late ninety.

746
00:34:31,800 --> 00:34:34,680
Speaker 5: Eight, so right before the real crazy part of what's

747
00:34:34,719 --> 00:34:38,080
Speaker 5: called now the dot com bubble, the spread between how

748
00:34:38,120 --> 00:34:42,160
Speaker 5: we define cheap and expensive, either just using quant multiples

749
00:34:42,239 --> 00:34:46,040
Speaker 5: or adjusting for forms of growth, set records. We had

750
00:34:46,200 --> 00:34:49,040
Speaker 5: fifty seventy five years of data and we lived through it,

751
00:34:49,120 --> 00:34:53,279
Speaker 5: blowing past those. Then after that, I'm like, all right,

752
00:34:53,320 --> 00:34:54,799
Speaker 5: that was a once in fifty year event.

753
00:34:54,880 --> 00:34:56,399
Speaker 3: We survived it. We ended up being right.

754
00:34:56,440 --> 00:34:59,080
Speaker 5: We ended up making money round trip that was excruciating

755
00:34:59,120 --> 00:35:02,640
Speaker 5: on the first leg. If you ask me back then,

756
00:35:02,800 --> 00:35:04,880
Speaker 5: am I ever going to see that in my career again?

757
00:35:05,600 --> 00:35:07,880
Speaker 5: I hope I'd be smart enough not to say never.

758
00:35:08,560 --> 00:35:11,439
Speaker 5: None of us in our field, yours or mine, should

759
00:35:11,440 --> 00:35:14,480
Speaker 5: say never. Markets are pretty hard things to say never about.

760
00:35:14,760 --> 00:35:16,560
Speaker 5: But I think I would have said it's highly unlikely.

761
00:35:16,960 --> 00:35:19,080
Speaker 5: For one, it was literally the most extreme event in

762
00:35:19,160 --> 00:35:23,759
Speaker 5: at least fifty years. Second, the question presupposes I and

763
00:35:23,800 --> 00:35:27,120
Speaker 5: people like me will still be around. Unpresumably, if we're

764
00:35:27,120 --> 00:35:31,080
Speaker 5: still around, we're in more supervisory and authoritative positions. So

765
00:35:31,120 --> 00:35:33,960
Speaker 5: how's it going to happen again? And then it happened

766
00:35:34,000 --> 00:35:37,960
Speaker 5: again almost exactly twenty years later, from twenty eighteen through

767
00:35:38,000 --> 00:35:41,880
Speaker 5: twenty twenty. You can point to COVID. It went absolutely

768
00:35:41,920 --> 00:35:44,640
Speaker 5: mad during COVID. You guys remember for the six month

769
00:35:44,719 --> 00:35:47,320
Speaker 5: period after COVID when the only two stocks you're supposed

770
00:35:47,320 --> 00:35:49,560
Speaker 5: to own were Peloton and Tesla. One of them has

771
00:35:49,600 --> 00:35:50,840
Speaker 5: still worked out mostly for you.

772
00:35:50,960 --> 00:35:51,719
Speaker 3: One of them did not.

773
00:35:52,280 --> 00:35:55,000
Speaker 5: But even before COVID, though you cannot pin it on COVID,

774
00:35:55,080 --> 00:35:58,759
Speaker 5: by late nineteen early twenty we were approaching tech bubble levels,

775
00:35:58,800 --> 00:36:01,720
Speaker 5: which again had not been seeing the prior fifty years before.

776
00:36:01,760 --> 00:36:05,960
Speaker 5: The tech bubble again, painful period for us, again made

777
00:36:06,040 --> 00:36:08,040
Speaker 5: money round trip. I'm going to brag about that we

778
00:36:08,160 --> 00:36:08,839
Speaker 5: survived it.

779
00:36:09,200 --> 00:36:09,760
Speaker 3: I think we're.

780
00:36:09,600 --> 00:36:12,960
Speaker 5: Stronger than we were beforehand. But if the first one

781
00:36:13,000 --> 00:36:16,600
Speaker 5: didn't make you start to go this behavioral stuff may

782
00:36:16,600 --> 00:36:19,000
Speaker 5: be real and maybe bigger than it.

783
00:36:18,960 --> 00:36:21,200
Speaker 3: Used to be. Yeah, and I wrote a piece.

784
00:36:20,960 --> 00:36:24,640
Speaker 5: Called the less efficient market hypothesis markets getting less efficient.

785
00:36:25,160 --> 00:36:27,200
Speaker 5: That may be true, but I think it's more accurate

786
00:36:27,200 --> 00:36:30,400
Speaker 5: to say there what my evidence is there prone to

787
00:36:30,840 --> 00:36:35,239
Speaker 5: some extreme bouts of craziness on occasion. It's not quite

788
00:36:35,280 --> 00:36:37,840
Speaker 5: the same as in steady state always being less efficient,

789
00:36:37,880 --> 00:36:39,960
Speaker 5: but probably some of both. But I do think that

790
00:36:40,040 --> 00:36:42,960
Speaker 5: has happened. I probably was not giving behavioral enough credit

791
00:36:43,000 --> 00:36:45,480
Speaker 5: early on, and I think it's gotten to be a

792
00:36:45,560 --> 00:36:46,160
Speaker 5: bigger part.

793
00:37:02,160 --> 00:37:04,319
Speaker 1: I want to read something from the Johnlely paper from

794
00:37:04,360 --> 00:37:07,839
Speaker 1: twenty fourteen. Suppose you imagine some investors get joy from

795
00:37:07,880 --> 00:37:10,359
Speaker 1: owning particular stocks, for example, being able to brag out

796
00:37:10,400 --> 00:37:12,200
Speaker 1: a cocktail party about the growth stocks that they own

797
00:37:12,239 --> 00:37:14,480
Speaker 1: that have done well. One way to describe this some

798
00:37:14,560 --> 00:37:17,239
Speaker 1: investors have a taste for growth stocks beyond simply their

799
00:37:17,280 --> 00:37:19,879
Speaker 1: effect on their portfolios. It can be rational for them

800
00:37:19,880 --> 00:37:22,960
Speaker 1: to accept somewhat lower returns for this pleasure. But even

801
00:37:23,000 --> 00:37:25,560
Speaker 1: if rational to the individuals who have this taste, if

802
00:37:25,560 --> 00:37:27,600
Speaker 1: some investors are willing to give up returns to others

803
00:37:27,600 --> 00:37:30,279
Speaker 1: because they care about cocktail party bragging, can we really

804
00:37:30,280 --> 00:37:32,920
Speaker 1: call that a rational market and feel this statement is useful.

805
00:37:33,080 --> 00:37:35,400
Speaker 1: I feel like that was like a little prescient about

806
00:37:35,600 --> 00:37:37,880
Speaker 1: how I have experienced some of the last five years,

807
00:37:38,080 --> 00:37:41,120
Speaker 1: which is that, like it seems to me that it

808
00:37:41,239 --> 00:37:43,279
Speaker 1: is hard to explain some of the stuff I write about,

809
00:37:43,280 --> 00:37:44,880
Speaker 1: which is maybe not like the most important thing in

810
00:37:44,880 --> 00:37:47,120
Speaker 1: the world. It's often like I think the most important

811
00:37:47,560 --> 00:37:49,120
Speaker 1: do too, but like it's like, you know, it's maybe

812
00:37:49,160 --> 00:37:51,799
Speaker 1: not like the biggest, you know, dollar, the test LIS's

813
00:37:51,800 --> 00:37:55,239
Speaker 1: pretty big. It does seem like a thing that is

814
00:37:55,360 --> 00:37:58,399
Speaker 1: regularly happening is that people have a taste for stocks

815
00:37:58,440 --> 00:38:03,560
Speaker 1: beyond any rational or irrational calculation about how we'll fight

816
00:38:03,560 --> 00:38:04,160
Speaker 1: their portfolio.

817
00:38:04,160 --> 00:38:07,600
Speaker 5: Well, particularly the whole meme world. Meme coins are clearly

818
00:38:07,760 --> 00:38:09,720
Speaker 5: a political statement slash.

819
00:38:11,000 --> 00:38:13,719
Speaker 1: Clearly there is a non economic motivation for some of that,

820
00:38:13,760 --> 00:38:14,600
Speaker 1: and you can callude rational.

821
00:38:14,640 --> 00:38:17,160
Speaker 3: But it's odd, Well a few things that paragraph.

