1
00:00:00,080 --> 00:00:06,080
Speaker 1: M. This is Mesters in Business with Very Renaults on

2
00:00:06,240 --> 00:00:11,479
Speaker 1: Bluebird Radio. This week on the podcast, I have an

3
00:00:11,520 --> 00:00:16,160
Speaker 1: extra special guest. What can I say about Sandy rat Trey.

4
00:00:16,239 --> 00:00:20,200
Speaker 1: He is the chief investment officer of the Man Group,

5
00:00:20,239 --> 00:00:25,160
Speaker 1: which manages over a hundred and twenty five billion dollars UH.

6
00:00:25,200 --> 00:00:28,960
Speaker 1: He's the co inventor of the VIX index. He has

7
00:00:29,480 --> 00:00:36,200
Speaker 1: an incredible career UH both in UH equity research and

8
00:00:36,600 --> 00:00:41,839
Speaker 1: for derivatives as well as systematic investing UM. He's just

9
00:00:41,920 --> 00:00:43,400
Speaker 1: a rock star. I don't know what else to say

10
00:00:43,440 --> 00:00:47,000
Speaker 1: about it. The track record that the Man Group has amassed,

11
00:00:47,479 --> 00:00:52,120
Speaker 1: as well as how they've pushed forward UM portfolio construction

12
00:00:52,880 --> 00:00:58,880
Speaker 1: theory is really incredibly, incredibly influential. Not only was he

13
00:00:59,000 --> 00:01:03,160
Speaker 1: the co inventor of the VIX index, but he's written

14
00:01:03,280 --> 00:01:08,280
Speaker 1: extensively about risk management and how to design portfolios and

15
00:01:08,880 --> 00:01:12,000
Speaker 1: you know what to expect when you're expecting a black swan,

16
00:01:12,080 --> 00:01:15,960
Speaker 1: which you by definition can't know what to expect, and

17
00:01:16,280 --> 00:01:18,840
Speaker 1: how to build strategies that give you some degree of

18
00:01:18,880 --> 00:01:23,000
Speaker 1: protection against this. If you're remotely interested in hedge funds,

19
00:01:23,000 --> 00:01:28,280
Speaker 1: asset management, quantitative strategies, the VIX and managing risk well

20
00:01:28,319 --> 00:01:32,319
Speaker 1: strap yourself in because this one, uh is a good one.

21
00:01:32,720 --> 00:01:37,560
Speaker 1: With no further ado, my conversation with Sandy Retrey of

22
00:01:37,600 --> 00:01:43,480
Speaker 1: the Man Group. This is Masters in Business with very

23
00:01:43,520 --> 00:01:49,520
Speaker 1: Renaults on Bluebird Radio. My extra special guest this week

24
00:01:49,680 --> 00:01:53,520
Speaker 1: is Sandy Retrey. He is the chief Investment Officer of

25
00:01:53,560 --> 00:01:56,520
Speaker 1: the Man Group. He is also on the Executive Committee

26
00:01:56,560 --> 00:02:00,600
Speaker 1: and the Responsible Investment Committee. He is per Apps most

27
00:02:00,600 --> 00:02:06,000
Speaker 1: famously the co inventor of the VIX index. He's also

28
00:02:06,360 --> 00:02:10,240
Speaker 1: run a number of different systematic strategies for Man and

29
00:02:10,360 --> 00:02:15,000
Speaker 1: other organizations. The Man Group's assets under management are over

30
00:02:15,080 --> 00:02:18,680
Speaker 1: a hundred and twenty billion dollars and Sandy is the

31
00:02:18,760 --> 00:02:23,120
Speaker 1: co author of the book Strategic Risk Management, Designing Portfolios

32
00:02:23,120 --> 00:02:28,360
Speaker 1: and Managing Risk. Sandy Retree, Welcome to Bloomberg. Great, Thank

33
00:02:28,360 --> 00:02:30,400
Speaker 1: you very much, Perry. It's good to be with you.

34
00:02:30,720 --> 00:02:34,080
Speaker 1: So you have a really interesting background. You're deeply steeped

35
00:02:34,080 --> 00:02:37,120
Speaker 1: in mathematics. How did you find your way into the

36
00:02:37,160 --> 00:02:42,000
Speaker 1: investment business? So it's a it's a long story. I

37
00:02:42,080 --> 00:02:45,400
Speaker 1: as a teenager, I thought I would become a theoretical

38
00:02:45,440 --> 00:02:49,280
Speaker 1: physicistem that was my ambition. I went to Cambridge University

39
00:02:49,360 --> 00:02:53,120
Speaker 1: to study physics, and I really discovered a number of

40
00:02:53,120 --> 00:02:56,840
Speaker 1: things at that time. Number one, UM, I was I

41
00:02:56,880 --> 00:02:59,200
Speaker 1: thought quite good at physics. Turned out that I surrounded

42
00:02:59,240 --> 00:03:00,880
Speaker 1: myself for the whole bund of other people are also

43
00:03:00,919 --> 00:03:05,919
Speaker 1: pretty good, and really standing out was hard Um. And Second,

44
00:03:06,320 --> 00:03:11,040
Speaker 1: I think at the time the amount of innovation that

45
00:03:11,120 --> 00:03:13,120
Speaker 1: was taking place in physics seemed to have sort of

46
00:03:13,280 --> 00:03:16,720
Speaker 1: dropped off a little bit from the nineteen seventies and

47
00:03:16,720 --> 00:03:18,280
Speaker 1: and it was a sort of slow period in the

48
00:03:18,320 --> 00:03:20,880
Speaker 1: late nineteen eighties, and so that made me think, well,

49
00:03:20,960 --> 00:03:24,400
Speaker 1: maybe I could use these math skills for something else.

50
00:03:24,720 --> 00:03:29,000
Speaker 1: And that really got me into thinking about finance. So

51
00:03:29,040 --> 00:03:33,000
Speaker 1: I some people duck out of physics having done PhD

52
00:03:33,160 --> 00:03:37,360
Speaker 1: s or or taught at universities. I ducked out a

53
00:03:37,360 --> 00:03:40,760
Speaker 1: little bit earlier on UM, and that got me into them,

54
00:03:40,760 --> 00:03:43,680
Speaker 1: thinking well I should use these skills. I end up

55
00:03:43,800 --> 00:03:48,360
Speaker 1: joining Golden Sacks, and and there again I learned something

56
00:03:48,480 --> 00:03:51,960
Speaker 1: which was I thought the exciting bit would be the

57
00:03:52,160 --> 00:03:57,600
Speaker 1: corporate finance areas, so advising on major corporate transactions, and

58
00:03:57,640 --> 00:04:00,080
Speaker 1: what I realized was that that didn't really use the

59
00:04:00,520 --> 00:04:02,600
Speaker 1: quantz skills that I had. So I did that for

60
00:04:02,640 --> 00:04:06,080
Speaker 1: a couple of years, and then I moved over to

61
00:04:06,320 --> 00:04:10,400
Speaker 1: first sixt income research, then equity derivators research, and then

62
00:04:10,480 --> 00:04:15,240
Speaker 1: finally transitioned out of that into UM into more proprietary trading,

63
00:04:15,280 --> 00:04:19,760
Speaker 1: and then into fund management. So let's build off of that.

64
00:04:19,839 --> 00:04:22,799
Speaker 1: You did a lot of work on derivatives and fundamental

65
00:04:22,839 --> 00:04:27,120
Speaker 1: strategy of government. You then go to MAN where you're

66
00:04:27,200 --> 00:04:32,080
Speaker 1: running things like systematic strategies and a h L. What

67
00:04:32,560 --> 00:04:36,480
Speaker 1: was man h L focused on when you were managing that.

68
00:04:37,200 --> 00:04:41,880
Speaker 1: So when I arrived at h L, which was the

69
00:04:42,080 --> 00:04:45,520
Speaker 1: end of two thousand and twelve, was really a futures

70
00:04:45,520 --> 00:04:50,280
Speaker 1: trend following business. It was c t A and c

71
00:04:50,400 --> 00:04:54,080
Speaker 1: t as have been having a pretty difficult time really

72
00:04:54,120 --> 00:04:56,360
Speaker 1: since the end of the financial crisis. They had a

73
00:04:56,360 --> 00:05:00,920
Speaker 1: tremendous two thousand eight and then that really done nothing

74
00:05:00,960 --> 00:05:05,440
Speaker 1: in the following years. So nine eleven, twelve, we're all

75
00:05:05,839 --> 00:05:08,960
Speaker 1: years which essentially added up to nothing. So the fantastic

76
00:05:09,440 --> 00:05:12,640
Speaker 1: crisis here. Very few as the marriagers could say that

77
00:05:12,640 --> 00:05:15,680
Speaker 1: two thousand eight was a great year, but HL could

78
00:05:15,680 --> 00:05:18,600
Speaker 1: definitely say that. But then you had a long dry

79
00:05:18,640 --> 00:05:22,520
Speaker 1: period and people were starting to say momentum doesn't work anymore.

80
00:05:22,640 --> 00:05:27,120
Speaker 1: It's broken. And I think what I really did when

81
00:05:27,120 --> 00:05:29,599
Speaker 1: I arrived in NHL it was to say, well, I've

82
00:05:29,600 --> 00:05:32,080
Speaker 1: been involved in all sorts of different quant strategies in

83
00:05:32,120 --> 00:05:37,240
Speaker 1: my golden years, and I can use a much broader

84
00:05:37,279 --> 00:05:41,160
Speaker 1: perspective than maybe the futures trend followers had and look

85
00:05:41,279 --> 00:05:45,760
Speaker 1: to develop a much wider range of systematic stranities. So

86
00:05:45,800 --> 00:05:49,600
Speaker 1: when I arrived, we had a handful of models UM.

87
00:05:49,640 --> 00:05:53,680
Speaker 1: Today we've got three or four models running in HL,

88
00:05:53,800 --> 00:05:57,279
Speaker 1: so we really expanded number of models we're using. UM,

89
00:05:57,320 --> 00:05:59,880
Speaker 1: we expanded the number of markets that were trading, so

90
00:06:00,080 --> 00:06:06,200
Speaker 1: we used to trade futures, UM and effects markets and

91
00:06:07,160 --> 00:06:09,680
Speaker 1: my group it started trading OTC markets, but it was

92
00:06:09,720 --> 00:06:12,880
Speaker 1: still quite small UM, and we really picked up and

93
00:06:12,960 --> 00:06:16,279
Speaker 1: start trading much wider variety of markets. So now we

94
00:06:16,360 --> 00:06:21,560
Speaker 1: trade around seven markets around the world using system addic models.

95
00:06:21,920 --> 00:06:24,760
Speaker 1: And then finally we came up with different many different

96
00:06:24,760 --> 00:06:27,920
Speaker 1: types of funds, so very short term funds, funds which

97
00:06:27,920 --> 00:06:31,960
Speaker 1: are more fundamentally driven, funds which are maybe trying to

98
00:06:32,560 --> 00:06:36,000
Speaker 1: provide more protection characteristics, or funds trying to maximize the

99
00:06:36,040 --> 00:06:39,800
Speaker 1: sharp ratio. So so we really tried to grow quant

100
00:06:39,880 --> 00:06:44,839
Speaker 1: into many different areas. And I suppose my advantage coming

101
00:06:44,839 --> 00:06:47,920
Speaker 1: into a place like h L is that most people

102
00:06:48,560 --> 00:06:52,080
Speaker 1: in those HL and in the competitors had really grown

103
00:06:52,160 --> 00:06:55,159
Speaker 1: up and had the whole careers in the in the

104
00:06:55,240 --> 00:06:57,800
Speaker 1: c t A or the futures trend following business, and

105
00:06:57,880 --> 00:06:59,960
Speaker 1: I had had none of my career and futures try

106
00:07:00,000 --> 00:07:02,880
Speaker 1: and following, and but I had all these other influences

107
00:07:02,920 --> 00:07:05,640
Speaker 1: that I could bring in, And so that really was

108
00:07:05,720 --> 00:07:09,120
Speaker 1: how I worked with the team now to significantly expand

109
00:07:10,000 --> 00:07:13,480
Speaker 1: UM the business which was having a very difficult time

110
00:07:13,480 --> 00:07:18,760
Speaker 1: and arrived and UM declined to lesson ten billion dollars

111
00:07:18,800 --> 00:07:22,480
Speaker 1: of assets in the h L unit and today were

112
00:07:23,320 --> 00:07:27,080
Speaker 1: many times launcher than now. So I want to focus

113
00:07:27,120 --> 00:07:30,400
Speaker 1: on something you mentioned in passing, but it's so relevant

114
00:07:30,800 --> 00:07:33,600
Speaker 1: to what we've been seeing in the markets recently. You

115
00:07:33,720 --> 00:07:38,120
Speaker 1: said that momentum as a factor seemed to have been fading.

116
00:07:38,760 --> 00:07:43,640
Speaker 1: We've seen other factor based investing like value, go long

117
00:07:43,760 --> 00:07:47,880
Speaker 1: long periods of underperforming. There's so many different questions I

118
00:07:48,240 --> 00:07:51,680
Speaker 1: can ask you about this. Let's just start with is

119
00:07:51,760 --> 00:07:56,840
Speaker 1: that the nature of any factor or any specific trading

120
00:07:56,920 --> 00:08:00,560
Speaker 1: edge that they only last so long before eventually everybody

121
00:08:00,560 --> 00:08:04,200
Speaker 1: wise is up to them and the alpha gets arbitraged away.

122
00:08:04,240 --> 00:08:07,400
Speaker 1: And do you see these sort of edges disappearing more

123
00:08:07,480 --> 00:08:13,080
Speaker 1: quickly these days then they used to in a perhaps kindler,

124
00:08:13,160 --> 00:08:17,480
Speaker 1: gentler era twenty or thirty years ago. So, um, I

125
00:08:17,560 --> 00:08:19,960
Speaker 1: agree actually with everything you said, Barry, except for the

126
00:08:20,040 --> 00:08:22,840
Speaker 1: last bit about kind of legenter era. From my memory

127
00:08:22,880 --> 00:08:24,880
Speaker 1: twenty or thirty years ago, it was probably less kind

128
00:08:24,960 --> 00:08:29,119
Speaker 1: and less gentle than it is today. Um, But let's

129
00:08:29,280 --> 00:08:33,040
Speaker 1: let's start with factors. So I think one of the

130
00:08:33,040 --> 00:08:36,360
Speaker 1: things which has been interesting over my career is nobody

131
00:08:36,360 --> 00:08:39,440
Speaker 1: really talked about factors apart from a very small sort

132
00:08:39,480 --> 00:08:42,520
Speaker 1: of quant group twenty thirty years ago. Today they're in,

133
00:08:42,920 --> 00:08:44,920
Speaker 1: you know that they're sort of part of all our

134
00:08:44,960 --> 00:08:48,480
Speaker 1: portfolio marriagers at Man Group, whether they're quants or discretion marriages.

135
00:08:48,520 --> 00:08:51,600
Speaker 1: Everybody talks factors and so that's been a big change,

136
00:08:51,600 --> 00:08:54,640
Speaker 1: and they've become sort of part of, you know, just

137
00:08:54,679 --> 00:08:58,840
Speaker 1: sort of general dialogue when people are talking about markets.

138
00:08:59,640 --> 00:09:04,280
Speaker 1: The thing I'd say is that I don't think that, um,

139
00:09:05,080 --> 00:09:07,680
Speaker 1: the core factors which have been around in a long

140
00:09:07,720 --> 00:09:11,960
Speaker 1: time are going to disappear. And for me, you know,

141
00:09:11,960 --> 00:09:14,680
Speaker 1: as a European, I was working in New York in

142
00:09:14,679 --> 00:09:17,720
Speaker 1: the late nine nineties and into the two thousand's, and

143
00:09:17,760 --> 00:09:20,079
Speaker 1: I remember looking at front page of the Wall Street

144
00:09:20,160 --> 00:09:24,599
Speaker 1: Journal one day and on it it said value investing

145
00:09:24,679 --> 00:09:28,360
Speaker 1: is for old people and um and I you know,

146
00:09:28,360 --> 00:09:30,680
Speaker 1: as the European obviously, you know Europeans want to live

147
00:09:30,720 --> 00:09:33,319
Speaker 1: in old houses. Clearly in the US people mostly want

148
00:09:33,320 --> 00:09:34,840
Speaker 1: to live in new houses. So it's a big sort

149
00:09:34,840 --> 00:09:40,600
Speaker 1: of didn't quite understand that the the the extent of

150
00:09:40,720 --> 00:09:43,600
Speaker 1: the statement there, and it was a ridiculous thing to say.

