1
00:00:10,880 --> 00:00:14,680
Speaker 1: Hello, and welcome to another episode of the Odd Thoughts podcast.

2
00:00:14,760 --> 00:00:18,200
Speaker 1: I'm Tracy Alloway and I'm Joe Wisn't all you know

3
00:00:18,239 --> 00:00:26,440
Speaker 1: what you never see go on? That's that's pretty open ended? Yeah, okay, well,

4
00:00:26,480 --> 00:00:28,760
Speaker 1: you know, as a journalist, I'm told to ask open

5
00:00:28,840 --> 00:00:32,960
Speaker 1: ended questions, but you never see you never see anyone

6
00:00:33,000 --> 00:00:38,760
Speaker 1: connect climate change with real yields and secular stagnation. But

7
00:00:38,880 --> 00:00:43,240
Speaker 1: I in large, those are pretty separate conversations. I mean,

8
00:00:43,360 --> 00:00:47,000
Speaker 1: maybe the secular stagnation part, but I definitely do not

9
00:00:47,200 --> 00:00:50,960
Speaker 1: see much conversation connecting the dodge between anything that's going

10
00:00:51,000 --> 00:00:53,840
Speaker 1: on in interest rates or real rates or anything like

11
00:00:53,880 --> 00:00:58,120
Speaker 1: that in climate Yeah, that's exactly right. So despite the

12
00:00:58,160 --> 00:01:01,520
Speaker 1: fact that every bank in the world seems to be

13
00:01:01,680 --> 00:01:04,600
Speaker 1: really high on E s G right now, and certainly

14
00:01:04,640 --> 00:01:06,960
Speaker 1: you and I are on the receiving ends of you know,

15
00:01:07,319 --> 00:01:10,440
Speaker 1: tons of press releases about what banks are doing in

16
00:01:10,480 --> 00:01:13,839
Speaker 1: the space, you actually don't see that many people talk

17
00:01:13,840 --> 00:01:17,839
Speaker 1: about climate change from a market's perspective, which is kind

18
00:01:17,880 --> 00:01:21,080
Speaker 1: of weird because if you think about the battle, the

19
00:01:21,120 --> 00:01:24,360
Speaker 1: climate change battle right now, so much of it is

20
00:01:24,400 --> 00:01:28,919
Speaker 1: about estimating the costs and the benefits or the cost

21
00:01:29,160 --> 00:01:34,000
Speaker 1: versus the benefits of actually tackling this issue. And if

22
00:01:34,040 --> 00:01:37,720
Speaker 1: you think about how you calculate costs and benefit, that's

23
00:01:37,720 --> 00:01:41,760
Speaker 1: basically a market question, right, Like you have to consider

24
00:01:42,080 --> 00:01:44,520
Speaker 1: how the market is functioning and at what rate the

25
00:01:44,560 --> 00:01:47,800
Speaker 1: market is actually rewarding action at any moment in time

26
00:01:47,880 --> 00:01:51,400
Speaker 1: in order to come up with those kind of estimates. Yeah,

27
00:01:51,480 --> 00:01:53,280
Speaker 1: I think that's ruing. I mean, I think if this

28
00:01:53,360 --> 00:01:56,000
Speaker 1: is where like I sort of like struggle with some

29
00:01:56,080 --> 00:02:00,680
Speaker 1: of this. There is this widespread expectation and of like

30
00:02:00,920 --> 00:02:09,839
Speaker 1: ongoing environmental degradation, potentially more um economic and environmental catastrophes,

31
00:02:10,280 --> 00:02:13,920
Speaker 1: horrendous things that will happen that may be associated with

32
00:02:14,040 --> 00:02:18,120
Speaker 1: climate change. On the other hand, like from a market standpoint,

33
00:02:18,240 --> 00:02:22,280
Speaker 1: it still seems as though, for the most part, most

34
00:02:22,320 --> 00:02:24,800
Speaker 1: of the things that get priced, whether we talk about

35
00:02:24,840 --> 00:02:29,800
Speaker 1: priced in, are more on the sort of like regulatory side.

36
00:02:30,360 --> 00:02:33,320
Speaker 1: It doesn't feel like there are many examples of the

37
00:02:34,000 --> 00:02:39,880
Speaker 1: of the environmental risks directly themselves manifesting in price, just

38
00:02:40,040 --> 00:02:43,720
Speaker 1: yet more about pricing in the sort of regulatory response

39
00:02:44,080 --> 00:02:48,600
Speaker 1: or policy response to this sort of ongoing threat. Right. So,

40
00:02:48,760 --> 00:02:51,040
Speaker 1: I know I started this conversation by saying that you'd

41
00:02:51,040 --> 00:02:54,400
Speaker 1: never see anyone talk about real yields and secularis TechNation