822
00:38:17,600 --> 00:38:19,480
Speaker 5: Even writing it, I think we went a little too

823
00:38:19,560 --> 00:38:22,719
Speaker 5: far because I don't think these people, we kind of

824
00:38:22,719 --> 00:38:24,920
Speaker 5: write it like they're consciously doing it, like I just

825
00:38:24,960 --> 00:38:26,880
Speaker 5: love owning these I know I'm gonna lose money. I

826
00:38:26,880 --> 00:38:30,480
Speaker 5: think they kind of end up they resolve cognitive dissonance

827
00:38:30,520 --> 00:38:33,319
Speaker 5: by saying I'm gonna make money even if that's what's

828
00:38:33,360 --> 00:38:34,000
Speaker 5: really going on.

829
00:38:34,360 --> 00:38:36,000
Speaker 1: Right, I think this is maybe a little harsh on

830
00:38:36,040 --> 00:38:38,840
Speaker 1: growth investors in twenty fourteen, but on MEME investors.

831
00:38:38,440 --> 00:38:42,600
Speaker 5: And now instead on meme investors. We were also given

832
00:38:42,640 --> 00:38:46,279
Speaker 5: a little mild shot to some academics who kind of

833
00:38:46,280 --> 00:38:48,759
Speaker 5: try to save market efficiency by saying it's more meta

834
00:38:48,800 --> 00:38:51,279
Speaker 5: efficient if you count people just have tastes for this.

835
00:38:51,360 --> 00:38:53,040
Speaker 5: That argument's been made. I think that's kind of a

836
00:38:53,040 --> 00:38:55,600
Speaker 5: cop out. You know, if we've done market efficiency down

837
00:38:55,600 --> 00:38:58,360
Speaker 5: to that, what do we actually even mean by market efficiency?

838
00:38:58,400 --> 00:39:00,359
Speaker 5: If you go, I know, I'm gonna do terrible on this,

839
00:39:00,480 --> 00:39:04,880
Speaker 5: but it's fun to me, that's functionally a fairly inefficient market.

840
00:39:05,640 --> 00:39:09,160
Speaker 5: So we proposed in that piece which did not catch

841
00:39:09,520 --> 00:39:13,840
Speaker 5: on that classically market efficiency and the testing of it

842
00:39:13,880 --> 00:39:18,200
Speaker 5: has a joint hypothesis problem. You're testing whether markets are efficient,

843
00:39:18,560 --> 00:39:20,600
Speaker 5: but you also need a model for how prices are set,

844
00:39:20,800 --> 00:39:23,759
Speaker 5: and if that model for how prices are set includes

845
00:39:24,080 --> 00:39:26,759
Speaker 5: this is fun to own, then you've pushed it too far.

846
00:39:27,080 --> 00:39:29,520
Speaker 3: It has to be a reasonable hypothesiz.

847
00:39:29,200 --> 00:39:30,719
Speaker 1: This is what I read about this is I want

848
00:39:30,800 --> 00:39:33,040
Speaker 1: someone and it's probably not me. To write a finance

849
00:39:33,080 --> 00:39:36,920
Speaker 1: textbook that incorporates the factor of this is fun to

850
00:39:36,920 --> 00:39:39,680
Speaker 1: own and gives some guidance how to price that.

851
00:39:39,960 --> 00:39:43,880
Speaker 3: Be and how long it's going to continue.

852
00:39:43,960 --> 00:39:47,719
Speaker 1: And people like someone is making money on this, oh yeah,

853
00:39:47,719 --> 00:39:49,680
Speaker 1: and like not just like fading it. Someone is like,

854
00:39:50,120 --> 00:39:52,759
Speaker 1: I have a model for which meme coins will go up,

855
00:39:53,239 --> 00:39:54,399
Speaker 1: And well, people have.

856
00:39:54,360 --> 00:39:55,040
Speaker 2: Tried to do that.

857
00:39:55,120 --> 00:39:58,359
Speaker 4: You think about all the different buzzy strategies, something about

858
00:39:58,360 --> 00:40:02,000
Speaker 4: ETFs real cool, you can do it. Yeah, but you

859
00:40:02,000 --> 00:40:04,680
Speaker 4: know there have been attempts if you scraped social media,

860
00:40:04,760 --> 00:40:06,840
Speaker 4: et cetera and find out what people are buzzing about.

861
00:40:06,920 --> 00:40:09,200
Speaker 3: But it's the old value manager's lament.

862
00:40:09,360 --> 00:40:12,960
Speaker 5: But it's also I think true if ultimately it's a

863
00:40:12,960 --> 00:40:17,320
Speaker 5: bad investment in return sense that will happen eventually.

864
00:40:17,480 --> 00:40:20,440
Speaker 3: I do think that time horizon and the extremes have

865
00:40:20,640 --> 00:40:22,359
Speaker 3: lengthened a lot of the point of this way, Why

866
00:40:22,400 --> 00:40:24,600
Speaker 3: will it happen eventually? Why will it happen eventually?

867
00:40:24,800 --> 00:40:29,040
Speaker 5: Well, if you buy a company that continues to perform

868
00:40:29,200 --> 00:40:31,719
Speaker 5: poorly and you paid a ton for it, I think

869
00:40:31,760 --> 00:40:33,719
Speaker 5: there's only so long that can go on. I think

870
00:40:33,719 --> 00:40:34,960
Speaker 5: it gets more and more obvious.

871
00:40:35,000 --> 00:40:35,520
Speaker 3: For one thing.

872
00:40:35,960 --> 00:40:38,160
Speaker 1: I think a lesson of the meme stack and frankly,

873
00:40:38,200 --> 00:40:42,320
Speaker 1: if crypto is that fundamental sety floor and valuation. But

874
00:40:43,040 --> 00:40:45,160
Speaker 1: you know, you know he's like doing LBI, right. But

875
00:40:45,440 --> 00:40:46,960
Speaker 1: I don't want to just be mean to bitcoin for

876
00:40:47,040 --> 00:40:50,160
Speaker 1: no reason. But like, you can certainly have a model

877
00:40:50,160 --> 00:40:52,280
Speaker 1: in which you say the fundamentals of bitcoin are nothing,

878
00:40:52,520 --> 00:40:55,319
Speaker 1: and you could then say and in fifty years it'll

879
00:40:55,320 --> 00:40:57,480
Speaker 1: go to zero, but like that's a weird thing to say. Now,

880
00:40:57,760 --> 00:40:58,120
Speaker 1: I don't know.

881
00:40:58,200 --> 00:41:02,239
Speaker 5: Well, let me be clear, one short bitcoin with my

882
00:41:02,280 --> 00:41:05,440
Speaker 5: worst enemies portfolio, and I one hundred percent sure I'll

883
00:41:05,480 --> 00:41:07,400
Speaker 5: never say money is a little bit of magic. What

884
00:41:07,480 --> 00:41:09,480
Speaker 5: becomes money is a little bit of magic. But I

885
00:41:09,520 --> 00:41:12,000
Speaker 5: think most of the probability is eventually zero, but it

886
00:41:12,000 --> 00:41:14,160
Speaker 5: may be a very long time arizon. I think what

887
00:41:14,200 --> 00:41:17,800
Speaker 5: we've learned watching this stuff is that time horizon is lengthened.

888
00:41:17,880 --> 00:41:18,160
Speaker 1: Okay.

889
00:41:19,840 --> 00:41:23,600
Speaker 5: In my piece on this, I try to hypothesize, and

890
00:41:23,760 --> 00:41:28,359
Speaker 5: I admit it's real opinionated hypothesizing. You cannot prove if

891
00:41:28,400 --> 00:41:32,600
Speaker 5: I'm right that markets are somewhat less efficient. Why gets

892
00:41:32,680 --> 00:41:35,719
Speaker 5: even harder. But one of my favorite explanations is an

893
00:41:35,760 --> 00:41:40,440
Speaker 5: old man complaining about social media and twenty four to

894
00:41:40,440 --> 00:41:43,719
Speaker 5: seven gamified trading. I don't think most people need a

895
00:41:43,719 --> 00:41:45,719
Speaker 5: lot of convincing that this stuff has made, say, our

896
00:41:45,760 --> 00:41:48,839
Speaker 5: politics worse, made us more in bubbles hate each other

897
00:41:48,880 --> 00:41:50,319
Speaker 5: more when it was supposed to make us love each

898
00:41:50,320 --> 00:41:53,200
Speaker 5: other more. Marcus are just voting mechanisms. I don't think

899
00:41:53,239 --> 00:41:57,040
Speaker 5: it's any different than politics. So this notion that things

900
00:41:57,120 --> 00:41:59,440
Speaker 5: get crazier and can go on for longer, and I

901
00:41:59,480 --> 00:42:02,840
Speaker 5: have a very cynical view of your statement that this

902
00:42:02,840 --> 00:42:08,080
Speaker 5: stuff might take fifty years. I think the more absolutely

903
00:42:08,160 --> 00:42:12,520
Speaker 5: unsubstantiated by anything something is, the longer the craziness can

904
00:42:12,520 --> 00:42:12,759
Speaker 5: go on.