151
00:09:43,640 --> 00:09:45,480
Speaker 1: I was a young person being included in the front

152
00:09:45,480 --> 00:09:48,760
Speaker 1: page of the Wall Street Journal, right in the tail

153
00:09:48,920 --> 00:09:51,280
Speaker 1: end of the tech bubble and just before a huge

154
00:09:51,440 --> 00:09:57,400
Speaker 1: outperformance of value stocks. So these factors and we should

155
00:09:57,440 --> 00:09:59,880
Speaker 1: talk about, you know, which are the which are the

156
00:10:00,000 --> 00:10:02,040
Speaker 1: actors that are likely to persist in which are not

157
00:10:02,720 --> 00:10:08,880
Speaker 1: um But factors like value for example, I think generally

158
00:10:08,920 --> 00:10:12,640
Speaker 1: don't have particularly high sharp ratios, so they're their returns

159
00:10:12,679 --> 00:10:16,840
Speaker 1: adjusted for risk are not particularly high. But the idea

160
00:10:16,920 --> 00:10:20,160
Speaker 1: that buying cheap stocks will never work again, I've never

161
00:10:20,200 --> 00:10:23,240
Speaker 1: thought that was a sensible thing to say or think,

162
00:10:23,559 --> 00:10:25,040
Speaker 1: but every now and then, you know, that's what this

163
00:10:25,120 --> 00:10:27,440
Speaker 1: fellow on the front page of the journal said um

164
00:10:27,600 --> 00:10:31,840
Speaker 1: twenty odd years ago. So I think factors, at least

165
00:10:31,920 --> 00:10:37,040
Speaker 1: a core set of factors are very likely to persist,

166
00:10:37,200 --> 00:10:40,079
Speaker 1: but it won't give particularly amazing risk adjusted returns, But

167
00:10:40,120 --> 00:10:42,760
Speaker 1: I think they are likely to give you positive risk

168
00:10:42,800 --> 00:10:47,960
Speaker 1: adjusted returns over you know, relatively long cycles. I would say, though,

169
00:10:48,000 --> 00:10:50,720
Speaker 1: that one of the things I've seen in the last

170
00:10:50,720 --> 00:10:54,520
Speaker 1: ten years or so is as people thought they understood factors,

171
00:10:54,559 --> 00:10:57,200
Speaker 1: and people started to find hundreds of these things. And

172
00:10:57,320 --> 00:11:00,240
Speaker 1: I don't think there are hundreds of real factors. Think

173
00:11:00,280 --> 00:11:05,520
Speaker 1: there's a small handful um of factors, and this explosion

174
00:11:05,640 --> 00:11:09,880
Speaker 1: really is an overfitting exercise. It's people finding patterns in

175
00:11:09,920 --> 00:11:13,080
Speaker 1: the data that don't really exist, and I think that's

176
00:11:13,120 --> 00:11:15,640
Speaker 1: something that people should be very wary of. You know,

177
00:11:15,679 --> 00:11:21,319
Speaker 1: I've seen data providers and firms sell their libraries of

178
00:11:21,400 --> 00:11:24,520
Speaker 1: factors with with you know, literally hundreds of these things,

179
00:11:24,559 --> 00:11:26,600
Speaker 1: and I don't think that's going to be a source

180
00:11:26,640 --> 00:11:29,440
Speaker 1: of returns. The final thing I should say is that

181
00:11:29,640 --> 00:11:32,280
Speaker 1: you mentioned momentum, and momentum is quite an interesting factor

182
00:11:33,040 --> 00:11:39,199
Speaker 1: because there are two very different definitions of momentum. One

183
00:11:39,679 --> 00:11:43,640
Speaker 1: is really used by equities people and they will go

184
00:11:43,760 --> 00:11:48,679
Speaker 1: along the positive momentum price momentum stocks and short the

185
00:11:48,720 --> 00:11:51,560
Speaker 1: negative price momentum stocks. They sometimes do it with earnings

186
00:11:51,559 --> 00:11:55,000
Speaker 1: as well, but basically go along the stocks which have

187
00:11:55,080 --> 00:11:57,720
Speaker 1: been out performing and short the stocks which have been

188
00:11:57,760 --> 00:12:01,880
Speaker 1: under performing. But then is a very different definition of momentum,

189
00:12:02,160 --> 00:12:03,720
Speaker 1: and that's what the c t as use, And they

190
00:12:03,760 --> 00:12:06,680
Speaker 1: don't they don't look at the price move against anything.

191
00:12:06,880 --> 00:12:08,960
Speaker 1: They just do it in absolute terms. And they tend

192
00:12:09,000 --> 00:12:11,800
Speaker 1: to do that in macro markets. So they'll create the

193
00:12:11,920 --> 00:12:19,000
Speaker 1: SMP or the or the the dacks or the euro

194
00:12:19,520 --> 00:12:22,560
Speaker 1: or gold or something like that. And and that's a

195
00:12:22,679 --> 00:12:25,720
Speaker 1: much different definition because it's not it's not going long

196
00:12:25,800 --> 00:12:28,320
Speaker 1: one set of markets and or one set of stocks

197
00:12:28,320 --> 00:12:30,480
Speaker 1: and short another set of stocks. It could be long everything,

198
00:12:30,559 --> 00:12:32,680
Speaker 1: or it could be short everything, and it gives you

199
00:12:32,720 --> 00:12:36,400
Speaker 1: a very different type of return profile. That second type

200
00:12:36,840 --> 00:12:42,679
Speaker 1: has a very nice feature, which is that it barely

201
00:12:42,800 --> 00:12:46,480
Speaker 1: reliably will do well in bad periods in markets. So

202
00:12:46,480 --> 00:12:49,640
Speaker 1: I talked about how, for example, HL two thousand eight

203
00:12:49,720 --> 00:12:53,760
Speaker 1: was an excellent year. Well, not many strategies that could

204
00:12:53,760 --> 00:12:55,680
Speaker 1: say two thousand eight was an excellent year. And that's

205
00:12:55,720 --> 00:12:59,160
Speaker 1: because that second definition of momentum what I would call

206
00:12:59,240 --> 00:13:04,480
Speaker 1: time series momentum the or univariate momentum. That that definition

207
00:13:04,559 --> 00:13:08,760
Speaker 1: momentum has very good protection like characteristics. It will pick

208
00:13:08,840 --> 00:13:11,800
Speaker 1: up on a trend, especially in negative trend in markets,

209
00:13:11,840 --> 00:13:15,360
Speaker 1: and jump on that trend. So it sounds like the

210
00:13:15,440 --> 00:13:21,640
Speaker 1: price momentum seems to be relative, while the time series

211
00:13:21,679 --> 00:13:25,439
Speaker 1: has a persistence that gives it a very different characteristic.

212
00:13:25,600 --> 00:13:28,320
Speaker 1: Or am I oversimplifying that? No? No, I think that's

213
00:13:28,320 --> 00:13:33,160
Speaker 1: exactly right. So I think you know, the most investing strategies,

214
00:13:33,360 --> 00:13:39,200
Speaker 1: including price momentum and equities, but most investing strategies have

215
00:13:39,400 --> 00:13:41,559
Speaker 1: what people like me would call a left tail, so

216
00:13:41,679 --> 00:13:43,760
Speaker 1: it make you money most of the time, and then

217
00:13:43,800 --> 00:13:46,400
Speaker 1: every now and then they serve your up an unpleasant surprise.

218
00:13:47,559 --> 00:13:52,080
Speaker 1: Time series momentum does the opposite. Time series amentum most

219
00:13:52,120 --> 00:13:55,280
Speaker 1: of the time gives you pretty boring returns, but every

220
00:13:55,320 --> 00:13:57,360
Speaker 1: now and then I'll give you a very positive surprise.

221
00:13:57,640 --> 00:14:02,720
Speaker 1: And and that's really rare in investing strategies. And from

222
00:14:02,720 --> 00:14:06,079
Speaker 1: my perspective, that's you know, that's a very attractive characteristic

223
00:14:06,120 --> 00:14:08,959
Speaker 1: when you're building portfolios, to have a bit of something

224
00:14:09,000 --> 00:14:12,959
Speaker 1: which does the opposite of most other investing strategies. Let's

225
00:14:13,000 --> 00:14:15,960
Speaker 1: talk a little bit about the VIX index, which you

226
00:14:16,480 --> 00:14:20,400
Speaker 1: were the co inventor of tell us. How does one

227
00:14:20,440 --> 00:14:25,000
Speaker 1: go about inventing the VIX index? Well, you know very

228
00:14:25,040 --> 00:14:29,240
Speaker 1: In short, you get lucky. So the the story behind

229
00:14:29,360 --> 00:14:33,200
Speaker 1: it is, as we talked about in our earlier segment.

230
00:14:33,400 --> 00:14:36,320
Speaker 1: I was working in New York basically and see clients

231
00:14:36,360 --> 00:14:38,560
Speaker 1: and every time you went to somebody's office to be

232
00:14:38,600 --> 00:14:42,520
Speaker 1: a TV screen in the in the fire in the

233
00:14:42,640 --> 00:14:45,560
Speaker 1: entrance area, and they would have prices of all sorts

234
00:14:45,600 --> 00:14:47,880
Speaker 1: of things coming across that screen, you know, the sort

235
00:14:47,880 --> 00:14:51,160
Speaker 1: of the price of crude ale, the treasury bond yield,

236
00:14:51,160 --> 00:14:53,000
Speaker 1: the sple hundred level, that sort of thing, and it

237
00:14:53,040 --> 00:14:55,720
Speaker 1: would have the VIX on it. And so there was

238
00:14:55,760 --> 00:14:58,560
Speaker 1: a VIX, but that was the VIX was the only

239
00:14:58,600 --> 00:15:00,600
Speaker 1: thing that seemed to come across the greens that you

240
00:15:00,640 --> 00:15:06,640
Speaker 1: couldn't trade. And so the story really came because a

241
00:15:06,800 --> 00:15:09,320
Speaker 1: colleague of mine at Golden Sacks came to me one

242
00:15:09,360 --> 00:15:12,320
Speaker 1: evening and said, you know, I've had this call um

243
00:15:12,360 --> 00:15:15,360
Speaker 1: from a client who wants to do a trade on

244
00:15:15,400 --> 00:15:20,040
Speaker 1: the VIX. Could we do that? And my colleague ran

245
00:15:20,720 --> 00:15:24,360
Speaker 1: options trading and I ran the drug to his research

246
00:15:24,920 --> 00:15:31,200
Speaker 1: a bit of Goldman and we got our heads together

247
00:15:31,360 --> 00:15:33,760
Speaker 1: and I went and found the formula for the VIX

248
00:15:33,880 --> 00:15:36,920
Speaker 1: as stood as as pre existed, and we worked at

249
00:15:37,000 --> 00:15:38,920
Speaker 1: it just wasn't possible for us to do a trade

250
00:15:38,960 --> 00:15:41,240
Speaker 1: on that. It wasn't designed in a way that you

251
00:15:41,320 --> 00:15:47,400
Speaker 1: could hedge a trade on the VIX. So we came

252
00:15:47,480 --> 00:15:49,800
Speaker 1: up with a kind of crazy idea which was, Okay,

253
00:15:49,800 --> 00:15:52,240
Speaker 1: well there's this VIX thing which was owned by the

254
00:15:52,280 --> 00:15:55,560
Speaker 1: cbo E, but you couldn't trade it. And there was

255
00:15:55,560 --> 00:15:57,120
Speaker 1: a good reason why you couldn't trade it, because you

256
00:15:57,120 --> 00:16:00,120
Speaker 1: couldn't hedge it. So why don't we change it? And

257
00:16:00,320 --> 00:16:03,880
Speaker 1: so we came up with a completely different formula. It

258
00:16:03,920 --> 00:16:08,200
Speaker 1: turned out to give fairly similar levels to the old VIX,

259
00:16:09,200 --> 00:16:11,680
Speaker 1: but but it was a completely different formula, didn't use

260
00:16:11,680 --> 00:16:15,680
Speaker 1: black sholes at all and um and we thought, well, hey,

261
00:16:15,720 --> 00:16:20,960
Speaker 1: this this formula actually you could hedge UM futures contract

262
00:16:21,080 --> 00:16:24,480
Speaker 1: on or something like that. And so what we then

263
00:16:24,520 --> 00:16:29,000
Speaker 1: did is I was quite friendly with a fellow called

264
00:16:29,000 --> 00:16:32,440
Speaker 1: Bill Speth, but the CBOE, who has had of research there.

265
00:16:32,920 --> 00:16:34,880
Speaker 1: And I called Bill and said, you know, you have

266
00:16:35,000 --> 00:16:36,720
Speaker 1: this VIX but you make no money out of it,

267
00:16:36,760 --> 00:16:39,240
Speaker 1: because you just published this thing and doesn't give you

268
00:16:39,280 --> 00:16:42,200
Speaker 1: any income. And we've got an idea how you could

269
00:16:42,280 --> 00:16:45,560
Speaker 1: change it to make it something that could be traded,

270
00:16:45,600 --> 00:16:47,680
Speaker 1: and maybe you could launch futures contracts on that, and

271
00:16:47,720 --> 00:16:50,120
Speaker 1: so that might be interesting to you. And then I

272
00:16:50,240 --> 00:16:52,840
Speaker 1: launched into a long description of the math behind that formula,

273
00:16:53,320 --> 00:16:56,680
Speaker 1: and UM Bill very wisely said, you know, maybe you

274
00:16:56,680 --> 00:16:59,160
Speaker 1: could send me a letter with that formula. So so

275
00:16:59,200 --> 00:17:02,640
Speaker 1: I sent him a letter and which was actually, with

276
00:17:02,680 --> 00:17:04,600
Speaker 1: the benefit of findset quite helpful because now I have

277
00:17:04,640 --> 00:17:07,920
Speaker 1: a quite clear record of when we communicated this formula

278
00:17:08,640 --> 00:17:12,119
Speaker 1: in two thousand three to the cbo E UM and

279
00:17:13,200 --> 00:17:16,360
Speaker 1: UM and within six months we had a new VIX

280
00:17:16,880 --> 00:17:21,320
Speaker 1: being calculated using this new formula, and another six months

281
00:17:21,359 --> 00:17:24,600
Speaker 1: later the cbo E, which until that point was only

282
00:17:24,640 --> 00:17:28,520
Speaker 1: an options exchange, had launched futures contracts on it. And

283
00:17:28,560 --> 00:17:30,919
Speaker 1: so there was a series of sort of you know,

284
00:17:31,040 --> 00:17:33,880
Speaker 1: lucky moments in there. A client came and asked a question.

285
00:17:34,680 --> 00:17:37,760
Speaker 1: I happened to know Bill at the cbo E, so

286
00:17:38,080 --> 00:17:41,640
Speaker 1: I knew somebody to call. They happened to be interested

287
00:17:41,720 --> 00:17:46,000
Speaker 1: in in launching futures contracts and our timing was was

288
00:17:46,040 --> 00:17:49,920
Speaker 1: spot on. So there was a whole series of bits

289
00:17:49,960 --> 00:17:53,280
Speaker 1: of luck along the way. And what really happened after

290
00:17:53,320 --> 00:17:57,240
Speaker 1: that is another good learning experience for me, which is

291
00:17:57,720 --> 00:17:59,680
Speaker 1: at a boss at the time who said, look, you

292
00:17:59,680 --> 00:18:02,320
Speaker 1: should talk to all the salespeople and find out if

293
00:18:02,320 --> 00:18:05,960
Speaker 1: they're gonna bring in business on this new Vix thing

294
00:18:06,080 --> 00:18:07,800
Speaker 1: that you've been working on. So I spoke to the

295
00:18:07,840 --> 00:18:10,720
Speaker 1: salespeople and we did a little survey, and you know,

296
00:18:10,760 --> 00:18:14,120
Speaker 1: I was going to retire on the proceeds of this survey.

297
00:18:14,119 --> 00:18:16,199
Speaker 1: It was just amazing how much business we were going

298
00:18:16,240 --> 00:18:19,000
Speaker 1: to do. Day one comes along, I wait for the

299
00:18:19,040 --> 00:18:22,359
Speaker 1: phone to ring, and it doesn't ring. They two comes along,

300
00:18:23,119 --> 00:18:26,320
Speaker 1: the phone still doesn't ring, and by day three you

301
00:18:26,400 --> 00:18:30,440
Speaker 1: sort of get it. You know, it's nothing's happening. And amazingly,

302
00:18:30,560 --> 00:18:33,119
Speaker 1: the first significant trades we got done were actually with

303
00:18:33,240 --> 00:18:37,439
Speaker 1: investors outside of the us UM and they were the

304
00:18:37,440 --> 00:18:39,480
Speaker 1: people that still have got the early stages of this

305
00:18:39,680 --> 00:18:43,080
Speaker 1: new market in in Vix going. The other thing that

306
00:18:43,160 --> 00:18:46,639
Speaker 1: happened was my boss said, you should call, you know,

307
00:18:46,680 --> 00:18:48,920
Speaker 1: some of the other banks and and see if they're

308
00:18:48,920 --> 00:18:52,040
Speaker 1: going to support this thing. So I called nine other banks,

309
00:18:52,080 --> 00:18:54,280
Speaker 1: and nine out of nine said they had no interest

310
00:18:54,320 --> 00:18:58,080
Speaker 1: in supporting it. So it wasn't a particularly auspicious beginning,

311
00:18:58,520 --> 00:19:00,920
Speaker 1: you know that, And it took a little bit of persistence.

312
00:19:00,920 --> 00:19:05,480
Speaker 1: Now today it's this huge market and it trades enormous volumes,

313
00:19:05,520 --> 00:19:09,240
Speaker 1: but it's a good sort of lesson in terms of

314
00:19:09,240 --> 00:19:12,639
Speaker 1: how difficult it is to get something new going. So

315
00:19:12,640 --> 00:19:14,800
Speaker 1: so let's talk a little bit about what the VIX

316
00:19:14,880 --> 00:19:20,360
Speaker 1: index does and some of the misunderstandings around it. When

317
00:19:20,440 --> 00:19:24,720
Speaker 1: when you talk to people, especially traders on the equity side,

318
00:19:25,160 --> 00:19:29,360
Speaker 1: they look at it as the fear index. It measures volatility,

319
00:19:29,400 --> 00:19:35,840
Speaker 1: but but to be more precise, it's really measuring volatility expectations. Right, Yes,

320
00:19:36,160 --> 00:19:40,720
Speaker 1: I think the the word fear gauge or fear index

321
00:19:40,960 --> 00:19:45,880
Speaker 1: is actually pretty accurate. But you're also absolutely correct that

322
00:19:46,480 --> 00:19:51,479
Speaker 1: what the VIX actually is is the market's expectation of

323
00:19:51,600 --> 00:19:56,679
Speaker 1: alterality over the next thirty calendar days. And so it

324
00:19:56,880 --> 00:20:04,000
Speaker 1: is a it's a market price, and inevitably because the

325
00:20:04,200 --> 00:20:07,840
Speaker 1: volatality cannot mathematically go below zero, there's no such thing

326
00:20:07,880 --> 00:20:11,680
Speaker 1: as vaultality below zero. But it's unlimited on the upside,

327
00:20:11,720 --> 00:20:14,720
Speaker 1: you know, there's no limit to how high vaultality can be.

328
00:20:15,080 --> 00:20:16,640
Speaker 1: Then if you go to the market and say, hey,

329
00:20:16,680 --> 00:20:19,920
Speaker 1: give me a price for the next thirty days of alterality,

330
00:20:20,000 --> 00:20:22,320
Speaker 1: it's generally going to overestimate because it's going to have

331
00:20:22,400 --> 00:20:25,960
Speaker 1: to protect itself a bit against the possibility of huge

332
00:20:25,960 --> 00:20:28,480
Speaker 1: swings upwards and the fact that there's a floor. It

333
00:20:28,520 --> 00:20:32,640
Speaker 1: can't go below zero. So your description is absolutely correct.