42
00:02:54,480 --> 00:02:57,400
Speaker 1: and climate change. But that's not strictly true, because there

43
00:02:57,520 --> 00:03:00,720
Speaker 1: is one person who has done it um and they

44
00:03:00,720 --> 00:03:04,880
Speaker 1: published a whole paper on the topic, which UM definitely

45
00:03:04,919 --> 00:03:07,320
Speaker 1: caught my eye. But we're going to be speaking to

46
00:03:07,919 --> 00:03:11,959
Speaker 1: Josh Younger again today. He's a managing director at JP

47
00:03:12,080 --> 00:03:16,959
Speaker 1: Morgan and also a multi appearance All Thoughts guest. I

48
00:03:16,960 --> 00:03:19,760
Speaker 1: think this is going to be his fourth time on

49
00:03:19,800 --> 00:03:22,960
Speaker 1: the podcast, so we're gonna get his thoughts on the

50
00:03:23,000 --> 00:03:26,280
Speaker 1: climate change and the link to yields, but we're also

51
00:03:26,280 --> 00:03:29,200
Speaker 1: going to go into some of the recent drama in

52
00:03:29,240 --> 00:03:32,600
Speaker 1: the treasuries market and I and in addition to him

53
00:03:32,639 --> 00:03:34,440
Speaker 1: as having done work on this, he's sort of the

54
00:03:34,440 --> 00:03:37,680
Speaker 1: perfect person to connect all these duds because in addition

55
00:03:37,720 --> 00:03:40,440
Speaker 1: to being a rape strategist managing director at JP Morgan,

56
00:03:40,720 --> 00:03:45,720
Speaker 1: he's also a trained astrophysicist, so the perfect person to

57
00:03:45,880 --> 00:03:51,000
Speaker 1: connect science and interest rates. So I'm really looking part

58
00:03:51,000 --> 00:03:56,119
Speaker 1: of this conversation excellent, Um, climate science meets interest rates.

59
00:03:56,160 --> 00:03:58,880
Speaker 1: I love it all right, Josh, welcome to All Thoughts

60
00:03:58,960 --> 00:04:02,520
Speaker 1: yet again. Yeah, thanks for having It's it's great to

61
00:04:02,520 --> 00:04:06,600
Speaker 1: be back. So I mentioned this idea that when we

62
00:04:06,640 --> 00:04:11,240
Speaker 1: talk about climate change policy, we generally think about trade

63
00:04:11,240 --> 00:04:15,560
Speaker 1: offs between cost versus benefit, And one of the things

64
00:04:15,560 --> 00:04:20,200
Speaker 1: I'm really interested in is what traditional analysis looks like

65
00:04:20,320 --> 00:04:22,880
Speaker 1: when it comes to cost benefits on on this topic,

66
00:04:22,960 --> 00:04:25,760
Speaker 1: because I don't know. When I think about climate change,

67
00:04:25,920 --> 00:04:29,200
Speaker 1: I think about the worst case scenario being that we

68
00:04:29,320 --> 00:04:32,440
Speaker 1: basically all die from global warming and you know, world

69
00:04:32,440 --> 00:04:36,200
Speaker 1: War three breaks out over resources and the entire world collapses.

70
00:04:36,240 --> 00:04:38,520
Speaker 1: So I've always been curious about how you put a

71
00:04:38,560 --> 00:04:42,400
Speaker 1: sort of a number around those sorts of scenarios. So

72
00:04:42,720 --> 00:04:46,200
Speaker 1: maybe just to begin with, walk us through how people

73
00:04:46,400 --> 00:04:49,320
Speaker 1: normally do that equation. How do they come up with

74
00:04:49,400 --> 00:04:53,159
Speaker 1: that cost benefit framework? Yeah? Sure, So it's it's a

75
00:04:53,200 --> 00:04:56,359
Speaker 1: bunch of different steps, and it's it's really about connecting

76
00:04:57,440 --> 00:05:03,200
Speaker 1: a few related but but sort institutionally separate fields of studies.

77
00:05:03,240 --> 00:05:05,919
Speaker 1: So on the one hand, there's the client science itself,

78
00:05:06,000 --> 00:05:11,120
Speaker 1: which is doing projections of global mean temperatures and sea level,

79
00:05:11,880 --> 00:05:17,359
Speaker 1: sea level rises and various other types of forecasting, and

80
00:05:17,400 --> 00:05:21,880
Speaker 1: those tended to go out hundred years um. That is

81
00:05:21,920 --> 00:05:25,279
Speaker 1: a highly nonlinear, very complicated process that's usually done with

82
00:05:25,320 --> 00:05:28,800
Speaker 1: supercomputers with these simulations, essentially you can't do it with

83
00:05:28,839 --> 00:05:32,679
Speaker 1: a pencil and paper. Um, So that's the climate science angle.