905
00:42:12,960 --> 00:42:15,120
Speaker 1: Right. I think that if you compare the longevity of

906
00:42:15,160 --> 00:42:17,239
Speaker 1: bitkind to like the game Stop premium, I think right.

907
00:42:17,440 --> 00:42:19,080
Speaker 3: But go back to the tech bubble.

908
00:42:19,160 --> 00:42:21,960
Speaker 5: Cisco Systems great company, but was selling at a very

909
00:42:21,960 --> 00:42:25,160
Speaker 5: stupid price, selling it about one hundred PE when the

910
00:42:25,200 --> 00:42:27,760
Speaker 5: e was gigantic. You can have a small startup company

911
00:42:27,840 --> 00:42:30,439
Speaker 5: it's growing super rapidly, sell at one hundred pee.

912
00:42:31,120 --> 00:42:32,480
Speaker 3: One of the largest companies in the.

913
00:42:32,440 --> 00:42:34,279
Speaker 5: World with some of the largest earnings in the world

914
00:42:34,360 --> 00:42:37,000
Speaker 5: selling at one hundred pe. You needed some very heroic

915
00:42:37,600 --> 00:42:40,080
Speaker 5: and I would argue near. I'll never say impossible, but

916
00:42:40,120 --> 00:42:45,080
Speaker 5: near impossible assumptions even if the tech bubble didn't break

917
00:42:45,200 --> 00:42:47,440
Speaker 5: in March of two thousand when it did, and by

918
00:42:47,440 --> 00:42:48,799
Speaker 5: the way, I still don't know why it broke in

919
00:42:48,840 --> 00:42:51,080
Speaker 5: March of two thousand and not a year earlier or

920
00:42:51,080 --> 00:42:53,680
Speaker 5: a year later. Once you're well passed what I would

921
00:42:53,680 --> 00:42:58,240
Speaker 5: consider rational saying you know exactly where but the time horizon,

922
00:42:58,280 --> 00:43:00,560
Speaker 5: you can imagine that going on for when the growth

923
00:43:00,680 --> 00:43:03,600
Speaker 5: is good, maybe even great, but not nearly enough to

924
00:43:03,719 --> 00:43:07,439
Speaker 5: justify that price. It gets more and more obvious if

925
00:43:07,440 --> 00:43:11,719
Speaker 5: something is based completely on air. One of the weird

926
00:43:11,760 --> 00:43:14,080
Speaker 5: of down Matt's point that there's there's no arm to

927
00:43:14,120 --> 00:43:17,840
Speaker 5: the upside, there's no LBO mechanism, and shorting it is

928
00:43:17,840 --> 00:43:19,000
Speaker 5: just frankly too dangerous.

929
00:43:19,560 --> 00:43:20,680
Speaker 3: It's a weird way to say.

930
00:43:20,680 --> 00:43:22,960
Speaker 5: Sometimes people say markets are efficient because you can't make

931
00:43:23,000 --> 00:43:26,000
Speaker 5: money from these things, and I'm like, it's another weird

932
00:43:26,000 --> 00:43:28,560
Speaker 5: way to defend efficient markets to go they can be

933
00:43:28,680 --> 00:43:33,399
Speaker 5: so friggin stupid that they're terrifying to make efficient. That

934
00:43:33,560 --> 00:43:36,520
Speaker 5: may be totally true, but it also is a weird

935
00:43:36,560 --> 00:43:38,120
Speaker 5: way to argue that markets are efficient.

936
00:43:38,280 --> 00:43:41,160
Speaker 4: What does it mean for AQR if markets are less efficient.

937
00:43:41,200 --> 00:43:45,560
Speaker 5: Well, any active management is an inherently arrogant act. You

938
00:43:45,600 --> 00:43:48,759
Speaker 5: cannot tell the average person we should all be active managers,

939
00:43:49,239 --> 00:43:50,560
Speaker 5: we should all have podcasts.

940
00:43:51,120 --> 00:43:58,920
Speaker 7: But I agree, So it's it's top to believe you

941
00:43:58,920 --> 00:44:01,040
Speaker 7: should be an active manager and to believe you're doing

942
00:44:01,080 --> 00:44:03,240
Speaker 7: something good for your clients, you have to believe two things.

943
00:44:03,800 --> 00:44:06,560
Speaker 5: That you have alpha, and that you're not charging the

944
00:44:06,560 --> 00:44:09,600
Speaker 5: full extent of that alpha through your fees. And we

945
00:44:09,640 --> 00:44:12,320
Speaker 5: do believe that, and a lot of active management managers

946
00:44:12,320 --> 00:44:16,080
Speaker 5: believe it. But it's an inherently arrogant act. It's consistent

947
00:44:16,120 --> 00:44:19,000
Speaker 5: to say most people shouldn't do this, but we should,

948
00:44:19,440 --> 00:44:20,920
Speaker 5: though the arrogance is obvious.

949
00:44:21,200 --> 00:44:23,600
Speaker 1: What percent of your alpha should in your face?

950
00:44:24,239 --> 00:44:27,200
Speaker 5: That is a super hard question. I've thought about writing

951
00:44:27,239 --> 00:44:30,359
Speaker 5: about this at one point. It kind of depends on

952
00:44:30,400 --> 00:44:31,759
Speaker 5: how unique your alpha.

953
00:44:31,520 --> 00:44:34,800
Speaker 1: Is, Okay, because I would think that if you got

954
00:44:34,920 --> 00:44:40,799
Speaker 1: a podshot manager really drunk, they would say, no.

955
00:44:40,920 --> 00:44:43,040
Speaker 5: That's what they can charge for their fees, not what

956
00:44:43,080 --> 00:44:45,400
Speaker 5: they should charge. I don't think they'd ever admit them

957
00:44:45,440 --> 00:44:49,640
Speaker 5: at ALGI, like Wilvers shall.

958
00:44:49,640 --> 00:44:52,160
Speaker 1: Deef in their soul. They want one hundred and ten percent.

959
00:44:52,239 --> 00:44:54,919
Speaker 5: It's a great question, but it's a hard question if

960
00:44:54,960 --> 00:44:57,680
Speaker 5: your alpha is doing FOM and French price to book.

961
00:44:58,120 --> 00:44:59,600
Speaker 5: By the way, I still think Farm and French are

962
00:44:59,600 --> 00:45:00,960
Speaker 5: going to be right in the next thirty years. They

963
00:45:00,960 --> 00:45:03,280
Speaker 5: haven't been right in a while, but spreads have gotten

964
00:45:03,320 --> 00:45:06,280
Speaker 5: wider and wider, and that's been a wind in their face.

965
00:45:06,320 --> 00:45:08,319
Speaker 5: I wrote a piece on this saying the long run

966
00:45:08,400 --> 00:45:11,400
Speaker 5: is lying to you, saying that x the spread widening

967
00:45:11,480 --> 00:45:15,280
Speaker 5: values to the simple Farm of French value has delivered alpha.

968
00:45:15,600 --> 00:45:17,279
Speaker 3: It's just lost on the repricing.

969
00:45:17,840 --> 00:45:20,040
Speaker 5: But what you can actually charge for doing price to

970
00:45:20,080 --> 00:45:23,040
Speaker 5: book should be very low, a very small fraction of

971
00:45:23,040 --> 00:45:23,759
Speaker 5: the expected return.

972
00:45:25,000 --> 00:45:26,719
Speaker 1: That's not alpha data, right, But.