334
00:20:33,080 --> 00:20:38,040
Speaker 1: It's the market's expectation of realized volatility over the next

335
00:20:38,080 --> 00:20:41,760
Speaker 1: thirty calendar days, but it has some features built into it.

336
00:20:41,760 --> 00:20:46,080
Speaker 1: It's always going to overestimate because you're effectively selling insurance

337
00:20:46,119 --> 00:20:48,880
Speaker 1: if you're selling the vics and and people don't sell

338
00:20:48,960 --> 00:20:51,720
Speaker 1: insurance cheap for the most part. That makes a whole

339
00:20:51,720 --> 00:20:54,120
Speaker 1: lot of sense. I know that a lot of traders

340
00:20:54,160 --> 00:20:59,560
Speaker 1: seem to conflate volatility with risk. How do you define

341
00:20:59,600 --> 00:21:02,359
Speaker 1: the different is between the two. It's a very good question,

342
00:21:02,359 --> 00:21:05,000
Speaker 1: and I think you know something I've thought about a

343
00:21:05,080 --> 00:21:09,840
Speaker 1: lot over the years, um and there's many different versions

344
00:21:09,880 --> 00:21:13,840
Speaker 1: of this. I think so most models, most risk models

345
00:21:13,920 --> 00:21:18,280
Speaker 1: do estimate volatility. They give you a var number, which

346
00:21:18,359 --> 00:21:24,240
Speaker 1: is basically a manipulated volatility number, or a risk in

347
00:21:24,359 --> 00:21:28,000
Speaker 1: stand deviations, or an expectation of loss or something like that,

348
00:21:28,720 --> 00:21:33,280
Speaker 1: and they're useful numbers, but in the end, I'm pretty

349
00:21:33,320 --> 00:21:37,359
Speaker 1: sure that both you and certainly I would never clue

350
00:21:37,400 --> 00:21:41,880
Speaker 1: what the volatility of our personal portfolios was last year

351
00:21:41,960 --> 00:21:45,040
Speaker 1: in a very volatile year. But we have a pretty

352
00:21:45,040 --> 00:21:47,359
Speaker 1: good idea of what the worst point is, you know,

353
00:21:47,400 --> 00:21:50,120
Speaker 1: when when we have the most losses of most pain

354
00:21:50,240 --> 00:21:53,720
Speaker 1: in our portfolios, and that's got nothing to do with voltility.

355
00:21:53,800 --> 00:21:57,160
Speaker 1: That's a draw down. So in the end, actually the

356
00:21:57,280 --> 00:21:59,840
Speaker 1: risk that a lot of us really experience and worry

357
00:21:59,880 --> 00:22:04,879
Speaker 1: of out his drawdowns. It's not volatility, which is mathematical

358
00:22:04,920 --> 00:22:08,000
Speaker 1: formula which describes the shape of a of a distribution,

359
00:22:08,600 --> 00:22:12,840
Speaker 1: and that I think is something which is difficult because

360
00:22:13,000 --> 00:22:18,760
Speaker 1: estimating drawdowns is extremely hard. Estimating volatility is actually fairly straightforward.

361
00:22:19,040 --> 00:22:21,760
Speaker 1: But the two don't really connect. And so why do

362
00:22:21,800 --> 00:22:24,680
Speaker 1: people estimate volatility? Because it's useful. It gives you how

363
00:22:24,680 --> 00:22:28,640
Speaker 1: wide the distribution will be. You can estimate it quite accurately,

364
00:22:28,680 --> 00:22:30,959
Speaker 1: you can forecast it quite well as well. You can

365
00:22:31,000 --> 00:22:34,439
Speaker 1: forecast volatality much better than you can forecast returns. So

366
00:22:34,480 --> 00:22:37,040
Speaker 1: all of that is useful. But what's not useful is

367
00:22:37,080 --> 00:22:41,640
Speaker 1: that we don't really worry in the end about what

368
00:22:41,680 --> 00:22:44,840
Speaker 1: I've altilt he was last year or what it will

369
00:22:44,880 --> 00:22:46,919
Speaker 1: be next year. What we really worry about is how

370
00:22:47,000 --> 00:22:49,560
Speaker 1: much we lost or how much we might lose. That

371
00:22:49,680 --> 00:22:53,760
Speaker 1: sort of that that pain threshold, and volatility doesn't really

372
00:22:53,840 --> 00:22:58,960
Speaker 1: connect with that. So unfortunately, really useful statistic drawdowns very

373
00:22:58,960 --> 00:23:02,159
Speaker 1: hard to estimate, and people don't really estimate it. The

374
00:23:02,200 --> 00:23:05,960
Speaker 1: other statistic volatility is useful, but it's in my view,

375
00:23:06,000 --> 00:23:09,760
Speaker 1: not the most useful estimate of of of risk because

376
00:23:09,800 --> 00:23:13,760
Speaker 1: we actually experience it. That's that's really kind of intriguing.

377
00:23:14,320 --> 00:23:18,960
Speaker 1: So so not only can you go long the vix um,

378
00:23:18,960 --> 00:23:21,720
Speaker 1: but some people can go short the VIX not not

379
00:23:21,880 --> 00:23:26,159
Speaker 1: exactly um a hedge if you have a long portfolio,

380
00:23:26,240 --> 00:23:29,200
Speaker 1: but I guess if you're short, maybe that that useful.

381
00:23:29,359 --> 00:23:33,000
Speaker 1: What do you think of of how people have been

382
00:23:33,240 --> 00:23:36,679
Speaker 1: using the VIX, either as a risk management tool or

383
00:23:36,720 --> 00:23:40,240
Speaker 1: as a way to UM get some non correlated exposure

384
00:23:40,280 --> 00:23:43,360
Speaker 1: to their to their equity holdings. Well, I think there's

385
00:23:43,359 --> 00:23:49,359
Speaker 1: a few things in there. Firstly, the the VIX and

386
00:23:49,440 --> 00:23:52,840
Speaker 1: the price of futures on the VIX will nearly always

387
00:23:53,240 --> 00:23:56,480
Speaker 1: disagree in generally, the futures will be higher than the

388
00:23:56,520 --> 00:23:59,359
Speaker 1: current level. So you know, as we're speaking now, the

389
00:23:59,440 --> 00:24:03,920
Speaker 1: VIX is a round sixteen better futures contract three months

390
00:24:03,920 --> 00:24:07,919
Speaker 1: out is trading at almost twenty one, so five points higher,

391
00:24:08,160 --> 00:24:10,600
Speaker 1: and that is a normal state of affairs. And as

392
00:24:10,600 --> 00:24:14,560
Speaker 1: we talked about earlier on voltialty, because voltality can't go

393
00:24:14,640 --> 00:24:17,359
Speaker 1: below zero than people, and it can go to an

394
00:24:17,440 --> 00:24:22,639
Speaker 1: unlimited upside level than people, generally the market will overestimate

395
00:24:23,440 --> 00:24:26,040
Speaker 1: volatilty to give it a little bit of an insurance

396
00:24:26,040 --> 00:24:30,240
Speaker 1: premium in there. Now, in terms of trading and investing

397
00:24:30,440 --> 00:24:33,160
Speaker 1: with the VIX, I think this thing is quite important

398
00:24:33,240 --> 00:24:35,560
Speaker 1: to know. So if you buy an ETS on the VIX,

399
00:24:35,640 --> 00:24:37,840
Speaker 1: for example, then it is going to have to hedge

400
00:24:37,840 --> 00:24:41,640
Speaker 1: itself with these futures contracts, and these futures contracts will

401
00:24:41,680 --> 00:24:45,560
Speaker 1: trade a lot higher. I said twenty one for September,

402
00:24:45,720 --> 00:24:49,680
Speaker 1: verus is sixteen now, so that's almost a thirty percent higher.

403
00:24:50,040 --> 00:24:52,920
Speaker 1: And by September, one of two things has to happen.

404
00:24:52,960 --> 00:24:54,800
Speaker 1: Either the VIX has to go up to twenty one,

405
00:24:55,080 --> 00:24:57,920
Speaker 1: or the futures contract will go down to sixteen. More lightly,

406
00:24:57,960 --> 00:25:00,360
Speaker 1: the futures contract will go down to sixty. And so

407
00:25:00,640 --> 00:25:03,639
Speaker 1: I think people often don't understand this when they're buying

408
00:25:03,720 --> 00:25:06,840
Speaker 1: a vix et f that they're not buying the level

409
00:25:06,880 --> 00:25:09,320
Speaker 1: of the vix they see on the on the screen

410
00:25:09,400 --> 00:25:12,320
Speaker 1: on their television screens or on their Bloomberg terminals or

411
00:25:12,320 --> 00:25:16,359
Speaker 1: wherever they see it. They're they're buying effectively a future

412
00:25:16,480 --> 00:25:18,920
Speaker 1: on the vix which generally is trading a much higher level,

413
00:25:18,960 --> 00:25:21,879
Speaker 1: and so they will have expected losses built in. That

414
00:25:22,160 --> 00:25:25,560
Speaker 1: is something which then some people on the other side

415
00:25:25,560 --> 00:25:27,119
Speaker 1: have said, well, this is very exciting. You know I

416
00:25:27,160 --> 00:25:31,560
Speaker 1: can I can sell the vix at twenty one in September,

417
00:25:31,600 --> 00:25:33,760
Speaker 1: and I expect it will go to sixteen. I can

418
00:25:33,800 --> 00:25:37,680
Speaker 1: make thirty percent in three months. And they're absolutely right,

419
00:25:38,640 --> 00:25:42,320
Speaker 1: you can and most likely will make thirty percent in

420
00:25:42,400 --> 00:25:45,000
Speaker 1: three months. But and the butt is a big thing.

421
00:25:45,359 --> 00:25:49,040
Speaker 1: If if something bad happens between now in September, then

422
00:25:49,160 --> 00:25:52,680
Speaker 1: you know, you can make very very significant losses because

423
00:25:52,880 --> 00:25:55,240
Speaker 1: the vix can just, you know, very very quickly go

424
00:25:55,400 --> 00:25:58,080
Speaker 1: up to very high levels. So what people often do

425
00:25:58,200 --> 00:26:02,159
Speaker 1: is they say that buying the vix that gives me

426
00:26:02,280 --> 00:26:05,840
Speaker 1: some protection against crises. If there's a crisis, then it

427
00:26:05,840 --> 00:26:08,000
Speaker 1: will lightly go up, and they're right, but they do

428
00:26:08,080 --> 00:26:11,040
Speaker 1: have to understand that it's not the VIX going from

429
00:26:11,080 --> 00:26:13,399
Speaker 1: the current sixteen to say eighteen. It's got to go

430
00:26:13,480 --> 00:26:16,520
Speaker 1: above that twenty one that's priced in in September before

431
00:26:16,560 --> 00:26:19,639
Speaker 1: you make any money. Um. But it can be a

432
00:26:19,680 --> 00:26:22,720
Speaker 1: protection strategy and insurance strategy. And then you've got people

433
00:26:22,720 --> 00:26:26,960
Speaker 1: on the other side who, let's say, look, I don't

434
00:26:27,000 --> 00:26:29,080
Speaker 1: expect something bad to happen, and I can earn this

435
00:26:29,320 --> 00:26:33,399
Speaker 1: very large insurance premium UM if I'm prepared to go

436
00:26:33,520 --> 00:26:35,680
Speaker 1: short the VIX and you of course can do that

437
00:26:35,800 --> 00:26:39,960
Speaker 1: through futures contracts, but you can also do it through ets.

438
00:26:40,560 --> 00:26:43,480
Speaker 1: I think my real observation on this those you know,

439
00:26:43,480 --> 00:26:45,679
Speaker 1: I've tried to give as clear an explanation as I

440
00:26:45,720 --> 00:26:48,240
Speaker 1: can of how this is working, but it's quite subtle.

441
00:26:48,480 --> 00:26:50,399
Speaker 1: This is not a simple thing. And I think a

442
00:26:50,440 --> 00:26:52,920
Speaker 1: lot of people that trade the VIX cts don't really

443
00:26:53,000 --> 00:26:57,320
Speaker 1: understand what's going on underneath the surface of the TS contract.

444
00:26:57,359 --> 00:27:00,080
Speaker 1: And there's a lot going on underneath the surface. So

445
00:27:00,080 --> 00:27:03,400
Speaker 1: so let's talk a little bit about um, people trading

446
00:27:03,560 --> 00:27:06,639
Speaker 1: products that they didn't really understand. And who better to

447
00:27:06,680 --> 00:27:10,600
Speaker 1: ask you the co inventor of the VIX back in

448
00:27:10,720 --> 00:27:15,600
Speaker 1: ten we saw the notes UH that were based on

449
00:27:15,640 --> 00:27:18,159
Speaker 1: the VIX and Credit Swiss was one of the larger

450
00:27:18,400 --> 00:27:21,879
Speaker 1: underwriters of these. UM just blow up and sent the

451
00:27:21,960 --> 00:27:25,439
Speaker 1: VIC spiking. I kind of remember we kissed fifty I

452
00:27:25,480 --> 00:27:28,760
Speaker 1: could I could be wrong about that, UM, but that

453
00:27:28,880 --> 00:27:32,240
Speaker 1: whole series of products, those short term E t N

454
00:27:32,359 --> 00:27:35,520
Speaker 1: s UM x I V was one and s v

455
00:27:35,800 --> 00:27:40,000
Speaker 1: x Y was another. UM. They just blew up spectacularly.

456
00:27:41,119 --> 00:27:44,920
Speaker 1: As you're watching this from from your seat, what are

457
00:27:44,920 --> 00:27:47,840
Speaker 1: you thinking about? Gee? Look what look? I lent the

458
00:27:47,920 --> 00:27:50,560
Speaker 1: keys to the car to the kids and they seem

459
00:27:50,640 --> 00:27:54,159
Speaker 1: to have wrecked it. Oh. I think I think your

460
00:27:54,240 --> 00:27:59,320
Speaker 1: description that Barry is pretty fair. When we did the works,

461
00:27:59,680 --> 00:28:01,840
Speaker 1: I'm Aaron. I did the work back in two thousand

462
00:28:01,880 --> 00:28:04,800
Speaker 1: three two four. The boss that asked me to do

463
00:28:04,800 --> 00:28:06,879
Speaker 1: the various other things that we talked about said, you know,

464
00:28:06,920 --> 00:28:09,760
Speaker 1: you should look at creating an ETF on this thing,

465
00:28:10,280 --> 00:28:12,720
Speaker 1: and so we did. Then we got together with one

466
00:28:12,720 --> 00:28:17,000
Speaker 1: of the very big E t F providers and they said, well, look,

467
00:28:17,040 --> 00:28:19,040
Speaker 1: maybe you could do some modeling of how this thing

468
00:28:19,080 --> 00:28:23,080
Speaker 1: will behave and we did that and we concluded this

469
00:28:23,160 --> 00:28:25,720
Speaker 1: is just not a good product. You know, people it's

470
00:28:25,760 --> 00:28:30,280
Speaker 1: got some nasty characteristics, and so we decided, along with

471
00:28:30,359 --> 00:28:33,560
Speaker 1: that large firm, that we should not sell E T

472
00:28:33,800 --> 00:28:37,400
Speaker 1: s on the vix. Now, other people took a different view,

473
00:28:37,520 --> 00:28:39,600
Speaker 1: and so your analogy of kind of the kids getting

474
00:28:39,600 --> 00:28:42,719
Speaker 1: the keys to the car more or less accurate. Actually,

475
00:28:43,200 --> 00:28:46,760
Speaker 1: I think the the the so I don't like the

476
00:28:46,800 --> 00:28:49,920
Speaker 1: E t F products mostly because they're very complicated. They

477
00:28:49,920 --> 00:28:53,160
Speaker 1: look simple on the outside, but underneath them they're very complicated,

478
00:28:53,400 --> 00:28:57,000
Speaker 1: and I don't think people always understand all that complexity.

479
00:28:57,080 --> 00:29:01,480
Speaker 1: So the events that happened in fabruy E two thousand

480
00:29:01,560 --> 00:29:06,680
Speaker 1: eighteen were these short vix et s and the short

481
00:29:06,760 --> 00:29:11,120
Speaker 1: vix ets trying to earn this insurance premium. So lots

482
00:29:11,160 --> 00:29:14,840
Speaker 1: of people, lots of all streets were happy owners of

483
00:29:15,000 --> 00:29:19,280
Speaker 1: these short vix ets. The problem with them is that

484
00:29:19,800 --> 00:29:22,760
Speaker 1: what the t F does then, is it it issues

485
00:29:22,880 --> 00:29:25,800
Speaker 1: units to people like you or I UM and then

486
00:29:25,840 --> 00:29:28,960
Speaker 1: has to sell futures contracts against it. If the price

487
00:29:29,000 --> 00:29:31,680
Speaker 1: of the vix starts going up, it needs to start

488
00:29:31,720 --> 00:29:34,680
Speaker 1: buying those contracts back, and if it goes up a lot,

489
00:29:34,720 --> 00:29:36,400
Speaker 1: it needs to buy a heck of a lot of them.

490
00:29:36,680 --> 00:29:39,880
Speaker 1: And so you have some volatility very late in the closing,

491
00:29:40,320 --> 00:29:42,880
Speaker 1: towards the end of the day UM, and I think

492
00:29:42,880 --> 00:29:45,160
Speaker 1: it was February five, if I remember rightly, two thousand

493
00:29:45,240 --> 00:29:48,000
Speaker 1: eighteen and UM, and they needed to buy a heck

494
00:29:48,000 --> 00:29:49,760
Speaker 1: of a lot of futures contracts in a very short

495
00:29:49,760 --> 00:29:51,840
Speaker 1: period of time. What did that do It push the

496
00:29:51,880 --> 00:29:54,360
Speaker 1: thing up even more, and so it kind of created

497
00:29:54,400 --> 00:29:59,280
Speaker 1: its own volatility and its own noise UM in that period.