84
00:05:33,080 --> 00:05:36,600
Speaker 1: And then when we think about policy, the question is

85
00:05:36,800 --> 00:05:39,800
Speaker 1: sort of what does that mean to humanity? What does

86
00:05:39,839 --> 00:05:42,479
Speaker 1: it mean to us? And how do we start start

87
00:05:42,520 --> 00:05:45,720
Speaker 1: to try to quantify that. So on the one hand,

88
00:05:45,800 --> 00:05:48,159
Speaker 1: you can think about the sort of more moral angle

89
00:05:48,240 --> 00:05:49,919
Speaker 1: to this, which is what you're alluding to, which is,

90
00:05:50,360 --> 00:05:54,039
Speaker 1: you know, people dying in wars and famine and disease

91
00:05:54,080 --> 00:05:56,880
Speaker 1: and so forth, and that's you know, obviously that the

92
00:05:57,200 --> 00:06:00,560
Speaker 1: none of it, but but also very hard to put

93
00:06:00,720 --> 00:06:03,920
Speaker 1: numbers around. So there's a tendency to kind of think

94
00:06:03,960 --> 00:06:06,760
Speaker 1: of this in economic terms, which is, you know, as

95
00:06:06,800 --> 00:06:10,280
Speaker 1: horrible as all those outcomes sound, that they do ultimately

96
00:06:10,279 --> 00:06:12,720
Speaker 1: boiled down in some sense to a cost. You can

97
00:06:12,720 --> 00:06:15,560
Speaker 1: put a dollar amount on it. Insurance companies do it

98
00:06:15,600 --> 00:06:20,080
Speaker 1: all the time. And uh, the economic impact of climate

99
00:06:20,120 --> 00:06:22,960
Speaker 1: change which incorporates all of these factors. So there's there's

100
00:06:23,000 --> 00:06:25,839
Speaker 1: kind of the purely economic impacts, which is, you know,

101
00:06:26,040 --> 00:06:30,400
Speaker 1: our coal mines productive, can we produce enough wheat? What's

102
00:06:30,640 --> 00:06:33,280
Speaker 1: what's the net impact on consumption? Like things like that,

103
00:06:33,360 --> 00:06:37,280
Speaker 1: And then you can have the more banal sounding mortality

104
00:06:37,920 --> 00:06:42,920
Speaker 1: projections and and other sorts of sort of nontraditional economic impacts,

105
00:06:42,920 --> 00:06:46,680
Speaker 1: and that ultimately gets boiled down to a dollar amount um.

106
00:06:46,800 --> 00:06:51,000
Speaker 1: And the process of of putting those things together is

107
00:06:51,360 --> 00:06:54,919
Speaker 1: what's generally referred to as as integrated assessment models. So

108
00:06:55,080 --> 00:06:59,440
Speaker 1: connecting climate science and economics with some with some other

109
00:06:59,480 --> 00:07:02,520
Speaker 1: inputs is well um. And the goal from that exercise

110
00:07:02,600 --> 00:07:06,159
Speaker 1: is to try to inform policy because there's a real

111
00:07:06,240 --> 00:07:09,440
Speaker 1: cost benefit here. Um. You know, if we want to

112
00:07:09,480 --> 00:07:11,800
Speaker 1: completely avoid climate change, we're all supposed to go back

113
00:07:12,160 --> 00:07:14,440
Speaker 1: to a pre industrial world where we don't burn carbon

114
00:07:14,480 --> 00:07:20,120
Speaker 1: at all and we live on sort of uh independently um,

115
00:07:20,280 --> 00:07:23,920
Speaker 1: self sufficient farms, and you know we walk everywhere or

116
00:07:23,920 --> 00:07:25,720
Speaker 1: maybe ride a bike, but even the bike's got to

117
00:07:25,760 --> 00:07:28,520
Speaker 1: get produced. So like, there's a cost that's not worth

118
00:07:28,560 --> 00:07:33,960
Speaker 1: bearing to most of the world. And so between doing

119
00:07:34,000 --> 00:07:39,360
Speaker 1: nothing and reversing the clock fifteen years um and undoing

120
00:07:39,360 --> 00:07:43,840
Speaker 1: all technological development, there's some number that represents the acceptable

121
00:07:43,880 --> 00:07:47,880
Speaker 1: cost to society, and the the process here is trying

122
00:07:47,920 --> 00:07:52,160
Speaker 1: to arrive at at that number through this rough mean.

123
00:07:52,200 --> 00:07:53,880
Speaker 1: There's many different ways to do it, but you're trying

124
00:07:53,880 --> 00:07:56,920
Speaker 1: to get ultimately to that break even to that number

125
00:07:56,920 --> 00:08:00,160
Speaker 1: that the acceptable cost to society UM, and how we

126
00:08:00,240 --> 00:08:04,160
Speaker 1: track and measure that it correct me if I'm wrong here.