973
00:45:26,719 --> 00:45:28,640
Speaker 5: That's kind of what we're saying. Everything exists on a

974
00:45:28,680 --> 00:45:31,799
Speaker 5: spectrum from one to the other. If you have discovered

975
00:45:32,320 --> 00:45:35,560
Speaker 5: and built the database yourself, an alternative data source that

976
00:45:35,640 --> 00:45:38,880
Speaker 5: you have built. This is extreme and I can't think

977
00:45:38,920 --> 00:45:41,520
Speaker 5: of an example at AQR that fits this. But you

978
00:45:41,560 --> 00:45:44,320
Speaker 5: can charge nearly one hundred percent of that alpha because

979
00:45:44,360 --> 00:45:47,719
Speaker 5: what they're getting is still absolutely unrelated to everything else

980
00:45:47,760 --> 00:45:50,279
Speaker 5: what they're doing. There is some deminimous notion that if

981
00:45:50,320 --> 00:45:53,360
Speaker 5: you charge ninety nine percent, maybe no one would bother

982
00:45:53,880 --> 00:45:55,920
Speaker 5: to do it. But you can charge a large fraction

983
00:45:56,000 --> 00:45:58,480
Speaker 5: of the alpha because what you're delivering is still ultimately

984
00:45:58,800 --> 00:46:02,160
Speaker 5: net returns that you can not get elsewhere that aren't

985
00:46:02,160 --> 00:46:05,440
Speaker 5: correlated to the rest of the portfolio are worth it.

986
00:46:06,040 --> 00:46:08,440
Speaker 1: We can talk abstractly about what percentage of your alpha

987
00:46:08,440 --> 00:46:10,000
Speaker 1: you should charge, but like, no one could send a

988
00:46:10,040 --> 00:46:11,640
Speaker 1: bill for like ninety percent on the alpha, right, Like

989
00:46:12,320 --> 00:46:17,320
Speaker 1: is pricing sort of like set by just like anchored norms.

990
00:46:17,440 --> 00:46:19,879
Speaker 5: Anchored norms is another way of saying, what are other

991
00:46:19,960 --> 00:46:20,879
Speaker 5: people charge or.

992
00:46:20,800 --> 00:46:22,200
Speaker 3: In the ballpark of what you're doing?

993
00:46:22,280 --> 00:46:25,080
Speaker 5: So of course that matters, right if you are way

994
00:46:25,120 --> 00:46:27,640
Speaker 5: off the anchor, way off on the high side, no

995
00:46:27,640 --> 00:46:28,719
Speaker 5: one should invest with you.

996
00:46:29,040 --> 00:46:30,399
Speaker 3: If you're way off on the low side.

997
00:46:30,480 --> 00:46:34,040
Speaker 1: People who charge really high fees and are pretty good

998
00:46:34,040 --> 00:46:36,480
Speaker 1: at demonstrating they have alpha, and oh.

999
00:46:36,400 --> 00:46:39,360
Speaker 5: Some of the block shops charge insane fees and have

1000
00:46:39,440 --> 00:46:42,160
Speaker 5: been very good, And that goes to that are they

1001
00:46:42,160 --> 00:46:43,080
Speaker 5: doing something unique?

1002
00:46:43,120 --> 00:46:44,680
Speaker 3: And I think to some extent they are.

1003
00:46:45,160 --> 00:46:47,680
Speaker 5: I think their problem becomes kind of in the direction

1004
00:46:47,800 --> 00:46:49,719
Speaker 5: of Medallion without going all the way. I don't think

1005
00:46:49,719 --> 00:46:53,680
Speaker 5: they've rebuilt medallion. Nothing is that, but they should charge

1006
00:46:53,680 --> 00:46:56,200
Speaker 5: a higher percent if they're doing something very unique, and

1007
00:46:56,239 --> 00:46:59,239
Speaker 5: they do. I have a very you can you know

1008
00:46:59,400 --> 00:47:02,719
Speaker 5: violin playing I think rosy view of how we think

1009
00:47:02,760 --> 00:47:06,319
Speaker 5: about fees. We're building a long term business. We think

1010
00:47:06,320 --> 00:47:08,279
Speaker 5: we have business value. We do not think we're just

1011
00:47:08,360 --> 00:47:11,359
Speaker 5: a hedge fund. We run a lot of traditional assets too,

1012
00:47:11,480 --> 00:47:12,520
Speaker 5: even the hedge funds.

1013
00:47:12,880 --> 00:47:14,319
Speaker 3: We were a big.

1014
00:47:14,080 --> 00:47:18,280
Speaker 5: Pioneer in doing the more obvious strategies and considerably lower fees,

1015
00:47:18,800 --> 00:47:21,520
Speaker 5: starting in merger, ARB and TREND. Following the way we

1016
00:47:21,600 --> 00:47:24,239
Speaker 5: broke into some of those as standalone products, not as

1017
00:47:24,360 --> 00:47:29,160
Speaker 5: part of our multistrats was charging less and saying, you know,

1018
00:47:29,239 --> 00:47:31,960
Speaker 5: this is real and it's good. But you know you

1019
00:47:31,960 --> 00:47:34,439
Speaker 5: can do one of every merger and make a fair

1020
00:47:34,440 --> 00:47:36,840
Speaker 5: amount of money. But that's not magic, and we shouldn't

1021
00:47:36,880 --> 00:47:40,400
Speaker 5: charge magic fees for it. But in that kind of kumbay,

1022
00:47:40,480 --> 00:47:45,120
Speaker 5: a big picture sense, charging fees such that clients are

1023
00:47:45,160 --> 00:47:48,160
Speaker 5: happy with the long term results is probably how you

1024
00:47:48,160 --> 00:47:50,560
Speaker 5: build business value if you step.

1025
00:47:50,320 --> 00:47:52,719
Speaker 4: Back from just AQR though, I'm curious to hear your

1026
00:47:52,719 --> 00:47:56,799
Speaker 4: thoughts on your industry overall and whether alternative strategies in

1027
00:47:56,840 --> 00:47:58,480
Speaker 4: general are too expensive.

1028
00:47:59,040 --> 00:47:59,719
Speaker 3: Yes, they are.

1029
00:48:00,640 --> 00:48:03,120
Speaker 4: I have some numbers to back up that statement. There's

1030
00:48:03,160 --> 00:48:05,400
Speaker 4: a new study out there. My understanding of it is

1031
00:48:05,400 --> 00:48:08,319
Speaker 4: that you basically take a sixty to forty since two

1032
00:48:08,400 --> 00:48:11,000
Speaker 4: thousand and eight, you add Alt's exposure to it in

1033
00:48:11,080 --> 00:48:16,440
Speaker 4: various proportions, and that blended portfolio basically trails that benchmark,

1034
00:48:16,760 --> 00:48:19,560
Speaker 4: basically in close proportions to the fees that are charged

1035
00:48:20,040 --> 00:48:21,799
Speaker 4: by some of the ALTS managers.

1036
00:48:21,960 --> 00:48:23,000
Speaker 2: And I don't know.

1037
00:48:23,080 --> 00:48:25,040
Speaker 4: I read that and I was just kind of wondering.

1038
00:48:25,080 --> 00:48:26,960
Speaker 4: I mean, you could make the case that why are

1039
00:48:27,040 --> 00:48:27,839
Speaker 4: we charging so much?

1040
00:48:27,840 --> 00:48:32,839
Speaker 5: Well, this is almost a mathematical certainty. Yeah, forget about

1041
00:48:32,880 --> 00:48:36,160
Speaker 5: ALTS for a second. You cannot tell me the average

1042
00:48:36,719 --> 00:48:42,000
Speaker 5: active portfolio beats the market after fees and costs.

1043
00:48:42,640 --> 00:48:44,520
Speaker 3: The average active adds up to the.