498
00:29:59,640 --> 00:30:03,000
Speaker 1: So I think that it's a pretty good example of

499
00:30:03,040 --> 00:30:08,000
Speaker 1: an unintended consequence from a financial product. And again, you know,

500
00:30:08,080 --> 00:30:14,280
Speaker 1: it's quite a complicated concept. Along vix CTF is complicated

501
00:30:14,320 --> 00:30:17,480
Speaker 1: as short vic CTF is very complicated, and I think

502
00:30:17,480 --> 00:30:21,479
Speaker 1: from that perspective and people probably had some surprising results

503
00:30:21,480 --> 00:30:22,880
Speaker 1: that you know, a lot of people lost a lot

504
00:30:22,920 --> 00:30:26,440
Speaker 1: of money, some of the issues of these short vix

505
00:30:26,480 --> 00:30:29,560
Speaker 1: cts made a lot of money UM at the same time,

506
00:30:29,600 --> 00:30:31,960
Speaker 1: and it felt like a pretty bad state of affairs

507
00:30:32,040 --> 00:30:35,480
Speaker 1: to me. Let's talk a little bit about your book, UM,

508
00:30:35,520 --> 00:30:38,360
Speaker 1: which was written with Harvey Campbell, who was a prior

509
00:30:38,440 --> 00:30:43,760
Speaker 1: guest and and just a delightful individual. What compelled you

510
00:30:43,760 --> 00:30:48,320
Speaker 1: guys to write this book? This is a pretty UM

511
00:30:48,360 --> 00:30:52,480
Speaker 1: in the Weeds Inside Baseball sort of sort of book.

512
00:30:53,200 --> 00:30:56,040
Speaker 1: Yeah you're you're right, it's absolutely in the Weeds type

513
00:30:56,080 --> 00:31:00,720
Speaker 1: of book. What got us going on this was really

514
00:31:00,720 --> 00:31:05,080
Speaker 1: a sense that for many portfolio managers, risk management is

515
00:31:05,120 --> 00:31:08,520
Speaker 1: something which comes afterwards they build their portfolio, and then

516
00:31:08,560 --> 00:31:11,840
Speaker 1: the risk team do something later on and tell them

517
00:31:11,840 --> 00:31:14,400
Speaker 1: whether it's okay or not. And we thought that that's

518
00:31:14,440 --> 00:31:16,880
Speaker 1: actually a very bad way to run portfolios. In a

519
00:31:16,960 --> 00:31:20,560
Speaker 1: much better way is to have the alpha side of

520
00:31:20,600 --> 00:31:24,000
Speaker 1: building portfolios in the risk side to be equal partners.

521
00:31:24,040 --> 00:31:26,440
Speaker 1: And that's something that we've really tried to build as

522
00:31:26,480 --> 00:31:29,640
Speaker 1: a culture of a man group that risk is part

523
00:31:29,680 --> 00:31:32,480
Speaker 1: of the investment team. It's, as they sometimes put it,

524
00:31:32,760 --> 00:31:35,040
Speaker 1: if risk is the police for other people that come

525
00:31:35,160 --> 00:31:37,440
Speaker 1: kind of knocking on your door telling you've done something wrong.

526
00:31:37,880 --> 00:31:41,000
Speaker 1: That's that's not really a good way of running portfolios.

527
00:31:41,240 --> 00:31:44,040
Speaker 1: What do you really want is as you're building the portfolios,

528
00:31:44,080 --> 00:31:47,920
Speaker 1: the risk and the alphabets to to come as as

529
00:31:47,960 --> 00:31:53,520
Speaker 1: equal partners. And the reason for that is, firstly, huge

530
00:31:53,520 --> 00:31:56,040
Speaker 1: amounts of damage tend to be done when there is

531
00:31:56,080 --> 00:31:58,840
Speaker 1: bad risk management in stress periods that people aren't prepared

532
00:31:58,880 --> 00:32:01,880
Speaker 1: for those stress periods, they make bad decisions, and those

533
00:32:01,880 --> 00:32:06,680
Speaker 1: stress periods lose a lot of money, often crystallize losses,

534
00:32:06,800 --> 00:32:09,520
Speaker 1: that sort of thing. So so firstly, if you don't

535
00:32:09,600 --> 00:32:13,720
Speaker 1: have a proper risk approach to building portfolios when you

536
00:32:14,120 --> 00:32:18,560
Speaker 1: enter chop your periods in markets, then you're likely to

537
00:32:18,600 --> 00:32:22,719
Speaker 1: make bad decisions. That The second, which I mentioned earlier

538
00:32:22,760 --> 00:32:26,880
Speaker 1: on is kind of surprisingly. It's actually much easier to

539
00:32:27,000 --> 00:32:31,440
Speaker 1: forecast risk than it is to forecast returns, and you

540
00:32:31,440 --> 00:32:35,800
Speaker 1: can use that to give yourself more stable portfolios. And

541
00:32:35,840 --> 00:32:39,040
Speaker 1: so we were really trying to say that you can

542
00:32:39,200 --> 00:32:43,760
Speaker 1: blend the alpha side of portfolio management with the risk

543
00:32:43,800 --> 00:32:46,280
Speaker 1: side as equal partners, and you'll build better and more

544
00:32:46,280 --> 00:32:49,520
Speaker 1: stable portfolios which will actually do better over the long

545
00:32:49,640 --> 00:32:52,440
Speaker 1: run because you won't end up making bad decisions during

546
00:32:52,480 --> 00:32:55,640
Speaker 1: the during the stress periods. So let's talk about that

547
00:32:55,720 --> 00:33:00,560
Speaker 1: last quote, it's easier to forecast risk then return. I'm

548
00:33:00,560 --> 00:33:02,880
Speaker 1: going to play devil's advocate and take the other side

549
00:33:02,880 --> 00:33:06,120
Speaker 1: of that argument. Hey, we know over long periods of

550
00:33:06,200 --> 00:33:11,160
Speaker 1: time what historical asset class returns are, and so we

551
00:33:11,240 --> 00:33:17,480
Speaker 1: can reasonably forecast eight percent over twenty years for equities UM,

552
00:33:17,520 --> 00:33:21,800
Speaker 1: but we can forecast the sort of UM black swans

553
00:33:21,800 --> 00:33:23,600
Speaker 1: that show up every now and then that are just

554
00:33:23,720 --> 00:33:31,400
Speaker 1: completely unexpected and are amongst the quote unknown unknowns. Tell

555
00:33:31,440 --> 00:33:35,520
Speaker 1: me what's wrong with that perspective that looks at extrapolating

556
00:33:35,560 --> 00:33:38,600
Speaker 1: long term returns versus hey, we have no idea what

557
00:33:38,680 --> 00:33:40,840
Speaker 1: the next random events is going to be. Yeah, I

558
00:33:40,840 --> 00:33:43,400
Speaker 1: think what's wrong with that is that you've tried to

559
00:33:43,400 --> 00:33:46,720
Speaker 1: compare forecasting very long run returns and then said that

560
00:33:46,760 --> 00:33:49,840
Speaker 1: I need to forecast short term risk. And so let

561
00:33:49,840 --> 00:33:53,360
Speaker 1: me sort of decompose that a little bit more. Many times,

562
00:33:53,480 --> 00:33:57,160
Speaker 1: people when they when they forecast returns, and we're all

563
00:33:57,160 --> 00:34:00,239
Speaker 1: guilty of this, we end up being pretty influence by

564
00:34:00,280 --> 00:34:03,360
Speaker 1: what happened in the last month or two, and so

565
00:34:03,400 --> 00:34:06,000
Speaker 1: we say, you know, people, many people, for example, are

566
00:34:06,000 --> 00:34:11,240
Speaker 1: pretty positive about equity market returns globally and maybe especially

567
00:34:11,239 --> 00:34:13,920
Speaker 1: in the US. And one of the reasons that really

568
00:34:14,040 --> 00:34:17,279
Speaker 1: is that we've had good returns, you know, really since

569
00:34:17,320 --> 00:34:20,800
Speaker 1: March shows since April last year, and people are extrapolating,

570
00:34:20,800 --> 00:34:23,640
Speaker 1: they're just stretching forwards. But if you look at the data,

571
00:34:23,920 --> 00:34:25,800
Speaker 1: and if you look at the data over long periods

572
00:34:25,800 --> 00:34:29,640
Speaker 1: of time, you find the correlation between past returns and

573
00:34:29,800 --> 00:34:33,920
Speaker 1: next month's returns is about zero in almost all markets

574
00:34:33,960 --> 00:34:37,360
Speaker 1: around the world, equity markets, bond markets, commodity markets just

575
00:34:37,400 --> 00:34:40,840
Speaker 1: about zero. There has no last month's returns have close

576
00:34:40,920 --> 00:34:45,040
Speaker 1: to zero predictive ability of telling you what next month's

577
00:34:45,040 --> 00:34:46,960
Speaker 1: returns are going to be. If you now do that

578
00:34:47,000 --> 00:34:50,680
Speaker 1: on risk and you calculate the fertility of markets last

579
00:34:50,719 --> 00:34:52,759
Speaker 1: month or the month before or the month before that,

580
00:34:53,000 --> 00:34:56,239
Speaker 1: lets hee has very high predictive ability. So people like

581
00:34:56,320 --> 00:34:59,440
Speaker 1: me would call this the serial correlation. So the serial

582
00:34:59,480 --> 00:35:02,920
Speaker 1: correlation the returns in other words, from last month or

583
00:35:02,960 --> 00:35:05,480
Speaker 1: two months ago or three months ago to this month's

584
00:35:05,480 --> 00:35:08,839
Speaker 1: returns is close to zero, you might as well call

585
00:35:08,880 --> 00:35:10,920
Speaker 1: it zero. So close to zero. If you do the

586
00:35:10,960 --> 00:35:14,120
Speaker 1: same thing in volatility, and you can do this across

587
00:35:14,200 --> 00:35:17,279
Speaker 1: equity markets in the US, but also across other parts

588
00:35:17,320 --> 00:35:21,360
Speaker 1: of the world, bond markets, commodity markets, currency markets, that

589
00:35:22,040 --> 00:35:25,040
Speaker 1: zerial correlation is around forty percent. So that's telling you

590
00:35:25,040 --> 00:35:27,080
Speaker 1: that last month's vulatility is actually telling you quite a

591
00:35:27,120 --> 00:35:30,759
Speaker 1: lot about this month's faultiality. And so whilst you might

592
00:35:30,800 --> 00:35:32,680
Speaker 1: be able to make a statement about you know, twenty

593
00:35:32,800 --> 00:35:36,719
Speaker 1: year expected returns um I suspect both you and I

594
00:35:36,800 --> 00:35:38,520
Speaker 1: will be, you know, at least twenty years older, and

595
00:35:38,600 --> 00:35:41,279
Speaker 1: that your point and whether anybody will really hold us

596
00:35:41,280 --> 00:35:45,239
Speaker 1: to it or whether it's the useful observation will be

597
00:35:45,719 --> 00:35:49,120
Speaker 1: tricky to to to really evaluate. But for most people

598
00:35:49,160 --> 00:35:52,000
Speaker 1: they need to have nearer term forecasts in order to

599
00:35:52,000 --> 00:35:53,799
Speaker 1: be able to make their decisions. And I think they

600
00:35:53,800 --> 00:35:57,520
Speaker 1: make a mistake that they think they can forecast returns

601
00:35:57,560 --> 00:36:00,680
Speaker 1: often by extrapolating from past months or turns, And the

602
00:36:00,719 --> 00:36:04,360
Speaker 1: evidence is that you shouldn't extractly. There's no extrapolation to

603
00:36:04,440 --> 00:36:08,400
Speaker 1: be done. Whilst a risk you can that that is

604
00:36:08,520 --> 00:36:10,840
Speaker 1: very parallel to what we were talking about earlier with

605
00:36:10,960 --> 00:36:15,160
Speaker 1: price versus time series. So equities follow a random walk,

606
00:36:15,719 --> 00:36:20,360
Speaker 1: but volatility and risk tends to be persistent and enhance

607
00:36:20,440 --> 00:36:25,800
Speaker 1: more likely to have some time series correlation. I think

608
00:36:26,000 --> 00:36:30,560
Speaker 1: that's what I'm hearing, Am I am I saying that right? Yeah, absolutely, absolutely,

609
00:36:30,600 --> 00:36:34,080
Speaker 1: the the there's a very small and what the c

610
00:36:34,280 --> 00:36:37,239
Speaker 1: t a s try and do, but the futures trend

611
00:36:37,320 --> 00:36:41,080
Speaker 1: followers is that there's there's a very small effect to

612
00:36:41,160 --> 00:36:44,680
Speaker 1: being able to pick up some persistence in returns, but

613
00:36:44,719 --> 00:36:46,840
Speaker 1: you have to do it across hundreds of markets and

614
00:36:46,920 --> 00:36:49,560
Speaker 1: you have to do it consistently over time, and you'll

615
00:36:49,640 --> 00:36:51,920
Speaker 1: just managed to eke out a little bit of alpha

616
00:36:51,920 --> 00:36:55,520
Speaker 1: about doing that, but altility is much easier. So what

617
00:36:55,680 --> 00:37:00,680
Speaker 1: should managers do proactively to prepare for are not the

618
00:37:00,800 --> 00:37:04,040
Speaker 1: known risk but the unknown risks? And and really just

619
00:37:04,160 --> 00:37:07,239
Speaker 1: over the past twenty years, we had nine eleven, we

620
00:37:07,320 --> 00:37:11,920
Speaker 1: had the Great Financial Crisis, we had the VICS meltdown,

621
00:37:11,960 --> 00:37:17,000
Speaker 1: and more recently we had the COVID pandemic. How how

622
00:37:17,080 --> 00:37:23,200
Speaker 1: can a fund manager build protection against these black swans

623
00:37:23,239 --> 00:37:28,000
Speaker 1: into their process? Well, so I think the first thing

624
00:37:28,120 --> 00:37:30,520
Speaker 1: that I portfolio managed to do is realize that you

625
00:37:30,600 --> 00:37:34,000
Speaker 1: cannot forecast these events. And you know, I speak, at

626
00:37:34,040 --> 00:37:35,879
Speaker 1: least I used to speak at conferences a lot when

627
00:37:35,920 --> 00:37:40,080
Speaker 1: we've still had conferences, and you know, people would ask me, well,

628
00:37:40,120 --> 00:37:42,879
Speaker 1: what's the black swan event that's going to happen this year?

629
00:37:43,520 --> 00:37:47,240
Speaker 1: And I also that was the most ridiculous question because

630
00:37:48,040 --> 00:37:52,120
Speaker 1: you know, clearly all of these events are unfoecastable, and

631
00:37:52,280 --> 00:37:55,319
Speaker 1: generally the ones that you forecast will happen don't end

632
00:37:55,400 --> 00:37:58,000
Speaker 1: up being the thing that takes place. So I think

633
00:37:58,040 --> 00:37:59,880
Speaker 1: the first thing is to show a lot of human

634
00:38:00,080 --> 00:38:04,080
Speaker 1: tear about our ability to forecast what what the bad

635
00:38:04,080 --> 00:38:06,240
Speaker 1: events will be. The only thing I can really feel

636
00:38:06,239 --> 00:38:08,400
Speaker 1: confident about is that there will be more bad events

637
00:38:08,440 --> 00:38:10,360
Speaker 1: in the future. You know, it seems that they just

638
00:38:10,440 --> 00:38:15,600
Speaker 1: keep coming, but they have different shapes and forms. And

639
00:38:15,680 --> 00:38:17,839
Speaker 1: maybe as a side anecdote on that, we have an

640
00:38:17,880 --> 00:38:20,160
Speaker 1: excellent risk manager at Man Group and at the end

641
00:38:20,200 --> 00:38:23,120
Speaker 1: of two thousand nineteen and a sort of planning exercise

642
00:38:23,160 --> 00:38:25,200
Speaker 1: who was giving all the risk that could affect markets.

643
00:38:25,239 --> 00:38:27,759
Speaker 1: And you had about twenty of these things, and one

644
00:38:27,800 --> 00:38:30,279
Speaker 1: of them was epidemic. And I looked at this at

645
00:38:30,320 --> 00:38:32,880
Speaker 1: the end of two thousand nineteen and I said, epidemic. Well,

646
00:38:32,920 --> 00:38:35,520
Speaker 1: I mean, like, you know, I don't know much about epidemics,

647
00:38:35,520 --> 00:38:38,000
Speaker 1: but I can't say it's impossible. But it doesn't seem

648
00:38:38,080 --> 00:38:40,279
Speaker 1: very lightly. So we did exactly nothing about the risk

649
00:38:40,320 --> 00:38:43,200
Speaker 1: of epidemic. And if he got the word almost right,

650
00:38:43,400 --> 00:38:46,480
Speaker 1: pandemic instead of epidemic. But you know, even if you

651
00:38:46,520 --> 00:38:48,600
Speaker 1: have it on your list of things, which my risk

652
00:38:48,680 --> 00:38:52,080
Speaker 1: manager did, um, it's very hard to act on it.

653
00:38:52,160 --> 00:38:55,600
Speaker 1: So I think humility in terms of ability to forecast

654
00:38:55,640 --> 00:38:58,600
Speaker 1: these things really important. If you think you can forecast,

655
00:38:58,920 --> 00:39:01,840
Speaker 1: you'll probably make them. State by protecting yourselves against the

656
00:39:01,880 --> 00:39:06,000
Speaker 1: wrong thing. Once you've got over that and said likely

657
00:39:06,000 --> 00:39:08,840
Speaker 1: it probably can't forecast the next bad thing, then I

658
00:39:08,840 --> 00:39:11,560
Speaker 1: think what becomes much more important is, Okay, now you

659
00:39:11,600 --> 00:39:15,920
Speaker 1: need a strategy that is going to be relatively insensitive

660
00:39:16,080 --> 00:39:20,320
Speaker 1: to the nature of the bad thing. In other words,

661
00:39:21,040 --> 00:39:26,759
Speaker 1: whether it's a tech bubble collapse or um credit crisis,

662
00:39:27,120 --> 00:39:31,400
Speaker 1: or you know, something entirely different to war, or a pandemic,

663
00:39:31,560 --> 00:39:33,560
Speaker 1: or it could be any of these things. Obviously you

664
00:39:33,600 --> 00:39:36,680
Speaker 1: need a strategy which is going to be robust to

665
00:39:36,840 --> 00:39:40,760
Speaker 1: any of those things coming along. And my own suggestion

666
00:39:40,800 --> 00:39:44,279
Speaker 1: on this would be that it's it's too expensive to

667
00:39:44,320 --> 00:39:48,799
Speaker 1: buy put options. Buying put options on the sp you

668
00:39:48,840 --> 00:39:50,680
Speaker 1: can do every now and then, but you can't do

669
00:39:50,719 --> 00:39:53,960
Speaker 1: it all the time. It's just it just becomes too costly.