127
00:08:04,280 --> 00:08:09,920
Speaker 1: My impression is that climate risk manifests itself in markets

128
00:08:09,960 --> 00:08:12,920
Speaker 1: in all different kinds of ways. What do you see

129
00:08:12,960 --> 00:08:18,000
Speaker 1: in commodities, whether you see it in insurance type of contracts, etcetera.

130
00:08:18,160 --> 00:08:20,960
Speaker 1: But is it the case that the what's being priced

131
00:08:21,120 --> 00:08:26,000
Speaker 1: is not the climate risk itself, but rather the regulatory

132
00:08:26,040 --> 00:08:31,760
Speaker 1: expectations or current policies of regulatory response towards climate change.

133
00:08:32,040 --> 00:08:34,960
Speaker 1: So like, imagine an environment in which there was no

134
00:08:35,080 --> 00:08:38,360
Speaker 1: climate science really or there was no will to action

135
00:08:38,480 --> 00:08:42,079
Speaker 1: at all to address climate would it be showing up?

136
00:08:42,320 --> 00:08:46,320
Speaker 1: Would we be seeing stresses regardless? So I think this

137
00:08:46,400 --> 00:08:49,160
Speaker 1: brings up a really important point, which is climate change

138
00:08:49,200 --> 00:08:52,080
Speaker 1: is the most important and obvious example of a real

139
00:08:52,120 --> 00:08:56,600
Speaker 1: market failure. So capital markets and markets in general are

140
00:08:56,640 --> 00:08:59,600
Speaker 1: are just really bad at long term planning, and long

141
00:08:59,720 --> 00:09:02,280
Speaker 1: term into case means fifty to a hundred years um.

142
00:09:02,440 --> 00:09:06,080
Speaker 1: So uh, it means that the risk price into markets

143
00:09:06,120 --> 00:09:08,000
Speaker 1: is much more along the lines of what you just described,

144
00:09:08,000 --> 00:09:11,000
Speaker 1: which is how is the government going to react today?

145
00:09:11,200 --> 00:09:14,120
Speaker 1: Rather than what is the net impact considering potential action,

146
00:09:15,280 --> 00:09:17,200
Speaker 1: whether or not it's enough, and if it's more than

147
00:09:17,320 --> 00:09:19,559
Speaker 1: enough or less than enough, like what's the net impact,

148
00:09:19,600 --> 00:09:24,000
Speaker 1: Like what's the real what's the real forecast? So, um,

149
00:09:24,040 --> 00:09:27,199
Speaker 1: there's a world in which this is actually a somewhat

150
00:09:27,200 --> 00:09:30,520
Speaker 1: exogenous process to the way markets think about risk um

151
00:09:30,559 --> 00:09:32,480
Speaker 1: and and that just because again it doesn't get really

152
00:09:32,480 --> 00:09:34,959
Speaker 1: priced in. This is a big part of the stern

153
00:09:34,960 --> 00:09:39,680
Speaker 1: instiguilers critique, which is, you know, market pricing is based

154
00:09:39,760 --> 00:09:44,160
Speaker 1: on individual incentives that are bay that are themselves the

155
00:09:44,160 --> 00:09:47,520
Speaker 1: product of like a single generation of people. And so

156
00:09:47,559 --> 00:09:50,160
Speaker 1: we're thinking about what is my preference as an individual

157
00:09:50,160 --> 00:09:53,600
Speaker 1: investor over the next fifty years as opposed to what

158
00:09:53,679 --> 00:09:58,640
Speaker 1: is the societal aggregate preference in an intergenerational context, which

159
00:09:58,679 --> 00:10:02,120
Speaker 1: prioritizes to the same extent the welfare of people who

160
00:10:02,120 --> 00:10:05,000
Speaker 1: are living a hundred years from now relative to my own.

161
00:10:05,040 --> 00:10:07,520
Speaker 1: And so markets are really bad at that UM, and

162
00:10:08,240 --> 00:10:10,720
Speaker 1: that's been a general critique of the way that climate

163
00:10:11,240 --> 00:10:14,320
Speaker 1: risk has been modeled. Um, it doesn't mean that markets

164
00:10:14,360 --> 00:10:17,080
Speaker 1: can't tell you something about kind of the base case.