1044
00:48:44,480 --> 00:48:47,759
Speaker 5: Market because for every deviation one way, there's deviation the

1045
00:48:47,760 --> 00:48:50,320
Speaker 5: other way. When I said it to inherently an arrogant

1046
00:48:50,320 --> 00:48:52,719
Speaker 5: act to be an active manager, it means you think

1047
00:48:52,760 --> 00:48:55,520
Speaker 5: you got it, even though if you buy one of

1048
00:48:55,560 --> 00:48:59,440
Speaker 5: each you can't have it. I mean a subset of

1049
00:48:59,480 --> 00:49:02,319
Speaker 5: the market like could. But I think a lot of

1050
00:49:02,320 --> 00:49:05,799
Speaker 5: that still applies. It's inherently arrogant act. But we've been

1051
00:49:05,840 --> 00:49:08,960
Speaker 5: saying this for at least twenty four years. We wrote

1052
00:49:08,960 --> 00:49:10,760
Speaker 5: a paper, and yes, I started a lot of sentences

1053
00:49:10,760 --> 00:49:13,080
Speaker 5: with We wrote a paper in two thousand and one

1054
00:49:13,239 --> 00:49:16,200
Speaker 5: called do hedge funds Hedge where we took the known

1055
00:49:16,239 --> 00:49:19,680
Speaker 5: indices of hedge funds and we tried to take out

1056
00:49:19,760 --> 00:49:21,640
Speaker 5: just the market beta. You could argue for a more

1057
00:49:21,680 --> 00:49:24,960
Speaker 5: sophisticated risk model, and that could as usual and go

1058
00:49:25,000 --> 00:49:27,799
Speaker 5: down that rabbit hole. But we found that betas were.

1059
00:49:27,840 --> 00:49:31,080
Speaker 5: First of all, this is more mundane but statistically underestimated,

1060
00:49:31,600 --> 00:49:34,520
Speaker 5: partly because a lot of hedge funds do some stuff

1061
00:49:34,520 --> 00:49:35,960
Speaker 5: that is not of perfect liquidity.

1062
00:49:36,880 --> 00:49:38,840
Speaker 3: And this was a small.

1063
00:49:39,800 --> 00:49:41,920
Speaker 5: Early version of what I got into at the private

1064
00:49:41,960 --> 00:49:44,319
Speaker 5: world later on. But if something doesn't trade all the

1065
00:49:44,360 --> 00:49:47,240
Speaker 5: time and you try to estimate its correlation with the market,

1066
00:49:47,760 --> 00:49:50,879
Speaker 5: you will underestimate it because one great way to look

1067
00:49:50,960 --> 00:49:53,480
Speaker 5: uncorrelated is to have a three day leg, and when

1068
00:49:53,480 --> 00:49:56,200
Speaker 5: you trade, your returns will be off. So the betas

1069
00:49:56,200 --> 00:49:59,600
Speaker 5: were underestimated by earlier studies. Given the correct betas, there

1070
00:49:59,600 --> 00:50:02,320
Speaker 5: was pretty much no alpha to the hedge fund world.

1071
00:50:02,719 --> 00:50:04,080
Speaker 5: First of all, I was a lot younger than a

1072
00:50:04,120 --> 00:50:06,919
Speaker 5: lot less well known, so I cared Nowadays, I quite

1073
00:50:06,920 --> 00:50:10,759
Speaker 5: obviously court controversy. But back then I probably had ten

1074
00:50:10,760 --> 00:50:13,880
Speaker 5: famous managers call me and yell at me about that paper.

1075
00:50:14,200 --> 00:50:16,360
Speaker 3: Yeah. The first time, I was, of course obnoxious.

1076
00:50:16,400 --> 00:50:18,839
Speaker 5: They called and said, why did you write this? And

1077
00:50:19,360 --> 00:50:22,040
Speaker 5: I gave the obnoxious answer because we think it's true.

1078
00:50:22,640 --> 00:50:25,640
Speaker 5: Now it was apparently not an acceptable answer. I will

1079
00:50:25,640 --> 00:50:28,680
Speaker 5: only call out one person on the positive side, Richard Perry,

1080
00:50:28,719 --> 00:50:29,960
Speaker 5: famous hedge fund manager.

1081
00:50:30,320 --> 00:50:30,920
Speaker 3: He called me up.

1082
00:50:30,960 --> 00:50:32,520
Speaker 5: And I already been yelled at by a whole bunch

1083
00:50:32,560 --> 00:50:34,960
Speaker 5: of people. And so when I heard Richard Perry on

1084
00:50:34,960 --> 00:50:36,880
Speaker 5: the phone, people like Richard Perry didn't call people like

1085
00:50:36,920 --> 00:50:37,920
Speaker 5: me back then.

1086
00:50:38,040 --> 00:50:39,840
Speaker 3: He was big, I was small. So I'm like, I

1087
00:50:39,880 --> 00:50:40,440
Speaker 3: know what this is.

1088
00:50:40,480 --> 00:50:42,640
Speaker 5: He's just gonna yell at me. He gets on the phone,

1089
00:50:42,640 --> 00:50:44,240
Speaker 5: he goes, that paper you wrote.

1090
00:50:43,960 --> 00:50:44,720
Speaker 3: That's just correct.

1091
00:50:44,719 --> 00:50:47,400
Speaker 5: Good job, right, And I'm only telling that. I'm not

1092
00:50:47,400 --> 00:50:49,120
Speaker 5: giving the names. And I do remember him, of.

1093
00:50:49,040 --> 00:50:53,680
Speaker 3: Course, in a book somewhere. It's all I have a

1094
00:50:53,719 --> 00:50:56,000
Speaker 3: list of Vedettas and my BRAINSU.

1095
00:50:56,040 --> 00:50:59,000
Speaker 1: Wait, so you've been You've been making fun of private

1096
00:50:59,000 --> 00:51:01,959
Speaker 1: equity managers on some lines recently. Do you get calls

1097
00:51:01,960 --> 00:51:02,600
Speaker 1: from them?

1098
00:51:02,880 --> 00:51:05,960
Speaker 5: No, they're so fat and happy, they don't even care

1099
00:51:06,040 --> 00:51:08,120
Speaker 5: about me making fun. No, I get yelled at by

1100
00:51:08,120 --> 00:51:11,759
Speaker 5: some usually friends. I live in Greenwich connection, so you

1101
00:51:11,840 --> 00:51:12,560
Speaker 5: all hang out.

1102
00:51:12,640 --> 00:51:14,719
Speaker 1: The country club. They're like, I'm.

1103
00:51:14,520 --> 00:51:19,239
Speaker 5: Actually a country club, but the old often used in

1104
00:51:19,280 --> 00:51:21,319
Speaker 5: a very bad way. Some of my best friends are

1105
00:51:21,360 --> 00:51:23,759
Speaker 5: so I'm off the hook. I have some good friends

1106
00:51:23,760 --> 00:51:27,240
Speaker 5: who are private equity managers. Most of them can accept

1107
00:51:27,320 --> 00:51:30,239
Speaker 5: the you're right about the industry but not our firm,

1108
00:51:30,440 --> 00:51:30,799
Speaker 5: which is.

1109
00:51:30,880 --> 00:51:34,399
Speaker 3: Essentially what I'm saying. So I can't not being mean

1110
00:51:34,600 --> 00:51:35,080
Speaker 3: about this.

1111
00:51:35,280 --> 00:51:38,480
Speaker 1: Right, You're not like, your criticism is volatility lunder right.

1112
00:51:38,520 --> 00:51:41,440
Speaker 1: Your criticism is that private equity seems to have a

1113
00:51:41,480 --> 00:51:43,919
Speaker 1: higher sharp because it has a lower of volatility because

1114
00:51:43,960 --> 00:51:45,040
Speaker 1: it doesn't report.

1115
00:51:46,480 --> 00:51:49,440
Speaker 5: Yes, that is my main criticism, which I think is

1116
00:51:49,520 --> 00:51:50,600
Speaker 5: quite obviously true.

1117
00:51:50,800 --> 00:51:52,440
Speaker 3: That's just me. Some people do disagree.

1118
00:51:53,320 --> 00:51:55,880
Speaker 5: The best disagreement I've heard, and I have some sympathy

1119
00:51:55,880 --> 00:51:58,120
Speaker 5: for this because I do not think markets are perfect,

1120
00:51:58,520 --> 00:52:02,839
Speaker 5: is we are right about the valuations. You are right

1121
00:52:03,239 --> 00:52:05,800
Speaker 5: that we move at a highly damped version of the market,

1122
00:52:06,400 --> 00:52:07,839
Speaker 5: but market moves too much.

1123
00:52:08,000 --> 00:52:08,759
Speaker 3: We are right.

1124
00:52:09,280 --> 00:52:11,360
Speaker 5: My response to that is, why don't we get to

1125
00:52:11,360 --> 00:52:11,600
Speaker 5: do that?

1126
00:52:11,920 --> 00:52:12,319
Speaker 1: It's fair?