670
00:39:54,400 --> 00:39:56,200
Speaker 1: And so you're gonna have to have a strategy which

671
00:39:56,239 --> 00:40:00,759
Speaker 1: relies a little bit more on either assets in your

672
00:40:00,760 --> 00:40:03,239
Speaker 1: portfolio which you think are likely to do well in

673
00:40:03,280 --> 00:40:06,200
Speaker 1: a stress period. That could be gold in my view,

674
00:40:06,239 --> 00:40:09,400
Speaker 1: it's not terribly reliable. It could be gold. It could

675
00:40:09,400 --> 00:40:12,400
Speaker 1: be U strategury bonds or other government bonds around the world.

676
00:40:12,719 --> 00:40:15,600
Speaker 1: Of course, if the problem emerges from the bond market,

677
00:40:15,640 --> 00:40:19,120
Speaker 1: it's not really going to help you. Um. And you know, people,

678
00:40:19,560 --> 00:40:22,040
Speaker 1: I think forget that they have definitely been problems from

679
00:40:22,040 --> 00:40:23,640
Speaker 1: the bond markets. You just need to look back a

680
00:40:23,719 --> 00:40:28,400
Speaker 1: little bit to the early es or before then to

681
00:40:28,480 --> 00:40:30,480
Speaker 1: see that actually there were plenty of problems that came

682
00:40:30,600 --> 00:40:33,239
Speaker 1: from the bond market. And or you could have some

683
00:40:33,320 --> 00:40:38,319
Speaker 1: sort of trading strategy, and certainly for me this time

684
00:40:38,360 --> 00:40:41,240
Speaker 1: series momentum, which we touched on in our first segment,

685
00:40:41,800 --> 00:40:46,759
Speaker 1: this idea that you can build a trading strategy like

686
00:40:46,800 --> 00:40:50,040
Speaker 1: the c t as have done, which relies on just

687
00:40:50,120 --> 00:40:53,960
Speaker 1: a little bit of persistence in returns and looks for

688
00:40:54,000 --> 00:40:56,360
Speaker 1: them everywhere. That can be a very good way of

689
00:40:56,400 --> 00:41:00,239
Speaker 1: building a defensive strategy. So, in other words, if we

690
00:41:00,360 --> 00:41:02,520
Speaker 1: describe a crisis, and I can't tell you whether it

691
00:41:02,560 --> 00:41:07,880
Speaker 1: comes from a war, credit crisis, an epidemic, something else,

692
00:41:08,239 --> 00:41:11,240
Speaker 1: but typically in a crisis like that, you will tend

693
00:41:11,280 --> 00:41:13,560
Speaker 1: to have equities going down, you will tend to have

694
00:41:13,640 --> 00:41:17,280
Speaker 1: bonds going up, you'll tend to have gold going up. Um.

695
00:41:17,320 --> 00:41:19,200
Speaker 1: It's a bit harder to tell what might happen to

696
00:41:19,360 --> 00:41:22,319
Speaker 1: energy prices, but those moves tend to persist for a bit.

697
00:41:22,480 --> 00:41:25,240
Speaker 1: They tend to know equities fall and then they keep falling,

698
00:41:25,280 --> 00:41:27,480
Speaker 1: and bonds might go up and they keep rising, and

699
00:41:27,600 --> 00:41:30,600
Speaker 1: gold the same thing. And you can build a strategy

700
00:41:30,640 --> 00:41:34,160
Speaker 1: which encapsulates and tries to capture that effect. Then you

701
00:41:34,160 --> 00:41:37,480
Speaker 1: can build something which is robust and is not depending

702
00:41:37,600 --> 00:41:39,719
Speaker 1: on your ability to put your finger on what the

703
00:41:39,760 --> 00:41:42,880
Speaker 1: next bad event might be. So in the book, you

704
00:41:42,880 --> 00:41:45,479
Speaker 1: you get into the nitty gritty, you you go over

705
00:41:46,000 --> 00:41:50,640
Speaker 1: details of a new risk management approach to portfolio design.

706
00:41:51,480 --> 00:41:55,640
Speaker 1: I want to ask, how did those strategies do in

707
00:41:55,960 --> 00:41:58,839
Speaker 1: back tests looking at OH eight or nine, and how

708
00:41:58,880 --> 00:42:04,879
Speaker 1: do they do in the real war world in March. Well,

709
00:42:04,880 --> 00:42:07,319
Speaker 1: so for us, actually, I don't think O eight O

710
00:42:07,440 --> 00:42:10,840
Speaker 1: nine really was back test because we were actually trading

711
00:42:10,880 --> 00:42:14,640
Speaker 1: most of these strategies at that time. But so I

712
00:42:14,719 --> 00:42:20,360
Speaker 1: think we actually have pretty good life experience. And the

713
00:42:20,400 --> 00:42:26,720
Speaker 1: first thing I should say, March of was an extraordinary crisis,

714
00:42:26,760 --> 00:42:30,440
Speaker 1: and all all crises are extraordinary, but one of the

715
00:42:30,480 --> 00:42:35,279
Speaker 1: things which was most extraordinary about March was that it

716
00:42:36,120 --> 00:42:38,799
Speaker 1: markets fell very quickly. What we've seen that before, but

717
00:42:38,880 --> 00:42:42,520
Speaker 1: then they reverted remarkably quickly, and really the most similar

718
00:42:42,640 --> 00:42:46,799
Speaker 1: crisis in terms of market action that you can put

719
00:42:46,800 --> 00:42:50,120
Speaker 1: your finger on since the Second World War was the

720
00:42:50,160 --> 00:42:55,080
Speaker 1: October seven crisis, so that that very rapid fall you

721
00:42:55,160 --> 00:43:00,160
Speaker 1: had followed by an almost equally rapid recovery. So so

722
00:43:00,480 --> 00:43:03,680
Speaker 1: that might say, well, you know, if you fitted, if

723
00:43:03,680 --> 00:43:06,879
Speaker 1: you've sort of tested your crisis protection on all these

724
00:43:06,880 --> 00:43:10,759
Speaker 1: slower crises, then maybe you wouldn't do too well in

725
00:43:10,840 --> 00:43:13,080
Speaker 1: this faster crisis. And it's actually not what we found.

726
00:43:13,080 --> 00:43:16,120
Speaker 1: So we found that, um, we had you know, quite

727
00:43:16,160 --> 00:43:22,840
Speaker 1: good strength of our strategies during um the March April period.

728
00:43:22,920 --> 00:43:27,239
Speaker 1: So for example, futures trend firing, something we talked about

729
00:43:27,280 --> 00:43:30,920
Speaker 1: quite a lot, um did you know, really rather well

730
00:43:31,120 --> 00:43:35,080
Speaker 1: in March and April of last year. But we also

731
00:43:35,120 --> 00:43:39,560
Speaker 1: talk about for example, rebouncing and and and trying to

732
00:43:39,600 --> 00:43:44,720
Speaker 1: stop rebouncing, which can can have the nasty effect of

733
00:43:44,920 --> 00:43:48,280
Speaker 1: buying the losers, and then if the losers carry on falling,

734
00:43:48,800 --> 00:43:51,000
Speaker 1: then you will, damn it, you just bought a whole

735
00:43:51,040 --> 00:43:53,880
Speaker 1: bunch of losers in time for another month's falls. And

736
00:43:53,920 --> 00:43:57,480
Speaker 1: we found that if you if you have UH strategies

737
00:43:57,560 --> 00:44:00,840
Speaker 1: which try and control your rebouncing, but they have to

738
00:44:00,840 --> 00:44:03,560
Speaker 1: have relatively rapid they have to be quite fast strategies,

739
00:44:03,600 --> 00:44:05,600
Speaker 1: and that would be my real point. So most of

740
00:44:05,600 --> 00:44:08,839
Speaker 1: our protection strategies are quite fast. The signals are quite quick.

741
00:44:08,880 --> 00:44:14,520
Speaker 1: They use data that goes back typically a few weeks um,

742
00:44:14,640 --> 00:44:18,440
Speaker 1: and they can move positions around quite rapidly. And that

743
00:44:18,480 --> 00:44:21,319
Speaker 1: worked pretty well in March and April last year. If

744
00:44:21,320 --> 00:44:23,680
Speaker 1: you do much slower strategies, you would not have had

745
00:44:23,680 --> 00:44:27,560
Speaker 1: the protection characteristics right and and and to put some

746
00:44:27,680 --> 00:44:35,439
Speaker 1: numbers on on the speed of March, we smp um.

747
00:44:35,480 --> 00:44:39,520
Speaker 1: That was the fastest drop in history, and I believe

748
00:44:40,200 --> 00:44:44,080
Speaker 1: it was just a day under a month, maybe a

749
00:44:44,080 --> 00:44:47,439
Speaker 1: few days under a month. And then the recovery from

750
00:44:47,640 --> 00:44:50,480
Speaker 1: the end of March beginning of April was back to

751
00:44:50,480 --> 00:44:54,759
Speaker 1: break even by August. That that's a pretty astounding turnaround,

752
00:44:55,640 --> 00:45:04,240
Speaker 1: arguably faster than the recovery from seven, which was itself

753
00:45:04,280 --> 00:45:09,719
Speaker 1: pretty quick, wasn't it Absolutely so, it was just totally extraordinary.

754
00:45:09,760 --> 00:45:12,520
Speaker 1: And from that perspective, you know, the past didn't give

755
00:45:12,520 --> 00:45:15,480
Speaker 1: you a particularly good guide as to how how that

756
00:45:15,520 --> 00:45:19,040
Speaker 1: crisis would would unfold. And then maybe that sort of

757
00:45:19,040 --> 00:45:22,759
Speaker 1: retrates my point a little bit that you can't you

758
00:45:22,880 --> 00:45:25,839
Speaker 1: can't build protection strategies which are really trying to put

759
00:45:25,880 --> 00:45:28,279
Speaker 1: your finger on exactly what's going to happen. You have

760
00:45:28,360 --> 00:45:31,800
Speaker 1: to you have to be aware that your forecasting ability

761
00:45:31,960 --> 00:45:34,520
Speaker 1: is poor, UM, and you've really got to have a

762
00:45:34,920 --> 00:45:37,440
Speaker 1: strategic response. And that's why we call the book strategic

763
00:45:37,480 --> 00:45:40,280
Speaker 1: risk Management. It's really a set of strategies that the plan.

764
00:45:41,000 --> 00:45:42,799
Speaker 1: And you can't make the plan up on the fly.

765
00:45:43,239 --> 00:45:45,080
Speaker 1: You know what. You really the worst thing you could

766
00:45:45,080 --> 00:45:47,800
Speaker 1: have been doing last year is making up your protection

767
00:45:47,840 --> 00:45:51,320
Speaker 1: strategy during March. It was too late by that point.

768
00:45:51,480 --> 00:45:53,880
Speaker 1: You have to make up your your protection strategy in

769
00:45:53,920 --> 00:45:56,200
Speaker 1: the months and years before then, and then you had

770
00:45:56,200 --> 00:45:59,160
Speaker 1: to be implementing it during March. So there was a

771
00:45:59,239 --> 00:46:01,600
Speaker 1: quote of yours really liked, and I want to ask

772
00:46:01,640 --> 00:46:05,600
Speaker 1: you about this quote. We are in a riskier environment

773
00:46:05,680 --> 00:46:09,759
Speaker 1: than we have been in the past twenty years for

774
00:46:09,880 --> 00:46:15,239
Speaker 1: the foreseeable future. Unquote the past twenty years. Really there

775
00:46:15,280 --> 00:46:18,399
Speaker 1: were a lot of risky events that took place, from

776
00:46:18,480 --> 00:46:21,480
Speaker 1: nine eleven to the Great Financial Crisis to the pandemic.

777
00:46:22,000 --> 00:46:26,000
Speaker 1: What makes this a riskier environment and why do you

778
00:46:26,080 --> 00:46:30,440
Speaker 1: see this as um being a persistent risk for the

779
00:46:30,680 --> 00:46:35,279
Speaker 1: for the next foreseeable future. So the reason that I

780
00:46:35,320 --> 00:46:39,200
Speaker 1: think we're in this highly or much riskier environment that

781
00:46:39,280 --> 00:46:43,359
Speaker 1: we've been in is because markets are much less diversified

782
00:46:43,400 --> 00:46:46,759
Speaker 1: than at any point in my career. And so you know,

783
00:46:46,800 --> 00:46:50,040
Speaker 1: I started trading markets when I was at high school

784
00:46:50,080 --> 00:46:53,919
Speaker 1: in the late nineties, and at that time people got

785
00:46:54,000 --> 00:46:58,640
Speaker 1: very worried that the Japanese market was round of the

786
00:46:58,719 --> 00:47:02,560
Speaker 1: MSCI world. Well, today the US market is two thirds

787
00:47:02,560 --> 00:47:06,120
Speaker 1: of the MSCI world. So it's and that's the highest

788
00:47:06,120 --> 00:47:07,920
Speaker 1: way to to ever be in such the highest way

789
00:47:08,000 --> 00:47:11,680
Speaker 1: that any one country has ever been in the global

790
00:47:11,719 --> 00:47:14,640
Speaker 1: Equity Index. And then if you now dig in within

791
00:47:14,719 --> 00:47:17,719
Speaker 1: the US market, and this is a little tougher to do,

792
00:47:18,560 --> 00:47:22,160
Speaker 1: but if you dig into the proportion of the U

793
00:47:22,200 --> 00:47:24,640
Speaker 1: s ecty market made up by tech, and the reason

794
00:47:24,640 --> 00:47:27,640
Speaker 1: it's difficult is they changed the classification system a couple

795
00:47:27,640 --> 00:47:30,680
Speaker 1: of years ago, then you'll see the tech is a

796
00:47:30,719 --> 00:47:33,160
Speaker 1: bigger portion of the US equity market than it has

797
00:47:33,200 --> 00:47:37,440
Speaker 1: ever been, including in the late in the tech bubble.

798
00:47:37,680 --> 00:47:42,640
Speaker 1: So you have an incredibly concentrated equity market both globally

799
00:47:42,800 --> 00:47:46,680
Speaker 1: into the US and also by sector within the US.

800
00:47:46,760 --> 00:47:49,480
Speaker 1: And for me, that means that you know, this is

801
00:47:49,520 --> 00:47:52,240
Speaker 1: not sort of looking at the vix today, tomorrow, yesterday, whatever.

802
00:47:52,440 --> 00:47:57,360
Speaker 1: More strategically, the market feels much more likely to be

803
00:47:57,400 --> 00:48:03,200
Speaker 1: able to produce unpleasant outcomes because the only freelance you

804
00:48:03,200 --> 00:48:07,359
Speaker 1: haven't financed the diversification you have the least diversification I've

805
00:48:07,360 --> 00:48:11,680
Speaker 1: ever seen. Huh, that's kind of interesting. So we have concentrated,

806
00:48:12,120 --> 00:48:16,359
Speaker 1: non diversified portfolios. And one of the things we've seen

807
00:48:16,520 --> 00:48:21,239
Speaker 1: is domestically, the US seems to be a higher poor

808
00:48:21,360 --> 00:48:25,080
Speaker 1: proportion of global equity markets. And then within the US,

809
00:48:25,320 --> 00:48:28,840
Speaker 1: the text actor continues to increase its waiting when the

810
00:48:29,000 --> 00:48:33,680
Speaker 1: sp What does that mean for the future of of

811
00:48:33,840 --> 00:48:40,520
Speaker 1: risk and managing it? Well, I look, I think it

812
00:48:40,560 --> 00:48:42,560
Speaker 1: means firstly that you know, you need to be just

813
00:48:42,640 --> 00:48:47,680
Speaker 1: aware of this, that that the market is so heavily concentrated.

814
00:48:48,239 --> 00:48:52,480
Speaker 1: Um the I think the what can you do about it?

815
00:48:52,360 --> 00:48:55,840
Speaker 1: It is probably the real question. And I think that

816
00:48:55,920 --> 00:48:59,520
Speaker 1: this is a pretty big challenge for people because historically

817
00:48:59,680 --> 00:49:04,239
Speaker 1: the answer was, well, if if I want to build

818
00:49:04,280 --> 00:49:07,759
Speaker 1: a balanced portfolio, then I'll hold some equities, and then

819
00:49:07,880 --> 00:49:13,839
Speaker 1: I'll hold some government bonds, often US treasury bonds um

820
00:49:14,200 --> 00:49:16,319
Speaker 1: as the sort of as the ballast as the thing

821
00:49:16,400 --> 00:49:19,440
Speaker 1: which gives a bit of stability to my equity portfolio.

822
00:49:19,880 --> 00:49:22,879
Speaker 1: But where we are today, I think people are are

823
00:49:23,440 --> 00:49:27,400
Speaker 1: much less convinced that treasury bonds will be the ballast

824
00:49:28,120 --> 00:49:31,359
Speaker 1: that they have been historically. In particular, you know, if

825
00:49:31,360 --> 00:49:35,759
Speaker 1: we continue to get high inflation numbers, then I don't

826
00:49:35,760 --> 00:49:38,200
Speaker 1: think anybody is good to argue that high inflation is

827
00:49:38,239 --> 00:49:41,680
Speaker 1: good for government bonds. It's is clearly bad for government bonds.

828
00:49:42,440 --> 00:49:46,080
Speaker 1: And so your challenge is that the way you built

829
00:49:46,120 --> 00:49:49,520
Speaker 1: stable portfolios in the past, there's balancing of equities and bonds,

830
00:49:50,560 --> 00:49:55,000
Speaker 1: is really much less suited to the current environment than

831
00:49:55,040 --> 00:49:57,640
Speaker 1: it was to the to the past environment. So what

832
00:49:57,680 --> 00:49:59,840
Speaker 1: can you do about it? But I think what most

833
00:50:00,000 --> 00:50:02,440
Speaker 1: people that I speak with at least to think of

834
00:50:02,520 --> 00:50:05,680
Speaker 1: doing about it is saying, well, I need to own

835
00:50:05,880 --> 00:50:10,840
Speaker 1: something other than treasury bonds to balance out my equity risk.