165
00:10:17,120 --> 00:10:19,400
Speaker 1: So what we're doing in this paper, I think is

166
00:10:19,760 --> 00:10:23,120
Speaker 1: in saying markets because they don't anticipate these types of

167
00:10:23,120 --> 00:10:26,000
Speaker 1: existential and long term risks, they're kind of giving you

168
00:10:26,040 --> 00:10:29,760
Speaker 1: a sense of the baseline rather than the the climate

169
00:10:29,840 --> 00:10:32,240
Speaker 1: change scenario. So they're giving a sense of how the

170
00:10:32,320 --> 00:10:36,920
Speaker 1: market would perceive risk and pricing and time preference in

171
00:10:36,960 --> 00:10:39,200
Speaker 1: the absence of climate change to some extent, because it's

172
00:10:39,280 --> 00:10:42,200
Speaker 1: it's not going to get efficiently priced, and that's a

173
00:10:42,400 --> 00:10:46,559
Speaker 1: problem because markets are not anticipating this like extremely important

174
00:10:47,040 --> 00:10:51,840
Speaker 1: risk factor, arguably the most important risk factor fundamentally. But

175
00:10:51,920 --> 00:10:55,120
Speaker 1: it's convenient in the sense that you can extract from

176
00:10:55,200 --> 00:10:59,480
Speaker 1: markets um this kind of baseline, time preference and other

177
00:10:59,559 --> 00:11:02,120
Speaker 1: fact in other variables that that lets you put that

178
00:11:02,240 --> 00:11:05,640
Speaker 1: price on carbon ultimately to to to get a sense

179
00:11:05,679 --> 00:11:08,719
Speaker 1: of what the cost benefit looks like, because to get

180
00:11:08,720 --> 00:11:11,640
Speaker 1: a cost benefit you need to have with and without

181
00:11:11,679 --> 00:11:13,920
Speaker 1: like they could do a cost benefit. It's what I

182
00:11:13,960 --> 00:11:16,440
Speaker 1: need to know is what's the world like without climate change,

183
00:11:16,440 --> 00:11:18,240
Speaker 1: and what's the world like with climate change? And how

184
00:11:18,280 --> 00:11:22,120
Speaker 1: do I mitigate that impact? And because climate change is

185
00:11:22,160 --> 00:11:25,800
Speaker 1: an indogenous process to the real world, but but arguably

186
00:11:25,800 --> 00:11:29,720
Speaker 1: in an exogenous process to the market perspective on the

187
00:11:29,720 --> 00:11:33,960
Speaker 1: real world, you can kind of do that somewhat efficiently. So,

188
00:11:34,320 --> 00:11:37,520
Speaker 1: I mean you mentioned the time preference of investors. There,

189
00:11:37,559 --> 00:11:39,880
Speaker 1: can we talk a little bit more about that, because

190
00:11:40,280 --> 00:11:42,440
Speaker 1: the point that you're making about yields is that we're

191
00:11:42,440 --> 00:11:45,240
Speaker 1: basically talking about the time value of money, and so

192
00:11:45,360 --> 00:11:49,200
Speaker 1: they're going to factor into any big action that you

193
00:11:49,240 --> 00:11:52,800
Speaker 1: actually undertake to fight climate change because it costs money,

194
00:11:52,880 --> 00:11:56,240
Speaker 1: and you have to make that calculation as an investor

195
00:11:56,240 --> 00:11:59,240
Speaker 1: how much your money is worth going into a climate

196
00:11:59,320 --> 00:12:02,880
Speaker 1: change initiative versus something else. So how are you thinking

197
00:12:02,920 --> 00:12:05,600
Speaker 1: about the discount rate and like how much does that

198
00:12:05,720 --> 00:12:11,600
Speaker 1: traditionally factor into cost benefit analyses for climate change? Yea,

199
00:12:11,679 --> 00:12:13,880
Speaker 1: So maybe just taking a step back, like why are

200
00:12:13,880 --> 00:12:17,280
Speaker 1: we discounting these impacts at all? Climate change costs? I

201
00:12:17,280 --> 00:12:19,440
Speaker 1: think the numbers like a quadrillion dollars over the next

202
00:12:19,600 --> 00:12:22,440
Speaker 1: hundred years. Why are we not just using that? Why?

203
00:12:22,480 --> 00:12:25,720
Speaker 1: Why why are we using this discounting effect? And from

204
00:12:25,720 --> 00:12:30,760
Speaker 1: a from a first principal standpoint, interest rates reflect time

205
00:12:30,760 --> 00:12:32,760
Speaker 1: preference in the sense that I can use my hundred

206
00:12:32,760 --> 00:12:35,600
Speaker 1: dollars today to go buy something, or I can use

207
00:12:35,640 --> 00:12:38,840
Speaker 1: it in a year, and if I'm going to delay

208
00:12:38,880 --> 00:12:42,080
Speaker 1: that consumption by a year. Then there's some value to

209
00:12:42,120 --> 00:12:45,440
Speaker 1: me in that right. There there's some there's some there's

210
00:12:45,480 --> 00:12:49,679
Speaker 1: some cost um in giving up that year of access