1127
00:52:12,360 --> 00:52:14,560
Speaker 5: You know when we've had a tough year because the

1128
00:52:14,640 --> 00:52:16,640
Speaker 5: market's gone crazy and we were on the wrong side

1129
00:52:16,640 --> 00:52:19,160
Speaker 5: of that. I have to tell my clients we're down

1130
00:52:19,160 --> 00:52:22,040
Speaker 5: twelve percent. I don't get to tell my clients we're

1131
00:52:22,160 --> 00:52:25,080
Speaker 5: up based on where I'd mark the portfolio. So why

1132
00:52:25,160 --> 00:52:27,080
Speaker 5: one group? And by the way, they could market just

1133
00:52:27,120 --> 00:52:29,759
Speaker 5: like we market. They're brilliant at valuing companies and they

1134
00:52:29,760 --> 00:52:31,400
Speaker 5: can tell you where they could sell it for today.

1135
00:52:31,800 --> 00:52:35,040
Speaker 5: So it's like an institutional legal quirk that they get

1136
00:52:35,040 --> 00:52:36,400
Speaker 5: to do it one way and we have to do

1137
00:52:36,440 --> 00:52:39,759
Speaker 5: it another. Everyone can mark their portfolio at what they

1138
00:52:39,760 --> 00:52:43,320
Speaker 5: could sell it for today or what they think it's worth,

1139
00:52:43,680 --> 00:52:45,319
Speaker 5: and one side gets to do it one one gets

1140
00:52:45,360 --> 00:52:48,560
Speaker 5: to do the other. So that's a reasonable argument, but

1141
00:52:48,760 --> 00:52:51,440
Speaker 5: I still think it leads to an unreasonable conclusion. I

1142
00:52:51,520 --> 00:52:54,600
Speaker 5: will say ninety percent of my critique is about the

1143
00:52:54,640 --> 00:52:57,560
Speaker 5: volatility or the beta, is about saying these things are

1144
00:52:57,560 --> 00:52:58,040
Speaker 5: low risk.

1145
00:52:58,480 --> 00:52:59,440
Speaker 3: Ten percent is.

1146
00:52:59,400 --> 00:53:04,239
Speaker 5: About percent if not trailing future returns, because if I'm

1147
00:53:04,320 --> 00:53:08,320
Speaker 5: right about the volatility laundering, it has implications for returns

1148
00:53:08,320 --> 00:53:11,480
Speaker 5: going forward. If when David Swenson was pioneering, which is

1149
00:53:11,520 --> 00:53:14,000
Speaker 5: what he called his book, Private equity is part of

1150
00:53:14,040 --> 00:53:16,000
Speaker 5: an institutional endownment portfolio.

1151
00:53:17,080 --> 00:53:18,880
Speaker 3: He's quite clear. It's a lot of it. It's an

1152
00:53:18,880 --> 00:53:20,919
Speaker 3: ill liquidity premium that no one wants.

1153
00:53:21,000 --> 00:53:23,680
Speaker 5: Illiquidity. Everyone's scared of it. So if you're willing to

1154
00:53:23,719 --> 00:53:26,440
Speaker 5: do that, you get paid extra. If I'm right that

1155
00:53:26,560 --> 00:53:29,960
Speaker 5: people love the fact that they don't have to look

1156
00:53:29,960 --> 00:53:34,560
Speaker 5: at the volatility, that means illiquidity is no longer a bug.

1157
00:53:34,600 --> 00:53:37,480
Speaker 5: It is now a feature and very simple model for

1158
00:53:37,520 --> 00:53:40,959
Speaker 5: how expective returns are set. You get paid a higher

1159
00:53:40,960 --> 00:53:42,840
Speaker 5: expective return in something if you have to bear a

1160
00:53:42,840 --> 00:53:46,960
Speaker 5: bug nobody wants and you pay through a lower expector return.

1161
00:53:47,320 --> 00:53:49,680
Speaker 5: If you have a feature everyone wants, then you have

1162
00:53:49,719 --> 00:53:52,360
Speaker 5: to pay up for it. So the chance that that

1163
00:53:52,480 --> 00:53:54,399
Speaker 5: is going on going forward, I think is quite high.

1164
00:53:54,400 --> 00:53:56,680
Speaker 5: Whether it means there's no edge to private equity or

1165
00:53:56,680 --> 00:54:00,359
Speaker 5: a negative edge or a smaller positive edge can tell

1166
00:54:00,360 --> 00:54:02,080
Speaker 5: you that. But I do think if you think of

1167
00:54:02,160 --> 00:54:05,719
Speaker 5: risk ajusy return as numerator of return denominator of risk,

1168
00:54:06,080 --> 00:54:09,840
Speaker 5: I think my statements are mostly about the denominator, but

1169
00:54:09,960 --> 00:54:11,800
Speaker 5: they're ten percent now about the numerator.

1170
00:54:11,840 --> 00:54:29,320
Speaker 1: Going forward. You're in my concernami private equity, private markets

1171
00:54:29,400 --> 00:54:32,080
Speaker 1: by which is that it seems to me that like

1172
00:54:32,520 --> 00:54:35,960
Speaker 1: there is a ton of fee pressure in public markets

1173
00:54:36,000 --> 00:54:38,600
Speaker 1: and everyone has kind of like learned the gospel of

1174
00:54:38,640 --> 00:54:42,080
Speaker 1: like Bilow cost index funds and even charging for farma

1175
00:54:42,120 --> 00:54:44,839
Speaker 1: French factors is not a two and twenty business. And

1176
00:54:45,760 --> 00:54:49,920
Speaker 1: private markets because they're not indexible because it's harder to

1177
00:54:50,000 --> 00:54:54,760
Speaker 1: like extract factors because they're not liquid, don't have those problems,

1178
00:54:54,800 --> 00:54:56,120
Speaker 1: and you can charge two and twenty for a lot

1179
00:54:56,120 --> 00:54:58,000
Speaker 1: of private stuff. And so like it seems to me

1180
00:54:58,400 --> 00:55:00,600
Speaker 1: that there is a move to you put a lot

1181
00:55:00,600 --> 00:55:04,160
Speaker 1: of stuff that would have previously been public into private markets,

1182
00:55:04,160 --> 00:55:06,640
Speaker 1: and to say we can put privates into four oh

1183
00:55:06,680 --> 00:55:09,840
Speaker 1: one ks and there's a good economic grastionals we're putting

1184
00:55:09,840 --> 00:55:11,960
Speaker 1: private assets into four one case because you don't need liquidity.

1185
00:55:12,600 --> 00:55:14,799
Speaker 1: But there's also like this like really overwhelming if.

1186
00:55:14,680 --> 00:55:19,600
Speaker 8: There's a positive illiquidity premium. Yeah too right, sorry, go no.

1187
00:55:19,680 --> 00:55:22,080
Speaker 8: I just like you know, you've written for years ago

1188
00:55:22,239 --> 00:55:25,759
Speaker 8: like criticizing people for charging alpha fees for beta, and

1189
00:55:25,800 --> 00:55:28,279
Speaker 8: it seems to me that like the reprivatization on the

1190
00:55:28,320 --> 00:55:31,720
Speaker 8: market is a way to sneak some alpha fies onto beta.

1191
00:55:31,840 --> 00:55:34,520
Speaker 5: I think a tremendous amount of the private world is

1192
00:55:34,600 --> 00:55:36,480
Speaker 5: charging massive alpha fees for beta.

1193
00:55:36,800 --> 00:55:38,520
Speaker 3: I won't mince words about that.

1194
00:55:38,600 --> 00:55:42,520
Speaker 5: If they outperform or underperform on net after all these

1195
00:55:42,560 --> 00:55:45,480
Speaker 5: massive fees, and if performance going forward, it is tougher.