836
00:50:11,480 --> 00:50:15,560
Speaker 1: And for some people that's private equity, for some people

837
00:50:16,680 --> 00:50:20,680
Speaker 1: that's hedge funds or alternatives. For some people it's infrastructure

838
00:50:20,960 --> 00:50:24,759
Speaker 1: or housing or other forms of real estate. But I

839
00:50:24,800 --> 00:50:27,200
Speaker 1: think it's reasonably clear in my mind that you need

840
00:50:27,239 --> 00:50:30,239
Speaker 1: to you need to think about balancing your portfolio, and

841
00:50:30,280 --> 00:50:33,400
Speaker 1: then you need to think pretty carefully about whether bonds

842
00:50:33,719 --> 00:50:37,560
Speaker 1: give you the same level of protection or balancing them

843
00:50:37,600 --> 00:50:40,200
Speaker 1: may then they would have done in the last twenty

844
00:50:40,200 --> 00:50:44,120
Speaker 1: thirty years. From my own perspective, if I look back

845
00:50:44,160 --> 00:50:48,480
Speaker 1: at the last twenty years in particular, it's a very

846
00:50:48,600 --> 00:50:51,480
Speaker 1: dangerous period to look at when you look at equities

847
00:50:51,480 --> 00:50:54,839
Speaker 1: and bonds. Over the last twenty years, when equities went down,

848
00:50:55,360 --> 00:50:58,640
Speaker 1: bonds nearly always went up in price, and and so

849
00:50:58,719 --> 00:51:00,840
Speaker 1: we've got used to this idea of bonds being the

850
00:51:01,000 --> 00:51:04,680
Speaker 1: protecting asset. But if you look before then, and you

851
00:51:04,719 --> 00:51:07,560
Speaker 1: can look back for in hundreds of years worth of data,

852
00:51:08,320 --> 00:51:10,920
Speaker 1: both in the US and then also the UK, whether

853
00:51:11,040 --> 00:51:14,600
Speaker 1: bond market started earlier than the US market, you see

854
00:51:14,600 --> 00:51:17,359
Speaker 1: that for almost all of history, except for the last

855
00:51:17,360 --> 00:51:20,520
Speaker 1: twenty years. When equities went down, bonds went down at

856
00:51:20,560 --> 00:51:23,879
Speaker 1: the same time. And so for me, I think it's

857
00:51:23,920 --> 00:51:27,200
Speaker 1: a very important question for investors, which is you need

858
00:51:27,239 --> 00:51:30,080
Speaker 1: to balance the risk in your portfolio. Are bonds the

859
00:51:30,120 --> 00:51:32,879
Speaker 1: answer to it? In my view, they're probably much less

860
00:51:32,920 --> 00:51:35,160
Speaker 1: the answer than they were historically, So then you need

861
00:51:35,239 --> 00:51:37,799
Speaker 1: to look at other asset classes and think more creatively

862
00:51:37,800 --> 00:51:40,000
Speaker 1: about how you do that. And the fact that I

863
00:51:40,000 --> 00:51:43,600
Speaker 1: think that equities public equities are as risky as they've

864
00:51:43,600 --> 00:51:46,800
Speaker 1: ever been from a strategic perspective, means that question is

865
00:51:46,800 --> 00:51:50,200
Speaker 1: actually as important as it's ever been to think about. Huh,

866
00:51:50,400 --> 00:51:54,280
Speaker 1: that's that's really kind of intriguing. Um. So, So, sticking

867
00:51:54,280 --> 00:51:58,160
Speaker 1: with the theme of risk, what does this concentration mean

868
00:51:58,640 --> 00:52:01,799
Speaker 1: and this lack of diversific asian mean in terms of,

869
00:52:02,480 --> 00:52:05,400
Speaker 1: you know, calibrating what we should be expecting. Should we

870
00:52:05,480 --> 00:52:09,759
Speaker 1: be looking for larger moves in the future, or you know,

871
00:52:10,080 --> 00:52:14,440
Speaker 1: how do you approach this um philosophically when you're thinking

872
00:52:14,440 --> 00:52:18,399
Speaker 1: about what you want to do with your portfolios? Yeah,

873
00:52:18,560 --> 00:52:20,400
Speaker 1: I think that you know, right now or in a

874
00:52:20,480 --> 00:52:24,279
Speaker 1: relatively sort of quiet summer period, and so markets have

875
00:52:24,400 --> 00:52:29,239
Speaker 1: been relatively stable. But but I looking beyond just a

876
00:52:29,280 --> 00:52:34,320
Speaker 1: short term than this lack of diversification, markets definitely means

877
00:52:34,360 --> 00:52:37,680
Speaker 1: that we should expect to see bigger moves in both directions.

878
00:52:37,719 --> 00:52:40,640
Speaker 1: Just to be clear, both upwards and downwards. Um, it

879
00:52:40,719 --> 00:52:42,799
Speaker 1: does not mean that markets couldn't go up a whole

880
00:52:42,840 --> 00:52:45,680
Speaker 1: bunch more from here. It just means that they're lately

881
00:52:45,800 --> 00:52:50,080
Speaker 1: to be more roletile than we've been used to in

882
00:52:50,120 --> 00:52:53,000
Speaker 1: the last few years. And so I think it's really

883
00:52:53,520 --> 00:52:56,040
Speaker 1: having a plan of action and being prepared for how

884
00:52:56,080 --> 00:52:58,640
Speaker 1: you respond to that, because in the end, for most

885
00:52:58,640 --> 00:53:01,120
Speaker 1: of us, the big up moves, I mean, maybe we

886
00:53:01,200 --> 00:53:03,480
Speaker 1: were underinvested in we have a bit of regret about it,

887
00:53:03,480 --> 00:53:05,839
Speaker 1: but you don't have much pain from the big up moves.

888
00:53:05,880 --> 00:53:07,920
Speaker 1: It's the big down moves that that caused all the

889
00:53:07,960 --> 00:53:10,359
Speaker 1: pain and caused cause the bad decisions to be made.

890
00:53:10,600 --> 00:53:12,640
Speaker 1: So I think it's having a plan of action. And

891
00:53:12,680 --> 00:53:14,879
Speaker 1: I would argue that you know, here we are sitting

892
00:53:14,880 --> 00:53:19,040
Speaker 1: in a relatively comfortable moment in markets currently, and you know,

893
00:53:19,080 --> 00:53:22,239
Speaker 1: if we're fortunate, then the summer will be enjoyable for

894
00:53:22,280 --> 00:53:26,520
Speaker 1: all of us and not too not too noisy. That's

895
00:53:26,560 --> 00:53:28,719
Speaker 1: a great time to be thinking about your plans for

896
00:53:28,760 --> 00:53:31,279
Speaker 1: how you'd respond if there was a if there was

897
00:53:31,320 --> 00:53:37,000
Speaker 1: a negative altality event. Really really quite quite interesting. So

898
00:53:37,120 --> 00:53:40,120
Speaker 1: let's talk a little bit about some strategies you mentioned,

899
00:53:40,560 --> 00:53:44,000
Speaker 1: alternatives like private equity and hedge funds. What can one

900
00:53:44,080 --> 00:53:49,480
Speaker 1: do to hedge against the risk of increased volatility in

901
00:53:49,520 --> 00:53:54,480
Speaker 1: the future. Well, I think you know, you can do

902
00:53:54,560 --> 00:53:57,440
Speaker 1: direct hedges, but they tend to be very expensive. So

903
00:53:57,600 --> 00:54:00,520
Speaker 1: you could go and buy futures on tracts on the

904
00:54:00,600 --> 00:54:02,680
Speaker 1: VIX or something like that, but they will turn out

905
00:54:02,680 --> 00:54:05,080
Speaker 1: to be very costly few over time, as we've talked

906
00:54:05,120 --> 00:54:08,279
Speaker 1: about now earlier sections, So more likely what you need

907
00:54:08,360 --> 00:54:10,719
Speaker 1: to do is to think about assets that will just

908
00:54:10,840 --> 00:54:16,080
Speaker 1: behave differently um to equity markets. And as equity markets

909
00:54:16,080 --> 00:54:21,000
Speaker 1: have become more concentrated, especially into tech um it's uh

910
00:54:22,120 --> 00:54:25,480
Speaker 1: to be precise about that. It's tech and communication services

911
00:54:25,520 --> 00:54:29,239
Speaker 1: of the two classifications that people use today. So as

912
00:54:29,280 --> 00:54:32,399
Speaker 1: it's become more concentrated into that into those sectors, then

913
00:54:32,440 --> 00:54:34,680
Speaker 1: you need, I think to think about things which will

914
00:54:34,719 --> 00:54:39,000
Speaker 1: be not so affected by a negative price move in

915
00:54:39,040 --> 00:54:43,920
Speaker 1: those and I think that private equity in the end,

916
00:54:43,960 --> 00:54:46,359
Speaker 1: it has a lot of equity market exposure into it,

917
00:54:46,400 --> 00:54:49,680
Speaker 1: but you tend to see the price action more slowly.

918
00:54:50,239 --> 00:54:54,680
Speaker 1: But infrastructure that's you know, could behave very differently. Hedge funds,

919
00:54:54,719 --> 00:54:57,000
Speaker 1: if they're good, hedge funds should have lots of protection

920
00:54:57,040 --> 00:55:00,480
Speaker 1: strategies built in and lots of short holding as well

921
00:55:00,480 --> 00:55:05,000
Speaker 1: as long holdings, and so should be less sensitive. We

922
00:55:05,120 --> 00:55:06,920
Speaker 1: talked about c T A S a little bit, one

923
00:55:06,960 --> 00:55:09,080
Speaker 1: of the few strategies which has a right tail to

924
00:55:09,160 --> 00:55:11,640
Speaker 1: it run the left tail that could be part of

925
00:55:11,680 --> 00:55:16,680
Speaker 1: your list of strategies as well. I think the core

926
00:55:16,760 --> 00:55:20,120
Speaker 1: things from my perspective would be recognize that you're not

927
00:55:20,160 --> 00:55:22,960
Speaker 1: gonna build a forecast the next difficult events number one.

928
00:55:23,640 --> 00:55:26,919
Speaker 1: Number two, you can't forecast it, then don't tin all

929
00:55:27,000 --> 00:55:30,920
Speaker 1: your diversification on a single thing. Have a range of

930
00:55:31,000 --> 00:55:36,279
Speaker 1: protection strategies out there. And number three would we make

931
00:55:36,280 --> 00:55:39,120
Speaker 1: sure those protection strategies are not too expensive to run.

932
00:55:39,480 --> 00:55:41,719
Speaker 1: And that, of course is the disadvantage of buying put

933
00:55:41,719 --> 00:55:44,200
Speaker 1: options on the SMP or buying VIC futures that they're

934
00:55:44,280 --> 00:55:46,400
Speaker 1: very expensive to run. So you need something that you

935
00:55:46,400 --> 00:55:48,399
Speaker 1: can actually put up with for a period of time.

936
00:55:48,960 --> 00:55:51,000
Speaker 1: I often see people that go and buy those more

937
00:55:51,040 --> 00:55:53,200
Speaker 1: expensive strategies and they do it for six months or

938
00:55:53,239 --> 00:55:55,719
Speaker 1: twelve months or eighteen months, and then they give up.

939
00:55:56,200 --> 00:55:58,480
Speaker 1: And oftentimes they managed to give up just before the

940
00:55:58,480 --> 00:56:02,200
Speaker 1: next bad event happens. That's been a terrible outcome. Yeah.

941
00:56:02,239 --> 00:56:05,440
Speaker 1: With with earthquake insurance, it turns out that there's always

942
00:56:05,480 --> 00:56:08,800
Speaker 1: a spike right after an earthquake, which is the least

943
00:56:09,120 --> 00:56:12,160
Speaker 1: likely time for there to be another earthquake, and by

944
00:56:12,160 --> 00:56:16,440
Speaker 1: the time enough time elapses where risk has gone up dramatically.

945
00:56:16,920 --> 00:56:20,480
Speaker 1: People have forgotten about it and they let that um,

946
00:56:20,560 --> 00:56:24,560
Speaker 1: that risk lapse. Um, I wanna. I want to emphasize

947
00:56:24,560 --> 00:56:26,760
Speaker 1: something you said, And it comes back to that question

948
00:56:26,800 --> 00:56:30,080
Speaker 1: someone ask you at the conference. You're not really thinking

949
00:56:30,120 --> 00:56:34,279
Speaker 1: about the specifics of the potential risk. It almost doesn't

950
00:56:34,320 --> 00:56:36,439
Speaker 1: matter if it's a bond risk or an equity miss

951
00:56:36,520 --> 00:56:41,360
Speaker 1: risk or some geopolitical risk. It's hey, we can expect

952
00:56:41,360 --> 00:56:44,320
Speaker 1: these asset classes to go down, these st classes to

953
00:56:44,360 --> 00:56:47,320
Speaker 1: go up with a whole lot of increase in volatility.

954
00:56:47,680 --> 00:56:52,520
Speaker 1: The black swan matters less than your preparation for some

955
00:56:53,239 --> 00:56:59,680
Speaker 1: unforeseen event? Am I stating that correctly? Absolutely? Yes. And

956
00:57:00,000 --> 00:57:02,879
Speaker 1: we've talked a little bit about how, and people ask

957
00:57:02,920 --> 00:57:05,239
Speaker 1: me at conferences forecast the next black swan. I think

958
00:57:05,239 --> 00:57:07,720
Speaker 1: it's actually the question I get asked the most. Because

959
00:57:08,000 --> 00:57:11,560
Speaker 1: I'm a strong believer in this phrase that there's no

960
00:57:11,680 --> 00:57:13,640
Speaker 1: such thing as a bad question. But I think that

961
00:57:13,680 --> 00:57:17,760
Speaker 1: one actually might be the bad question, because by definition,

962
00:57:17,840 --> 00:57:19,960
Speaker 1: you can't forecast the black swan. That's kind of what

963
00:57:20,000 --> 00:57:22,960
Speaker 1: a black swant is forecast of all. So So let's

964
00:57:23,000 --> 00:57:27,000
Speaker 1: talk a little bit about the technology you guys use

965
00:57:27,240 --> 00:57:31,040
Speaker 1: to create these models too, and to model out risks

966
00:57:31,120 --> 00:57:35,680
Speaker 1: and and other strategies. You build all that stuff in house. There,

967
00:57:35,840 --> 00:57:38,600
Speaker 1: you're not really a big buyer of off the shelf

968
00:57:39,360 --> 00:57:43,000
Speaker 1: UM risk management technology. Tell us a little bit about

969
00:57:43,760 --> 00:57:48,880
Speaker 1: your approach, which seems to be pretty comprehensive to thinking

970
00:57:48,920 --> 00:57:54,800
Speaker 1: about and planning for unforeseen risk. Yes, so, I mean

971
00:57:54,800 --> 00:57:56,440
Speaker 1: that's the first thing i'd say is that we essentially

972
00:57:56,480 --> 00:58:00,480
Speaker 1: build all of our own technology. We don't really buy technology.

973
00:58:00,520 --> 00:58:02,720
Speaker 1: We we buy the hardware, of course, but but we

974
00:58:02,840 --> 00:58:05,880
Speaker 1: write all the software, all the code ourselves. And that's

975
00:58:05,920 --> 00:58:09,040
Speaker 1: because we think that the things you can buy off

976
00:58:09,040 --> 00:58:11,080
Speaker 1: the shelf, or everyone can buy it off the shelf,

977
00:58:11,120 --> 00:58:14,400
Speaker 1: and therefore it's not really going to be a competitive advantage.

978
00:58:14,440 --> 00:58:16,800
Speaker 1: It's going to be it's going to be maybe a

979
00:58:16,880 --> 00:58:20,720
Speaker 1: base standard. And not trying to criticize the external products.

980
00:58:20,720 --> 00:58:22,320
Speaker 1: I just think that if you really want to have

981
00:58:22,360 --> 00:58:28,000
Speaker 1: an edge UM in building risk models or building short

982
00:58:28,120 --> 00:58:32,280
Speaker 1: term forecasters of risk or return or whatever, you need

983
00:58:32,320 --> 00:58:36,600
Speaker 1: to write your own code, your own software, and and

984
00:58:36,600 --> 00:58:38,400
Speaker 1: and you need to put a lot of effort into that,

985
00:58:38,920 --> 00:58:40,919
Speaker 1: and you need to create an environment where you can

986
00:58:41,000 --> 00:58:44,440
Speaker 1: hire the best software developers. And I often see people

987
00:58:44,440 --> 00:58:46,280
Speaker 1: saying well, you know, I hired a bunch of quants

988
00:58:46,280 --> 00:58:49,200
Speaker 1: and I had a bunch of developers as if you know,

989
00:58:49,240 --> 00:58:51,480
Speaker 1: that's the sort of a generic thing like buying a

990
00:58:51,480 --> 00:58:55,120
Speaker 1: loaf of bread or something. It's just not it's you know,

991
00:58:55,160 --> 00:58:59,080
Speaker 1: the best developers hundreds maybe even thousands of times as

992
00:58:59,120 --> 00:59:03,040
Speaker 1: productive as the average developers. So getting those best de

993
00:59:03,160 --> 00:59:06,680
Speaker 1: plenty your organization really important, and thinking about why they

994
00:59:06,680 --> 00:59:09,280
Speaker 1: would want to work for you and not want to

995
00:59:09,320 --> 00:59:13,400
Speaker 1: work for somebody else, that's pretty important. UM. So I

996
00:59:13,440 --> 00:59:15,720
Speaker 1: think for us, we felt that we can have an

997
00:59:15,800 --> 00:59:19,080
Speaker 1: edge by building better technology than you can buy off

998
00:59:19,080 --> 00:59:23,040
Speaker 1: the shelf and UM and then in order to get that,

999
00:59:23,080 --> 00:59:27,960
Speaker 1: we've invested a huge amount in providing UM a good

1000
00:59:28,080 --> 00:59:33,600
Speaker 1: environment for UH quantitative researchers and technologists to operate in.