211
00:12:49,720 --> 00:12:51,640
Speaker 1: to my money, and so that's the interest rate I'm

212
00:12:51,640 --> 00:12:53,599
Speaker 1: going to charge for it. So if interest rates to

213
00:12:53,640 --> 00:12:56,840
Speaker 1: one percent, it means that the time preference reflects a

214
00:12:56,880 --> 00:13:00,800
Speaker 1: one percent annualized cost to delay in can some ship uh,

215
00:13:00,880 --> 00:13:03,360
Speaker 1: interest rates are observable, like we have many interest rate

216
00:13:04,120 --> 00:13:06,800
Speaker 1: driven instruments. Treasury bonds are are kind of a good

217
00:13:06,800 --> 00:13:10,680
Speaker 1: example of that. There's also a rich and deep derivatives

218
00:13:10,679 --> 00:13:13,960
Speaker 1: market which is tied to other interest rates, and we

219
00:13:14,000 --> 00:13:17,120
Speaker 1: can use that to infer time preference in that sense.

220
00:13:17,520 --> 00:13:20,720
Speaker 1: So let's say we have an observation of time preference UM,

221
00:13:20,960 --> 00:13:24,920
Speaker 1: and we've soolidated that into some running rate of interest

222
00:13:25,679 --> 00:13:29,439
Speaker 1: two percent three adjusted for inflation, because most of these

223
00:13:29,520 --> 00:13:33,280
Speaker 1: estimates are done in inflation adjusted terms UM. So then

224
00:13:34,120 --> 00:13:36,079
Speaker 1: what we can do is say, well, climate change is

225
00:13:36,080 --> 00:13:38,560
Speaker 1: going to cost a certain amount of money every year

226
00:13:38,720 --> 00:13:42,320
Speaker 1: for the next hundred years. Because that money in the

227
00:13:42,320 --> 00:13:45,120
Speaker 1: future is worth less or we're gonn actually talk about

228
00:13:45,120 --> 00:13:47,839
Speaker 1: it probably after this but potentially more, but it's worth

229
00:13:47,880 --> 00:13:50,200
Speaker 1: a different amount in the future. Than it is today,

230
00:13:50,760 --> 00:13:55,720
Speaker 1: than the discounted future value of those damages is what

231
00:13:55,760 --> 00:13:57,559
Speaker 1: I should be willing to pay today to avoid them.

232
00:13:58,000 --> 00:14:00,040
Speaker 1: And so that's the social cost of carbon, which is

233
00:14:00,080 --> 00:14:03,800
Speaker 1: an incremental metric ton of carbon dioxide. What is the

234
00:14:03,840 --> 00:14:07,440
Speaker 1: economic impact into the future, and what is the discounted

235
00:14:07,520 --> 00:14:11,320
Speaker 1: value of that damage today? And if I'm thinking in

236
00:14:11,400 --> 00:14:15,240
Speaker 1: terms of pure economics, the moral angles is important but

237
00:14:15,320 --> 00:14:18,960
Speaker 1: not necessarily part of that analysis, then I should be

238
00:14:18,960 --> 00:14:21,600
Speaker 1: willing to pay up to that discounted amount to to

239
00:14:21,680 --> 00:14:24,760
Speaker 1: avoid negative outcomes. I guess put a different way. We

240
00:14:24,840 --> 00:14:27,520
Speaker 1: usually talk about the time value money as how much

241
00:14:27,560 --> 00:14:29,560
Speaker 1: is the money worth to me today versus in a year?

242
00:14:29,640 --> 00:14:32,000
Speaker 1: But you can also think of it equivalently as how

243
00:14:32,040 --> 00:14:33,640
Speaker 1: much is it worked to me to avoid a loss

244
00:14:33,880 --> 00:14:36,800
Speaker 1: in a year versus today? Those are it's it's symmetric

245
00:14:36,840 --> 00:14:40,280
Speaker 1: in that respect. So what is thinking about this framework

246
00:14:40,600 --> 00:14:43,320
Speaker 1: get you? Like? Okay, so you work through this, you

247
00:14:43,400 --> 00:14:48,680
Speaker 1: like calculate the theoretical damages do this sort of this

248
00:14:48,840 --> 00:14:52,160
Speaker 1: time value analysis? Where does it get you? And what

249
00:14:52,280 --> 00:14:55,560
Speaker 1: does it tell you in terms of carbon pricing or

250
00:14:55,640 --> 00:14:59,160
Speaker 1: connecting it to uh to your specialty like rates? Like

251
00:14:59,440 --> 00:15:02,640
Speaker 1: where does what does the analysis say, yeah, I think

252
00:15:02,640 --> 00:15:05,600
Speaker 1: it's important to keep in mind that models are flawed.