1196
00:55:45,800 --> 00:55:47,720
Speaker 5: And by the way, your point about the fee compression

1197
00:55:47,800 --> 00:55:50,719
Speaker 5: in the public world only makes my hypothesis that they

1198
00:55:50,760 --> 00:55:53,239
Speaker 5: won't beat the public world by the same amount going

1199
00:55:53,239 --> 00:55:56,200
Speaker 5: forward stronger. I think privates have a big function in

1200
00:55:56,200 --> 00:55:58,040
Speaker 5: the world. I don't think they're going away what you

1201
00:55:58,080 --> 00:55:59,600
Speaker 5: started out with me, what do you guys do for

1202
00:55:59,640 --> 00:56:01,400
Speaker 5: the world. There are things that are in between what

1203
00:56:01,480 --> 00:56:04,000
Speaker 5: should be public and what's mom and pop. But I

1204
00:56:04,000 --> 00:56:06,600
Speaker 5: think where we are now, I think a lot of

1205
00:56:06,640 --> 00:56:12,799
Speaker 5: institutions are giving up some amount of expected return for

1206
00:56:13,000 --> 00:56:16,400
Speaker 5: the ease of limit, of reducing their agency problem of

1207
00:56:16,440 --> 00:56:19,040
Speaker 5: sticking with something. Now, if they're going to be terrible

1208
00:56:19,120 --> 00:56:21,719
Speaker 5: and not stick with things, it might be rational to

1209
00:56:21,719 --> 00:56:25,080
Speaker 5: give up some return. But you can't double count and

1210
00:56:25,120 --> 00:56:28,240
Speaker 5: say we're making ourselves better investors giving up some expected

1211
00:56:28,280 --> 00:56:30,359
Speaker 5: return and oh yeah, our expected returns are going.

1212
00:56:30,280 --> 00:56:30,760
Speaker 1: To be higher.

1213
00:56:31,080 --> 00:56:34,680
Speaker 5: If that's the rationale, then you've accepted my argument, and

1214
00:56:34,719 --> 00:56:36,560
Speaker 5: I think you have to say, this is what we

1215
00:56:36,600 --> 00:56:38,480
Speaker 5: get paid for by making your life easier.

1216
00:56:39,120 --> 00:56:41,239
Speaker 4: I am curious what you make of the push to

1217
00:56:41,280 --> 00:56:44,719
Speaker 4: put privates into more retail accessible wrappers. I'm talking about

1218
00:56:44,719 --> 00:56:47,840
Speaker 4: et avs, but I guess I'm also talking about interval

1219
00:56:47,840 --> 00:56:49,000
Speaker 4: funds a little bit.

1220
00:56:49,239 --> 00:56:51,359
Speaker 2: It just seems like that's where the world is headed.

1221
00:56:51,440 --> 00:56:53,680
Speaker 5: I'm going to hedge this and say it very carefully.

1222
00:56:53,840 --> 00:56:54,879
Speaker 5: I think it's a terrible idea.

1223
00:56:55,120 --> 00:56:55,719
Speaker 6: Okay, go on.

1224
00:56:56,000 --> 00:56:58,600
Speaker 5: There are things that end up in retail in a

1225
00:56:58,680 --> 00:57:00,920
Speaker 5: very good way eventually, but we've done some of this

1226
00:57:00,920 --> 00:57:03,399
Speaker 5: when we introduce mutual funds. I can't say always going

1227
00:57:03,440 --> 00:57:05,719
Speaker 5: to retail is a bad thing. You can price it reasonably,

1228
00:57:05,760 --> 00:57:08,360
Speaker 5: you can say these are strategies you've never had before.

1229
00:57:09,000 --> 00:57:13,120
Speaker 5: This feels a little bit more like we've exhausted the institutions.

1230
00:57:13,280 --> 00:57:15,440
Speaker 5: I think I saw a number saying endown. It's like

1231
00:57:15,480 --> 00:57:18,560
Speaker 5: forty three percent for a number I can't verify. That's

1232
00:57:18,640 --> 00:57:21,840
Speaker 5: wildly specific, but that's the number I remember. So it

1233
00:57:21,880 --> 00:57:23,880
Speaker 5: has a feel of who else we're going to get

1234
00:57:23,880 --> 00:57:25,240
Speaker 5: to own this stuff.

1235
00:57:25,240 --> 00:57:27,440
Speaker 1: It seems so explicitly, and it's really wild.

1236
00:57:27,880 --> 00:57:31,560
Speaker 5: I think it's explicitly that it's one of these things

1237
00:57:31,560 --> 00:57:33,960
Speaker 5: that for the cynics, I don't think they'll ever be

1238
00:57:34,000 --> 00:57:36,280
Speaker 5: a satisfying moment where we get to say we were right.

1239
00:57:36,400 --> 00:57:40,040
Speaker 5: It'll just be somewhat worse over the next decade. It's

1240
00:57:40,040 --> 00:57:42,880
Speaker 5: not one of these things like a tech bubble in

1241
00:57:43,000 --> 00:57:45,920
Speaker 5: ninety nine two thousand that at least, I don't think

1242
00:57:46,120 --> 00:57:50,560
Speaker 5: there are scenarios where things get worse rapidly and secondary sales.

1243
00:57:50,600 --> 00:57:52,960
Speaker 5: We got close to some of that in the GFC.

1244
00:57:53,520 --> 00:57:56,080
Speaker 5: I was on some investment committees where we were talking

1245
00:57:56,080 --> 00:57:57,640
Speaker 5: we didn't do it, but we were talking about whether

1246
00:57:57,680 --> 00:57:59,560
Speaker 5: we would have to do that. So I'm not saying

1247
00:57:59,560 --> 00:58:01,280
Speaker 5: it's a chance of ugliness, but I think the meat

1248
00:58:01,320 --> 00:58:05,640
Speaker 5: of the probability is is just somewhat worse. But it

1249
00:58:05,760 --> 00:58:09,320
Speaker 5: does cause me some worry sadness to say, yeah, what

1250
00:58:09,400 --> 00:58:11,520
Speaker 5: we really need is to stick a liquids in four.

1251
00:58:11,360 --> 00:58:12,000
Speaker 3: To one case.

1252
00:58:12,200 --> 00:58:12,439
Speaker 2: Yeah.

1253
00:58:12,640 --> 00:58:15,320
Speaker 3: So yeah, bluntly, I think it's a bad idea.

1254
00:58:15,520 --> 00:58:17,439
Speaker 4: Well, I just wanted to talk about and I don't

1255
00:58:17,440 --> 00:58:18,920
Speaker 4: know if this goes too far, but it feels like

1256
00:58:18,960 --> 00:58:20,600
Speaker 4: you had a change of heart when it comes to

1257
00:58:20,720 --> 00:58:22,560
Speaker 4: machine learning and AI in general.

1258
00:58:22,840 --> 00:58:25,040
Speaker 2: This is something we've spoken about before.

1259
00:58:25,120 --> 00:58:27,520
Speaker 4: I did well, was that like a light bulb moment

1260
00:58:27,640 --> 00:58:29,720
Speaker 4: or was that just you know, maybe people on your

1261
00:58:29,800 --> 00:58:31,320
Speaker 4: team wearing you down over time?

1262
00:58:31,720 --> 00:58:32,840
Speaker 3: It was more the ladder.

1263
00:58:33,120 --> 00:58:36,040
Speaker 5: Yeah, the ladder is the second one, right, I gotta

1264
00:58:36,040 --> 00:58:38,400
Speaker 5: think that through every time I tell that people this.

1265
00:58:38,480 --> 00:58:40,600
Speaker 5: I've said in a lot of public arenas I sell

1266
00:58:40,640 --> 00:58:42,680
Speaker 5: it to clients. I think I probably slowed us down

1267
00:58:42,720 --> 00:58:45,040
Speaker 5: by a couple of years in machine learning. I think

1268
00:58:45,080 --> 00:58:46,880
Speaker 5: it probably costs us some money because the stuff has

1269
00:58:46,920 --> 00:58:50,400
Speaker 5: worked pretty well. We've always described ourselves and we were

1270
00:58:50,440 --> 00:58:53,240
Speaker 5: getting into this a little bit earlier. As a blend

1271
00:58:53,480 --> 00:58:57,480
Speaker 5: of data, back tests are nice out of sample long

1272
00:58:57,560 --> 00:59:01,000
Speaker 5: periods are even nicer, but also theory.

1273
00:59:01,640 --> 00:59:02,280
Speaker 3: Theory can be a.