1001
00:59:33,680 --> 00:59:35,520
Speaker 1: And just to give you a sort of a side

1002
00:59:35,520 --> 00:59:38,520
Speaker 1: example of that, we open source quite a lot of

1003
00:59:38,560 --> 00:59:41,360
Speaker 1: our code. So that means that, you know, we pay

1004
00:59:41,440 --> 00:59:43,880
Speaker 1: our developers to write code for us, and then we

1005
00:59:43,960 --> 00:59:45,640
Speaker 1: go and stick it on a website for other people

1006
00:59:45,680 --> 00:59:47,959
Speaker 1: to download it if they want. So why on earth,

1007
00:59:48,080 --> 00:59:50,160
Speaker 1: you know, what would possess you to do something like that?

1008
00:59:50,480 --> 00:59:54,080
Speaker 1: And the reason that we do it is that that

1009
00:59:54,120 --> 00:59:57,840
Speaker 1: then provides advertising to people that we're actually really serious

1010
00:59:58,480 --> 01:00:02,880
Speaker 1: software develop opers and that we take our code really seriously.

1011
01:00:03,240 --> 01:00:05,560
Speaker 1: And if you're a young software developer, you'll probably see

1012
01:00:05,600 --> 01:00:07,560
Speaker 1: the stuff that we've published and say, well, you know,

1013
01:00:07,600 --> 01:00:09,360
Speaker 1: actually I wouldn't mind working in a place like that

1014
01:00:09,440 --> 01:00:14,800
Speaker 1: because code and and and technology and standards and all

1015
01:00:14,800 --> 01:00:18,800
Speaker 1: those sorts of things are really high at at this firm.

1016
01:00:19,080 --> 01:00:24,960
Speaker 1: So that's how we think about UM investing in technology,

1017
01:00:25,200 --> 01:00:30,400
Speaker 1: investing in developers, creating a culture where developers and quant

1018
01:00:30,440 --> 01:00:35,280
Speaker 1: researchers really want to work. And the reason for that, UM,

1019
01:00:35,760 --> 01:00:38,600
Speaker 1: why would we carry on doing all this investment? It

1020
01:00:38,800 --> 01:00:41,040
Speaker 1: is really that you need you It's a very competitive

1021
01:00:41,080 --> 01:00:43,120
Speaker 1: business and you need to stay ahead all the time

1022
01:00:43,480 --> 01:00:45,479
Speaker 1: and you need to carry on innovating all the time.

1023
01:00:45,520 --> 01:00:49,040
Speaker 1: And if you don't, then somebody will eat your lunch.

1024
01:00:49,400 --> 01:00:52,760
Speaker 1: So from our perspective, we're always building new models. We're

1025
01:00:52,800 --> 01:00:55,840
Speaker 1: always coming up with new approaches to estimate risk. We're

1026
01:00:55,840 --> 01:00:58,800
Speaker 1: always worrying about, you know, how can we find a

1027
01:00:58,800 --> 01:01:03,000
Speaker 1: new alpha source, what my go wrong? UM? And how

1028
01:01:03,040 --> 01:01:06,960
Speaker 1: a market is changing in their structure this uh, you know,

1029
01:01:07,040 --> 01:01:11,080
Speaker 1: big effect of more retail investment in retail investors in

1030
01:01:11,600 --> 01:01:14,720
Speaker 1: equity markets today, how should we respond to that. That's

1031
01:01:14,800 --> 01:01:17,680
Speaker 1: really something which is just a very ongoing and continuous

1032
01:01:17,720 --> 01:01:20,520
Speaker 1: form of a place for us to invest, and when

1033
01:01:20,560 --> 01:01:22,520
Speaker 1: we try and get the benefits out of out of

1034
01:01:22,560 --> 01:01:26,120
Speaker 1: that over long term, m really really kind of interesting.

1035
01:01:26,400 --> 01:01:28,960
Speaker 1: Let's talk a little bit about machine learning. You guys

1036
01:01:28,960 --> 01:01:33,720
Speaker 1: have been on the cutting edge of that, including a

1037
01:01:33,840 --> 01:01:39,160
Speaker 1: collaboration with the University of Oxford at the Oxford Man Institute.

1038
01:01:40,000 --> 01:01:43,760
Speaker 1: Tell us what you guys are doing with machine learning

1039
01:01:43,840 --> 01:01:50,120
Speaker 1: and does any of this relate back to volatility? So

1040
01:01:50,640 --> 01:01:53,960
Speaker 1: what we're doing with machine learning is we're really saying

1041
01:01:54,120 --> 01:01:59,360
Speaker 1: that financial markets have patterns in them which you can

1042
01:01:59,520 --> 01:02:02,360
Speaker 1: dig out often profit from if you um if you

1043
01:02:02,600 --> 01:02:06,360
Speaker 1: look hard enough. The problem in financial markets is that

1044
01:02:06,400 --> 01:02:10,240
Speaker 1: the patterns are are pretty weak. You know, they're not

1045
01:02:10,440 --> 01:02:14,000
Speaker 1: They're not simple patterns. There are people like me would

1046
01:02:14,000 --> 01:02:16,880
Speaker 1: say there's a low signal to noise ratio. There's a

1047
01:02:16,920 --> 01:02:19,240
Speaker 1: lot of noise and not very much signal out there.

1048
01:02:19,960 --> 01:02:22,680
Speaker 1: So what are we using machine learning for. We're using

1049
01:02:22,680 --> 01:02:25,760
Speaker 1: it for a number of different things, but the underlying

1050
01:02:25,920 --> 01:02:31,480
Speaker 1: theme is that most models that people use in markets,

1051
01:02:31,520 --> 01:02:32,800
Speaker 1: and you could even think of it just as a

1052
01:02:32,880 --> 01:02:37,160
Speaker 1: value model. You know, you pee, for example, price over earnings.

1053
01:02:37,720 --> 01:02:39,920
Speaker 1: That's a linear model. It seems to sort of assume

1054
01:02:40,000 --> 01:02:43,160
Speaker 1: that price goes up in line with earnings. But we

1055
01:02:43,200 --> 01:02:45,480
Speaker 1: all know that when you look at markets, if there's

1056
01:02:45,520 --> 01:02:47,800
Speaker 1: one thing they don't ever do, it's move in a

1057
01:02:47,920 --> 01:02:50,720
Speaker 1: linear or straight line fashion. They move into every shape

1058
01:02:50,760 --> 01:02:54,440
Speaker 1: you could imagine except for the straight line. And UM,

1059
01:02:54,480 --> 01:02:56,440
Speaker 1: And what machine learning is really trying to do is

1060
01:02:56,480 --> 01:03:00,680
Speaker 1: to say, can I find much more subtle patterns? Um

1061
01:03:00,840 --> 01:03:03,840
Speaker 1: the straight line, which is what most of finance actually

1062
01:03:03,960 --> 01:03:07,800
Speaker 1: ends up using for modeling, And here are some examples

1063
01:03:07,800 --> 01:03:11,640
Speaker 1: of that. One that I'm particularly excited about is UM

1064
01:03:11,760 --> 01:03:15,560
Speaker 1: what we would call natural language processing, which is having

1065
01:03:15,640 --> 01:03:19,880
Speaker 1: machines read text. Now we all know that there's far

1066
01:03:19,960 --> 01:03:22,040
Speaker 1: too much for us all to read. You know, nobody

1067
01:03:22,080 --> 01:03:27,479
Speaker 1: can read every analyst report, every company earning statement, every

1068
01:03:27,520 --> 01:03:32,280
Speaker 1: annual report, attend every investor or day. Those are too much.

1069
01:03:32,760 --> 01:03:36,240
Speaker 1: So wouldn't it be wonderful if you could have machines

1070
01:03:36,320 --> 01:03:37,920
Speaker 1: to all that reading for you and tell you what

1071
01:03:38,000 --> 01:03:40,040
Speaker 1: to think at the end of it. And that might

1072
01:03:40,120 --> 01:03:44,280
Speaker 1: sound like science fiction, and at a certain level, I

1073
01:03:44,280 --> 01:03:48,439
Speaker 1: think it probably is. Science fiction today. But five years ago,

1074
01:03:48,560 --> 01:03:52,880
Speaker 1: if you said, well, machines will process images, will process

1075
01:03:52,960 --> 01:03:55,680
Speaker 1: pictures better than humans, I think people would look at

1076
01:03:55,720 --> 01:03:57,720
Speaker 1: you a bit funny and say, you know, well, no, no,

1077
01:03:57,880 --> 01:04:00,520
Speaker 1: not really. Now you know you have that on your phone.

1078
01:04:00,680 --> 01:04:02,960
Speaker 1: You just type a word into your phone, into your

1079
01:04:02,960 --> 01:04:05,960
Speaker 1: photos library and just watch it happen in action. It's

1080
01:04:06,120 --> 01:04:09,680
Speaker 1: just extraordinary how it will find all the pictures that

1081
01:04:09,840 --> 01:04:13,480
Speaker 1: reflect the words that you've that you've typed in. So

1082
01:04:13,840 --> 01:04:18,240
Speaker 1: machines definitely do process images better than humans. It's well known.

1083
01:04:18,360 --> 01:04:21,720
Speaker 1: For example, processing X ray images looking at for cancers

1084
01:04:21,880 --> 01:04:25,240
Speaker 1: is much better done today by humans, by machines and

1085
01:04:25,280 --> 01:04:27,120
Speaker 1: by humans. You might want the human at the end,

1086
01:04:27,440 --> 01:04:29,320
Speaker 1: but you want the machine to do all the sifting

1087
01:04:30,040 --> 01:04:33,040
Speaker 1: type work. And so one example of machine learning is

1088
01:04:33,080 --> 01:04:36,160
Speaker 1: getting machines to understand text and tell you, you know,

1089
01:04:36,880 --> 01:04:40,400
Speaker 1: go ahead, machine read every analyst report on you know,

1090
01:04:40,680 --> 01:04:43,840
Speaker 1: these hundred stocks, and tell me you know, not only

1091
01:04:43,880 --> 01:04:46,640
Speaker 1: the analyst reports, but all the newspapers in every language

1092
01:04:46,680 --> 01:04:49,040
Speaker 1: around the world, all the ending schools, read all of it,

1093
01:04:49,320 --> 01:04:51,760
Speaker 1: and then tell me what to think. Um, that is

1094
01:04:51,800 --> 01:04:54,040
Speaker 1: something which today is science fiction, but I don't think

1095
01:04:54,080 --> 01:04:56,840
Speaker 1: it will be in five years time. M that's that's

1096
01:04:56,880 --> 01:05:00,760
Speaker 1: really kind of intriguing. My last regg your question I

1097
01:05:01,200 --> 01:05:03,960
Speaker 1: have to ask you is, so you spent most of

1098
01:05:04,000 --> 01:05:08,680
Speaker 1: your career developing quantitative strategies. What are some of the

1099
01:05:08,680 --> 01:05:17,919
Speaker 1: biggest changes you've noticed. I think the firstly is actually

1100
01:05:17,920 --> 01:05:21,120
Speaker 1: a bit striking what hasn't changed? So a lot of

1101
01:05:21,120 --> 01:05:25,680
Speaker 1: things haven't changed. You know, the many of the standard

1102
01:05:25,720 --> 01:05:28,920
Speaker 1: models today are basically the same as the standard models

1103
01:05:28,960 --> 01:05:31,440
Speaker 1: twenty thirty years ago. We still use black sholes for

1104
01:05:31,480 --> 01:05:34,800
Speaker 1: pricing options. We'll still use the Barre risk model for

1105
01:05:34,960 --> 01:05:41,000
Speaker 1: calculating equity risk in portfolios. We still send data in

1106
01:05:41,120 --> 01:05:43,040
Speaker 1: very similar ways to the way that we sent it

1107
01:05:43,080 --> 01:05:46,520
Speaker 1: twenty thirty years ago, for the most part, in really

1108
01:05:46,640 --> 01:05:48,920
Speaker 1: terrible file formats. But nobody seems to come up with

1109
01:05:48,920 --> 01:05:53,320
Speaker 1: a better convention that everyone will accept. Um. So there's

1110
01:05:53,320 --> 01:05:56,320
Speaker 1: a lot of stuff that hasn't changed, But I think

1111
01:05:56,840 --> 01:05:59,160
Speaker 1: there are some things which have changed. The first is

1112
01:05:59,240 --> 01:06:02,480
Speaker 1: that people often like to sort of characterize the world.

1113
01:06:02,480 --> 01:06:08,040
Speaker 1: There's there's the quantz versus the discretionary people. So the quants,

1114
01:06:08,160 --> 01:06:10,720
Speaker 1: you know, the model driven people in some sort of

1115
01:06:10,760 --> 01:06:14,000
Speaker 1: battle with the discretionary people. And I say, I don't

1116
01:06:14,080 --> 01:06:17,480
Speaker 1: view it that way at all. I think that everyone

1117
01:06:17,640 --> 01:06:19,720
Speaker 1: is becoming much more quantitative in the way that they

1118
01:06:19,760 --> 01:06:22,320
Speaker 1: build and run their portfolios. The tools that we all

1119
01:06:22,360 --> 01:06:27,560
Speaker 1: have today on our Bloomberg terminals or on websites or

1120
01:06:28,480 --> 01:06:32,080
Speaker 1: products that we can buy from third parties or build ourselves,

1121
01:06:32,360 --> 01:06:36,800
Speaker 1: were essentially unimaginable um five or ten years ago, and

1122
01:06:36,840 --> 01:06:41,200
Speaker 1: everybody's got them. So give you an example. Fifteen twenty

1123
01:06:41,280 --> 01:06:45,480
Speaker 1: years ago, I was building quite sophisticated screening tools that

1124
01:06:45,520 --> 01:06:49,480
Speaker 1: would search equity markets for opportunities. Today you can basically

1125
01:06:49,520 --> 01:06:52,640
Speaker 1: do what I built on a Bloomberg terminal. So everybody's

1126
01:06:52,680 --> 01:06:54,640
Speaker 1: got it, or everybody that's got a Bloomberg terminal has

1127
01:06:54,640 --> 01:06:59,160
Speaker 1: got it. So there's been you know, all investors, not

1128
01:06:59,320 --> 01:07:01,760
Speaker 1: just the quants, but the discretion marriage as well. They've

1129
01:07:01,800 --> 01:07:05,000
Speaker 1: all become more quantitative. We've seen it was trading as well.

1130
01:07:05,280 --> 01:07:08,840
Speaker 1: Trading used to be people shouting each other on a

1131
01:07:08,880 --> 01:07:11,840
Speaker 1: trading floor. Today it's it's all machines and in almost

1132
01:07:11,920 --> 01:07:15,320
Speaker 1: every market around the world, and very sophisticated machines and

1133
01:07:15,680 --> 01:07:19,680
Speaker 1: very sophisticated algorithms trading with each other. So what I

1134
01:07:19,880 --> 01:07:22,920
Speaker 1: think I've really seen is everything in markets has become

1135
01:07:23,000 --> 01:07:26,560
Speaker 1: more quantitative. But then there are some things which have

1136
01:07:26,680 --> 01:07:32,160
Speaker 1: been kind of, you know, unattainable. So far, credit markets

1137
01:07:32,200 --> 01:07:37,080
Speaker 1: have remained stubbornly sort of immune to being taken over

1138
01:07:37,200 --> 01:07:42,360
Speaker 1: by more quantitative strategies. So far, private equity is the same.

1139
01:07:42,480 --> 01:07:44,920
Speaker 1: It's it's really done in the same way as it

1140
01:07:45,040 --> 01:07:48,520
Speaker 1: was twenty or thirty years ago. Um, And I think

1141
01:07:48,520 --> 01:07:50,920
Speaker 1: that will probably change over time. Was certainly starting to

1142
01:07:50,960 --> 01:07:54,000
Speaker 1: see that in credit, for example, where the markets are

1143
01:07:54,040 --> 01:07:57,800
Speaker 1: starting to trade more like equity markets or futures markets,

1144
01:07:58,080 --> 01:08:00,720
Speaker 1: and it is starting to be possible to build the

1145
01:08:00,800 --> 01:08:03,960
Speaker 1: same sorts of risk models and the same sorts of

1146
01:08:03,960 --> 01:08:08,240
Speaker 1: of alpha models, of return forecasting models in credit that

1147
01:08:08,320 --> 01:08:10,600
Speaker 1: you've had in equities for a long time. So I

1148
01:08:10,600 --> 01:08:13,320
Speaker 1: think my real thing is that everything has become more quantitative,

1149
01:08:13,360 --> 01:08:14,720
Speaker 1: and I think it is going to become a whole

1150
01:08:14,720 --> 01:08:18,519
Speaker 1: bunch more quantitative over the next twenty years. Alright, so

1151
01:08:18,600 --> 01:08:20,960
Speaker 1: let's jump to our favorite questions that we ask all

1152
01:08:21,000 --> 01:08:25,080
Speaker 1: our guests. Tell us what you've been streaming this past year,

1153
01:08:25,320 --> 01:08:29,400
Speaker 1: or anything interesting that you're watching on Netflix or Amazon

1154
01:08:29,439 --> 01:08:35,639
Speaker 1: Prime or or whatever. Um. I'm not a huge watcher

1155
01:08:35,720 --> 01:08:39,840
Speaker 1: of things on streaming services. But you know, I have

1156
01:08:40,040 --> 01:08:45,240
Speaker 1: been um watching a rewatching a series of films by

1157
01:08:45,280 --> 01:08:49,360
Speaker 1: a famous British actor called Bill Nike, written by a

1158
01:08:49,400 --> 01:08:52,559
Speaker 1: playwright called David Hair. The first of them is called

1159
01:08:52,600 --> 01:08:57,920
Speaker 1: Page eight and there are quite sophisticated plays about um

1160
01:08:58,880 --> 01:09:02,760
Speaker 1: An m I six agent and his life struggles. So

1161
01:09:02,880 --> 01:09:06,479
Speaker 1: that's what I've been streaming, and I you can find

1162
01:09:06,520 --> 01:09:10,439
Speaker 1: it on on on Netflix. Page eight is the is

1163
01:09:10,479 --> 01:09:12,840
Speaker 1: the first of them. Well, we'll definitely check that out.