253
00:15:06,360 --> 00:15:09,040
Speaker 1: These models are very flawed, But we have to make

254
00:15:09,080 --> 00:15:12,160
Speaker 1: a decision. We don't have the option to not make

255
00:15:12,160 --> 00:15:14,360
Speaker 1: a call on this because climate change is going to

256
00:15:14,440 --> 00:15:16,720
Speaker 1: happen one way or the other and we need to

257
00:15:16,720 --> 00:15:19,280
Speaker 1: take some steps to mitigate it. But we're lots of

258
00:15:19,280 --> 00:15:23,520
Speaker 1: steps to mitigate it. But if we're unwilling to de industrialize,

259
00:15:23,520 --> 00:15:26,960
Speaker 1: which is kind of the the ultimate and only complete solution,

260
00:15:27,560 --> 00:15:30,080
Speaker 1: then we need to come up with what we're willing

261
00:15:30,160 --> 00:15:32,560
Speaker 1: to do, and so this analysis is an attempt to

262
00:15:32,560 --> 00:15:37,520
Speaker 1: do that, and we're basically making our best guests as

263
00:15:37,560 --> 00:15:39,440
Speaker 1: to how much we should we should be willing to

264
00:15:39,480 --> 00:15:43,280
Speaker 1: spend to avoid damage down the line. And it's convenient

265
00:15:43,320 --> 00:15:46,040
Speaker 1: as a framework UM in two ways. One is it's

266
00:15:46,240 --> 00:15:49,280
Speaker 1: it really does get you to policy, right. It gets

267
00:15:49,280 --> 00:15:51,560
Speaker 1: you to a set of policies that the cost of

268
00:15:51,600 --> 00:15:54,720
Speaker 1: which is immediate, that you can at least justify on

269
00:15:54,760 --> 00:15:56,840
Speaker 1: the basis of some framework. Right. We need to come

270
00:15:56,920 --> 00:15:58,600
Speaker 1: up with a reason why it's worth spending a trillion

271
00:15:58,600 --> 00:16:02,120
Speaker 1: dollars today, um, because everyone who is contributing those tax

272
00:16:02,120 --> 00:16:04,520
Speaker 1: always want to know why a trillion? Why not to

273
00:16:04,720 --> 00:16:07,960
Speaker 1: why not half? Where did this number come from? Um

274
00:16:08,120 --> 00:16:11,040
Speaker 1: flawed though those models maybe they get us to that

275
00:16:11,400 --> 00:16:15,040
Speaker 1: and two is at least from a more near term standpoint.

276
00:16:15,200 --> 00:16:18,760
Speaker 1: From an observers standpoint, we get a sense of how

277
00:16:18,800 --> 00:16:21,480
Speaker 1: the government is likely to react, so we're not what

278
00:16:21,560 --> 00:16:24,360
Speaker 1: we're focused a little bit on the policy making process,

279
00:16:24,360 --> 00:16:27,680
Speaker 1: but there's also the more near term impact on how

280
00:16:27,760 --> 00:16:30,560
Speaker 1: government policy evolves. And to the extent that this framework,

281
00:16:30,600 --> 00:16:34,840
Speaker 1: this approach is central to the government's policy making process.

282
00:16:35,000 --> 00:16:38,280
Speaker 1: It lets us kind of think like they think and

283
00:16:38,320 --> 00:16:40,600
Speaker 1: come up with a sense of what the likely range

284
00:16:40,600 --> 00:16:43,560
Speaker 1: of outcomes is. And the Biden administration has been very

285
00:16:43,600 --> 00:16:46,640
Speaker 1: clear that the social cost of carbon is the single

286
00:16:46,640 --> 00:16:50,160
Speaker 1: most important number in their climate change agenda. So you know,

287
00:16:50,760 --> 00:16:52,560
Speaker 1: we should we should get a better sense of how

288
00:16:52,600 --> 00:16:55,960
Speaker 1: to how to get behind a value of that and

289
00:16:56,000 --> 00:16:59,640
Speaker 1: what the right set of assumptions are than rather than

290
00:16:59,720 --> 00:17:02,360
Speaker 1: sort of necessarily arguing about the utility of that number

291
00:17:02,360 --> 00:17:05,119
Speaker 1: relative to other numbers. The last thing I'd say is

292
00:17:05,600 --> 00:17:09,359
Speaker 1: this discounting effect. Um it is possible. It's kind of

293
00:17:09,359 --> 00:17:12,119
Speaker 1: two approaches to this. What we've focused on here is

294
00:17:12,119 --> 00:17:15,160
Speaker 1: a market driven discount rate, which is really measuring pure

295
00:17:15,200 --> 00:17:18,880
Speaker 1: time preference, and and that's extracted from in this case

296
00:17:18,920 --> 00:17:21,240
Speaker 1: derivative markets, but you can get it from treasuries or