1274
00:59:02,320 --> 00:59:04,480
Speaker 5: Formal economic theory, but it can also just mean a

1275
00:59:04,600 --> 00:59:06,960
Speaker 5: common sense story where you think you understand why you're

1276
00:59:07,000 --> 00:59:10,760
Speaker 5: making money. And there's no way to say exactly what

1277
00:59:10,880 --> 00:59:13,400
Speaker 5: percentage of both. But I've often described it as an

1278
00:59:13,400 --> 00:59:15,600
Speaker 5: attempt to be fifty to fifty. We want something to

1279
00:59:15,680 --> 00:59:19,240
Speaker 5: make sense to us as economists and have strong data

1280
00:59:19,720 --> 00:59:21,960
Speaker 5: when you move into the machine learning world. I don't

1281
00:59:22,000 --> 00:59:23,720
Speaker 5: think you have to abandon theory, and this is something

1282
00:59:23,760 --> 00:59:26,080
Speaker 5: I think we do a little different than most I

1283
00:59:26,120 --> 00:59:27,640
Speaker 5: think there are some in the machine learning world to

1284
00:59:27,680 --> 00:59:30,320
Speaker 5: kind of throw theory out the window. We're still using

1285
00:59:30,400 --> 00:59:33,320
Speaker 5: common sense and theory to kind of limit the scope,

1286
00:59:33,480 --> 00:59:36,600
Speaker 5: but you are leaning more on the data. And again

1287
00:59:36,760 --> 00:59:39,360
Speaker 5: I'm making up these numbers. But if regular stuff is

1288
00:59:39,400 --> 00:59:43,000
Speaker 5: fifty to fifty, machine learning seventy five twenty five data

1289
00:59:43,280 --> 00:59:45,400
Speaker 5: even for us, that was uncomfortable for me.

1290
00:59:45,640 --> 00:59:47,000
Speaker 3: When you've been telling a story.

1291
00:59:46,840 --> 00:59:49,400
Speaker 5: For twenty five years and it's worked for you and

1292
00:59:49,480 --> 00:59:53,000
Speaker 5: your clients, it's not easy to move. I also think

1293
00:59:53,560 --> 00:59:56,080
Speaker 5: it's not improper for the role for the old man

1294
00:59:56,160 --> 00:59:57,840
Speaker 5: of the firm to go, let's.

1295
00:59:57,640 --> 00:59:58,240
Speaker 3: Just slow down.

1296
00:59:58,680 --> 01:00:01,120
Speaker 5: You guys come in here doing this with a Southern

1297
01:00:01,240 --> 01:00:05,600
Speaker 5: droll and your new fangled machine learning. No, that's the

1298
01:00:05,720 --> 01:00:06,400
Speaker 5: villainous rule.

1299
01:00:06,480 --> 01:00:07,680
Speaker 2: Oh no, I think it's kind of cool.

1300
01:00:07,720 --> 01:00:11,880
Speaker 5: Yeah, okay, so I actually can defend it as entirely appropriate.

1301
01:00:11,960 --> 01:00:13,960
Speaker 5: But yeah, I had to be convinced. It wasn't a

1302
01:00:14,080 --> 01:00:16,920
Speaker 5: light bulb moment. It was people like Brian Kelly, Andrea Ferzini,

1303
01:00:17,000 --> 01:00:20,440
Speaker 5: Laura Serb and all partners of mine presenting great results

1304
01:00:20,480 --> 01:00:23,080
Speaker 5: that made increase it and I did a lot more reading.

1305
01:00:23,120 --> 01:00:24,200
Speaker 3: I was probably I.

1306
01:00:24,280 --> 01:00:27,600
Speaker 5: Programmed in a programming language called LISP in the nineteen eighties.

1307
01:00:27,640 --> 01:00:31,280
Speaker 5: That was an early machine learning or AI language, not

1308
01:00:31,280 --> 01:00:33,800
Speaker 5: even machine learning, and then I didn't do anything about

1309
01:00:33,880 --> 01:00:36,720
Speaker 5: that for the next thirty years. So I think part

1310
01:00:36,760 --> 01:00:38,520
Speaker 5: of it was just me getting up to speed.

1311
01:00:38,600 --> 01:00:41,160
Speaker 1: Frankly, the thing I find compelling in the Kelly paper

1312
01:00:41,280 --> 01:00:43,920
Speaker 1: is like you seed like a sort of toy set

1313
01:00:43,960 --> 01:00:46,720
Speaker 1: of factors into like a ten thousand neuron model, And

1314
01:00:47,280 --> 01:00:49,479
Speaker 1: what he argues is basically you might get a better

1315
01:00:50,320 --> 01:00:53,720
Speaker 1: view of the actual like function that generates the results

1316
01:00:53,760 --> 01:00:56,200
Speaker 1: than if you just try to like use your economic

1317
01:00:56,280 --> 01:00:58,040
Speaker 1: intuition in a lineary aggression.

1318
01:00:58,160 --> 01:00:58,240
Speaker 3: Right.

1319
01:00:58,400 --> 01:01:00,240
Speaker 1: There's like theory behind it. In the theory is like

1320
01:01:00,320 --> 01:01:04,080
Speaker 1: things are not as linear as you know traditional methods, and.

1321
01:01:04,280 --> 01:01:07,760
Speaker 5: It's basically saying it's still a sin to only look

1322
01:01:07,840 --> 01:01:11,400
Speaker 5: for patterns. You still need some economics, but machine learning

1323
01:01:11,520 --> 01:01:13,840
Speaker 5: is better at balancing those trade offs. So the idea

1324
01:01:13,840 --> 01:01:16,320
Speaker 5: of throwing more at it when you have a better

1325
01:01:16,400 --> 01:01:20,200
Speaker 5: technique for that, you lean in that direction. I did

1326
01:01:20,360 --> 01:01:23,120
Speaker 5: have a better title, I think than him. I wanted

1327
01:01:23,200 --> 01:01:25,960
Speaker 5: him that's not better. His titles great pretty well, no,

1328
01:01:26,160 --> 01:01:29,000
Speaker 5: but I wanted to call it simply. Ackham was wrong.

1329
01:01:31,360 --> 01:01:34,320
Speaker 3: Okay, all right, and maybe not bad, but I still

1330
01:01:34,360 --> 01:01:36,480
Speaker 3: want him to use that for maybe a follow up paper.

1331
01:01:37,360 --> 01:01:38,560
Speaker 1: All right, Cliff, thanks for coming.

1332
01:01:38,560 --> 01:01:39,080
Speaker 3: Oh, this was fun.

1333
01:01:39,160 --> 01:01:40,560
Speaker 2: Thanks for video with us.

1334
01:01:47,040 --> 01:01:48,480
Speaker 1: And that was the Money Stuff Podcast.

1335
01:01:48,680 --> 01:01:50,600
Speaker 2: I'm Matt Levi and I'm Katie Grifeld.

1336
01:01:51,040 --> 01:01:53,080
Speaker 1: You can find my work by subscribing to The Money

1337
01:01:53,120 --> 01:01:54,920
Speaker 1: Stuff newsletter on Bloomberg.

1338
01:01:54,560 --> 01:01:57,600
Speaker 4: Dot com, and you can find me on Bloomberg TV

1339
01:01:57,800 --> 01:02:01,440
Speaker 4: every day on Open Interest between nine to eleven am Eastern.

1340
01:02:01,840 --> 01:02:03,560
Speaker 1: We'd love to hear from you. You can send an

1341
01:02:03,560 --> 01:02:06,760
Speaker 1: email to Moneypod at Bloomberg dot net, ask us a

1342
01:02:06,840 --> 01:02:08,200
Speaker 1: question and we might answer it on air.

1343
01:02:08,520 --> 01:02:10,680
Speaker 4: You can also subscribe to our show wherever you're listening

1344
01:02:10,760 --> 01:02:12,760
Speaker 4: right now and leave us a review. It helps more

1345
01:02:12,800 --> 01:02:13,640
Speaker 4: people find the show.

1346
01:02:14,400 --> 01:02:17,600
Speaker 1: The Money Stuff Podcast is produced by Anamasarakus and Moses

1347
01:02:17,640 --> 01:02:19,919
Speaker 1: ONAM Special. Thanks this week to stay se Won.

1348
01:02:20,240 --> 01:02:22,320
Speaker 4: Our theme music was composed by Blake Maples.

1349
01:02:22,680 --> 01:02:25,280
Speaker 1: Brandon Francis Newdam is our executive producer.

1350
01:02:25,040 --> 01:02:27,160
Speaker 4: And Stage Bauman is Bloomberg's head of Podcasts.

1351
01:02:27,400 --> 01:02:29,720
Speaker 1: Thanks for listening to The Money Stuff Podcasts. We'll be

1352
01:02:29,800 --> 01:02:31,240
Speaker 1: back next week with more stuff