1164
01:09:13,280 --> 01:09:16,720
Speaker 1: Tell me us, tell us a little bit about your

1165
01:09:16,800 --> 01:09:22,720
Speaker 1: mentors who helped to shape your career. So I was

1166
01:09:22,840 --> 01:09:27,920
Speaker 1: very fortunate in the first fifteen years of my career

1167
01:09:27,960 --> 01:09:30,479
Speaker 1: I was at Golden Sacks. It was an outstanding place

1168
01:09:30,520 --> 01:09:36,599
Speaker 1: to work. The people that I particularly worked with over

1169
01:09:36,640 --> 01:09:43,120
Speaker 1: that period were really two or three folks. Um Manny Roman,

1170
01:09:43,240 --> 01:09:46,599
Speaker 1: who today is the CEO of Pimco, I worked with

1171
01:09:46,960 --> 01:09:51,720
Speaker 1: pretty much twenty five years, both at Golden Sacks as

1172
01:09:51,760 --> 01:09:54,080
Speaker 1: well as at Man Group. Before he went off to PIMCO.

1173
01:09:54,880 --> 01:09:58,760
Speaker 1: Worked with a fellow called Garish Ready who went on

1174
01:09:58,840 --> 01:10:02,040
Speaker 1: to run a fund of funds business called Prisma Um

1175
01:10:02,120 --> 01:10:04,320
Speaker 1: and a fellow called Mark Zorak, who went on to

1176
01:10:04,400 --> 01:10:08,600
Speaker 1: be a professor at Cornell, and I think it was

1177
01:10:08,640 --> 01:10:11,679
Speaker 1: important to me to have a variety of different people

1178
01:10:12,080 --> 01:10:15,800
Speaker 1: to learn from and to sort of build different experiences.

1179
01:10:15,840 --> 01:10:18,760
Speaker 1: All three of them are extremely different people, but I

1180
01:10:18,840 --> 01:10:22,920
Speaker 1: really learned, I think, both how to manage people and

1181
01:10:22,960 --> 01:10:26,040
Speaker 1: how to get to the core of a problem. I think,

1182
01:10:26,040 --> 01:10:28,519
Speaker 1: how to work out what was important and what was

1183
01:10:28,600 --> 01:10:32,519
Speaker 1: not important, and not to give different practice of equal

1184
01:10:32,560 --> 01:10:36,960
Speaker 1: weight when you're making decisions. And I learned that differently

1185
01:10:37,080 --> 01:10:40,559
Speaker 1: from from each of those um three people who are

1186
01:10:40,600 --> 01:10:44,439
Speaker 1: really sort of my core mentors. Very interesting, tell us

1187
01:10:44,479 --> 01:10:46,759
Speaker 1: about some of your favorite books. What are you reading

1188
01:10:46,840 --> 01:10:51,760
Speaker 1: right now and what are some of your all time faves. Well, so,

1189
01:10:51,920 --> 01:10:56,640
Speaker 1: I'm I am very interested in architecture, and so I

1190
01:10:56,680 --> 01:11:02,439
Speaker 1: tend to read relatively quirky and eclectic books. I'm currently

1191
01:11:02,479 --> 01:11:05,920
Speaker 1: reading something about Nordic Modernism, which I suspect will not

1192
01:11:05,960 --> 01:11:08,880
Speaker 1: be that popular with your audience. It's a niche area.

1193
01:11:09,439 --> 01:11:15,680
Speaker 1: UM but um So, I I think architectural theory is

1194
01:11:15,680 --> 01:11:18,000
Speaker 1: something which I'm very interested in. In another life, I

1195
01:11:18,080 --> 01:11:20,840
Speaker 1: might have been an architect. Um So that's sort of

1196
01:11:20,880 --> 01:11:24,720
Speaker 1: one area of interest for me. Um. I'm actually a

1197
01:11:24,840 --> 01:11:27,280
Speaker 1: keen piano player as well. So I know you asked

1198
01:11:27,320 --> 01:11:29,680
Speaker 1: about books, but I play a lot of music, and

1199
01:11:29,880 --> 01:11:34,559
Speaker 1: at the moment I'm um playing um, some early twentieth

1200
01:11:34,600 --> 01:11:40,280
Speaker 1: century music by Debussy, which drives my family nuts because

1201
01:11:40,280 --> 01:11:42,920
Speaker 1: it's very hard to play, but if you play it well,

1202
01:11:42,960 --> 01:11:44,720
Speaker 1: it sounds good. I'm not sure I've mastered how to

1203
01:11:44,720 --> 01:11:47,160
Speaker 1: play it well yet, So that's another sort of core

1204
01:11:47,240 --> 01:11:49,360
Speaker 1: part of my life is try and play the piano

1205
01:11:49,439 --> 01:11:52,559
Speaker 1: for at least an hour every day and that sort

1206
01:11:52,600 --> 01:11:55,800
Speaker 1: of straightens out my mind at the end of the day. Um.

1207
01:11:55,960 --> 01:12:00,760
Speaker 1: And in fiction, I amid I'm really a sort of

1208
01:12:00,840 --> 01:12:06,479
Speaker 1: enthusiast for mid twentieth century writers, So Treuman, Capote or

1209
01:12:06,680 --> 01:12:09,360
Speaker 1: Gray and Green or sort of that. That group of

1210
01:12:10,280 --> 01:12:14,400
Speaker 1: writers are all my favorites. Very interesting. You know, there's

1211
01:12:14,439 --> 01:12:18,080
Speaker 1: some really fascinating I know you're not a big video guy,

1212
01:12:18,200 --> 01:12:23,839
Speaker 1: but on YouTube there are some really fascinating shows about music,

1213
01:12:24,160 --> 01:12:29,479
Speaker 1: like Song Exploder or Polyphonic or there's just a run

1214
01:12:29,600 --> 01:12:36,280
Speaker 1: of things that take apart various genres of music from

1215
01:12:36,360 --> 01:12:41,320
Speaker 1: a musicologists or historians perspective. There if you're if you're

1216
01:12:41,320 --> 01:12:44,479
Speaker 1: a classical music fan and play the piano, you may

1217
01:12:44,600 --> 01:12:48,200
Speaker 1: find some of that stuff interesting because the parallels back

1218
01:12:48,280 --> 01:12:52,559
Speaker 1: and forth between classical music and pop music. The pop

1219
01:12:52,600 --> 01:12:57,320
Speaker 1: audience is not familiar with it, but the classical fans

1220
01:12:57,439 --> 01:12:59,759
Speaker 1: clearly are. You you might find some of that stuff

1221
01:12:59,760 --> 01:13:04,280
Speaker 1: in interesting. Um. And actually there are very strong parallels

1222
01:13:04,320 --> 01:13:07,599
Speaker 1: between music and architecture as well. Architecture is all about rhythms.

1223
01:13:07,680 --> 01:13:09,960
Speaker 1: People often don't see it, but they're there. Well you,

1224
01:13:10,200 --> 01:13:13,559
Speaker 1: I'm gonna assume that someone like you read godal escher

1225
01:13:13,640 --> 01:13:18,280
Speaker 1: Bach twenty years ago. Um, am, I uh so they're

1226
01:13:18,439 --> 01:13:23,320
Speaker 1: the same parallels between pattern and repetition and and how

1227
01:13:23,479 --> 01:13:27,519
Speaker 1: they things morph over time. It's it's it's really and

1228
01:13:27,720 --> 01:13:30,840
Speaker 1: and I never thought about architecture, um the way music

1229
01:13:30,960 --> 01:13:33,439
Speaker 1: and math show up and and art show up. But

1230
01:13:33,920 --> 01:13:37,160
Speaker 1: I guess in a lot of ways, especially with UM,

1231
01:13:37,880 --> 01:13:42,760
Speaker 1: with with larger buildings, clearly some of those fractal progressions

1232
01:13:42,800 --> 01:13:45,880
Speaker 1: are are there. I'm we're really off on a on

1233
01:13:46,000 --> 01:13:49,400
Speaker 1: a digression. Let me let me go to my next question. UM,

1234
01:13:49,520 --> 01:13:51,479
Speaker 1: what sort of advice would you give to a recent

1235
01:13:51,560 --> 01:13:57,000
Speaker 1: college grad who is interested in a career in quantitative

1236
01:13:57,080 --> 01:14:03,599
Speaker 1: strategies or risk management and find it so My advice

1237
01:14:03,680 --> 01:14:06,360
Speaker 1: would be not to get too narrow too quickly, and

1238
01:14:06,600 --> 01:14:09,800
Speaker 1: to try and build as broad a range of experience

1239
01:14:09,880 --> 01:14:12,680
Speaker 1: as you can. In my case, I did quite a

1240
01:14:12,760 --> 01:14:15,400
Speaker 1: few things in the early years of my career and

1241
01:14:15,479 --> 01:14:18,920
Speaker 1: they really turned out to be differentiated for me later on.

1242
01:14:19,240 --> 01:14:21,880
Speaker 1: So I worked a bill in corporate finance. I worked

1243
01:14:21,920 --> 01:14:23,720
Speaker 1: out it wasn't for me, but I learned to heck

1244
01:14:23,760 --> 01:14:25,880
Speaker 1: of a lot in my a couple of years of

1245
01:14:25,960 --> 01:14:29,480
Speaker 1: doing corporate finance, I worked and fixed income and equities

1246
01:14:29,520 --> 01:14:31,599
Speaker 1: and credit. I work on the cell side as well

1247
01:14:31,680 --> 01:14:35,000
Speaker 1: as the buy side, and that was incredibly valuable to me.

1248
01:14:35,720 --> 01:14:39,000
Speaker 1: It gave me just different approaches to problems when I

1249
01:14:39,640 --> 01:14:41,519
Speaker 1: came across them. So I think that's the first piece

1250
01:14:41,560 --> 01:14:44,320
Speaker 1: of advice. The second piece of advice, I think is

1251
01:14:44,479 --> 01:14:48,040
Speaker 1: that like most quants, I thought I was good at

1252
01:14:48,080 --> 01:14:51,720
Speaker 1: math and um and you know it probably wasn't bad

1253
01:14:51,760 --> 01:14:53,640
Speaker 1: at it. But it turns out that actually there are

1254
01:14:53,640 --> 01:14:57,040
Speaker 1: a lot of people that are good at math, and I,

1255
01:14:57,640 --> 01:15:00,280
Speaker 1: from my perspective, learned that I had some skills which

1256
01:15:00,320 --> 01:15:03,719
Speaker 1: maybe differentiated me a little bit from the core skill

1257
01:15:04,000 --> 01:15:05,840
Speaker 1: of all these other people that were good at math

1258
01:15:06,280 --> 01:15:09,120
Speaker 1: and those in my case where I was good at

1259
01:15:09,160 --> 01:15:11,839
Speaker 1: making decisions I think I remained good at making decisions.

1260
01:15:11,880 --> 01:15:13,560
Speaker 1: I can see things, I don't have to spend a

1261
01:15:13,600 --> 01:15:15,800
Speaker 1: lot of time thinking about it. I can decide and

1262
01:15:15,960 --> 01:15:18,920
Speaker 1: move on and have too much regret. And I learned

1263
01:15:18,960 --> 01:15:23,160
Speaker 1: that I was, I think, better than many quants a

1264
01:15:23,320 --> 01:15:27,759
Speaker 1: communicating a quantitative things in straightforward English, which most quants

1265
01:15:27,800 --> 01:15:29,760
Speaker 1: are not pretty good at. And so I really tried

1266
01:15:29,800 --> 01:15:32,560
Speaker 1: to work out what are my other strengths that my

1267
01:15:32,720 --> 01:15:36,479
Speaker 1: differentiating strengths, and and and and tried to use those.

1268
01:15:37,120 --> 01:15:39,600
Speaker 1: And I would recommend any quant workout. If you're just

1269
01:15:39,640 --> 01:15:41,720
Speaker 1: going to go straight for a math competition, you know

1270
01:15:41,800 --> 01:15:43,680
Speaker 1: there's some people who are pretty damn good at math

1271
01:15:43,760 --> 01:15:45,360
Speaker 1: out there, so you're going to have to think about

1272
01:15:45,400 --> 01:15:48,240
Speaker 1: your extra strengths to things which separate you from the crowd.

1273
01:15:48,960 --> 01:15:51,120
Speaker 1: That makes a lot of sense in our final question,

1274
01:15:51,840 --> 01:15:54,519
Speaker 1: what do you know about the world of investing today?

1275
01:15:54,880 --> 01:15:57,920
Speaker 1: You wish you knew thirty or so years ago when

1276
01:15:58,000 --> 01:16:03,800
Speaker 1: you were first getting started. So I think thirty or

1277
01:16:03,840 --> 01:16:08,479
Speaker 1: so years ago. Um. I think the thing that I

1278
01:16:08,680 --> 01:16:13,320
Speaker 1: most didn't realize was how much more tech and quant

1279
01:16:13,439 --> 01:16:18,840
Speaker 1: focused the world was going to become. And I slightly

1280
01:16:19,040 --> 01:16:24,040
Speaker 1: under emphasized my quantum tech skills at that time. So

1281
01:16:24,400 --> 01:16:26,960
Speaker 1: that's the first thing I think, and I shouldn't have done.

1282
01:16:27,000 --> 01:16:29,479
Speaker 1: You know, today it all feels very obvious that tech

1283
01:16:29,520 --> 01:16:32,920
Speaker 1: has dominated our lives so much, and behind most tech

1284
01:16:33,000 --> 01:16:35,360
Speaker 1: through a lot of algorithms, but thirty years ago it

1285
01:16:35,520 --> 01:16:39,000
Speaker 1: wasn't so obvious. So that's the first thing. I think.

1286
01:16:39,080 --> 01:16:41,960
Speaker 1: The second thing that I wish i'd known thirty years

1287
01:16:42,000 --> 01:16:46,640
Speaker 1: ago was that it's very easy to gravitate towards the

1288
01:16:46,720 --> 01:16:51,120
Speaker 1: glamorous businesses. So when I was sitting on trading floors

1289
01:16:51,160 --> 01:16:55,360
Speaker 1: that called them sacks, then the glamorous bit was was

1290
01:16:55,479 --> 01:16:58,000
Speaker 1: trading what we call the exotic derivatives or the more

1291
01:16:58,080 --> 01:17:03,400
Speaker 1: complicated derivatives contracts. But actually that business almost doesn't exist anymore.

1292
01:17:03,880 --> 01:17:06,320
Speaker 1: It does exist, but it's really much smaller than it

1293
01:17:06,520 --> 01:17:09,120
Speaker 1: was twenty years ago. And that's what everyone sort have

1294
01:17:09,160 --> 01:17:11,679
Speaker 1: wanted to do. And then there were other things which

1295
01:17:11,720 --> 01:17:17,800
Speaker 1: were apparently less interesting, like understanding equity in disease um

1296
01:17:18,320 --> 01:17:21,960
Speaker 1: or um or building quant equity models or something like that,

1297
01:17:22,040 --> 01:17:24,160
Speaker 1: and those turned out to be much bigger things. So

1298
01:17:24,240 --> 01:17:26,519
Speaker 1: I think the second thing which I've realized, which I

1299
01:17:26,600 --> 01:17:29,720
Speaker 1: wish i'd realized, and understood twenty years ago is the

1300
01:17:29,800 --> 01:17:32,720
Speaker 1: glamorous stuff is not always the stuff to go for.

1301
01:17:33,240 --> 01:17:36,559
Speaker 1: Often it's the stuff that actually people sort of think

1302
01:17:36,560 --> 01:17:39,200
Speaker 1: maybe it's a bit boring or something like that. Um,

1303
01:17:39,600 --> 01:17:42,800
Speaker 1: but there are very often the big opportunities lie in

1304
01:17:42,880 --> 01:17:45,200
Speaker 1: the stuff that people think is a bit boring. M

1305
01:17:45,920 --> 01:17:49,840
Speaker 1: really quite quite interesting. Thank you, Sandy for being so

1306
01:17:50,160 --> 01:17:53,280
Speaker 1: generous with your time. We have been speaking with Sandy

1307
01:17:53,600 --> 01:17:56,640
Speaker 1: rat Trey. He is the chief investment officer of the

1308
01:17:56,800 --> 01:18:00,519
Speaker 1: hundred billion dollar Man Group, as well as the co

1309
01:18:00,760 --> 01:18:03,920
Speaker 1: inventor of the VIX index and the author of the

1310
01:18:04,000 --> 01:18:09,120
Speaker 1: book Strategic Risk Management. If you enjoyed this conversation, well,

1311
01:18:09,240 --> 01:18:12,080
Speaker 1: be sure and check out any of our previous three

1312
01:18:12,479 --> 01:18:16,200
Speaker 1: d and fifty or so UH interviews that we've done

1313
01:18:16,280 --> 01:18:20,880
Speaker 1: over the past seven years. You can find those at iTunes, Spotify,

1314
01:18:21,720 --> 01:18:26,400
Speaker 1: a cast, wherever final podcasts are sold. We love your comments,

1315
01:18:26,439 --> 01:18:31,360
Speaker 1: feedback and suggestions right to us at m IB podcast

1316
01:18:31,600 --> 01:18:34,880
Speaker 1: at Bloomberg dot net. You can sign up for my

1317
01:18:35,160 --> 01:18:38,760
Speaker 1: daily reading list at Rit Halts dot com. Check out

1318
01:18:38,840 --> 01:18:43,360
Speaker 1: my weekly column on Bloomberg dot com slash Opinion. Follow

1319
01:18:43,400 --> 01:18:47,479
Speaker 1: me on Twitter at rit Halts. I would be remiss

1320
01:18:47,520 --> 01:18:50,000
Speaker 1: if I did not thank the crack staff that helps

1321
01:18:50,040 --> 01:18:55,479
Speaker 1: put these conversations together each week. Charlie Vollmer is my

1322
01:18:55,840 --> 01:19:00,400
Speaker 1: audio engineer. Paris Walt is my producer. A tick to

1323
01:19:00,520 --> 01:19:04,479
Speaker 1: Val Brunn is our project manager. Michael Batnick is our

1324
01:19:04,520 --> 01:19:09,080
Speaker 1: head of research. I'm Barry Ridholtz. You've been listening to

1325
01:19:09,240 --> 01:19:11,840
Speaker 1: Masters in Business on Bloomberg Radio.