297
00:17:21,280 --> 00:17:24,560
Speaker 1: other places. UM. That's been pretty standard practice for a

298
00:17:24,640 --> 00:17:27,600
Speaker 1: very long time. A common critique of that approach is

299
00:17:27,640 --> 00:17:31,359
Speaker 1: that it prioritizes the sort of talked about it a

300
00:17:31,359 --> 00:17:35,040
Speaker 1: second ago, you know, the current generation's individual perceptions of

301
00:17:35,080 --> 00:17:39,760
Speaker 1: time preference, rather than societal aggregate UM intergenerational perceptions of

302
00:17:40,040 --> 00:17:43,280
Speaker 1: time preference. And it's explicitly a moral because it's just

303
00:17:44,040 --> 00:17:46,200
Speaker 1: a number extracted from markets. And so what you can

304
00:17:46,200 --> 00:17:48,720
Speaker 1: do is you can use that discount rate and you

305
00:17:48,760 --> 00:17:52,680
Speaker 1: can take a more prescriptive approach and normative approach, where

306
00:17:52,720 --> 00:17:54,840
Speaker 1: you argue for what the right discount rate should be,

307
00:17:55,920 --> 00:18:02,840
Speaker 1: incorporating factors like like inequality betweenations and within nations, incorporating

308
00:18:03,240 --> 00:18:05,920
Speaker 1: the interaction between different countries. You can incorporate the risk

309
00:18:05,960 --> 00:18:08,439
Speaker 1: of extinction. People have done that, UM, and so you

310
00:18:08,440 --> 00:18:14,080
Speaker 1: can use that to quantify again the right number, and

311
00:18:14,080 --> 00:18:16,399
Speaker 1: and that becomes in some sense the thing you argue about.

312
00:18:16,480 --> 00:18:18,639
Speaker 1: So there, on the one hand, there is the modeling process,

313
00:18:18,640 --> 00:18:21,800
Speaker 1: which has its own uncertainties and legitimate critiques of that

314
00:18:21,840 --> 00:18:24,199
Speaker 1: but we have to come up with something. And on

315
00:18:24,240 --> 00:18:26,280
Speaker 1: the other you can sort of say, well, well, the

316
00:18:26,280 --> 00:18:28,520
Speaker 1: market saying the discount rate should be one percent, but

317
00:18:28,520 --> 00:18:30,480
Speaker 1: I think it should be negative ten percent because I'm

318
00:18:30,600 --> 00:18:32,960
Speaker 1: highly risk averse when it comes to these things, and

319
00:18:33,000 --> 00:18:35,680
Speaker 1: you can make a normative case for that, and therefore

320
00:18:35,680 --> 00:18:38,919
Speaker 1: it's worth spending more money than than than another discount

321
00:18:39,000 --> 00:18:42,080
Speaker 1: RaSE but potentially imply so it sort of consolidates that

322
00:18:42,080 --> 00:18:46,040
Speaker 1: whole argument to a single number, which is convenient. It's reductive,

323
00:18:46,119 --> 00:18:49,400
Speaker 1: but it's it's convenient for for having this conversation. So

324
00:18:49,600 --> 00:18:53,800
Speaker 1: is the implication here that whatever you decide as the

325
00:18:53,840 --> 00:18:58,360
Speaker 1: discount factor is going to be like as much, it's

326
00:18:58,400 --> 00:19:01,320
Speaker 1: going to be as important as an input into your

327
00:19:01,359 --> 00:19:06,040
Speaker 1: model as your climate change projections and things like that. Like,

328
00:19:06,440 --> 00:19:12,080
Speaker 1: is the discount factor perhaps an underappreciated input into our

329
00:19:12,119 --> 00:19:18,160
Speaker 1: models of climate change and cost benefit analysis? Yeah, it's

330
00:19:18,200 --> 00:19:21,560
Speaker 1: massively important because of the long time frames involved. So

331
00:19:21,600 --> 00:19:24,679
Speaker 1: if you have a one percent annualized discount factor and

332
00:19:24,760 --> 00:19:27,640
Speaker 1: you compound that over a hundred years, you're talking about

333
00:19:27,640 --> 00:19:32,120
Speaker 1: a very large effect. The the thing that I think

334
00:19:32,119 --> 00:19:34,040
Speaker 1: we wanted to highlight in this which is in the

335
00:19:34,200 --> 00:19:37,959
Speaker 1: past there was this debate, which is, normative discount factors

336
00:19:37,960 --> 00:19:41,639
Speaker 1: tend to result in lower numbers, market based discount factors

337
00:19:41,680 --> 00:19:45,199
Speaker 1: at the time tended to result in higher numbers. So

338
00:19:45,640 --> 00:19:47,560
Speaker 1: if you go back to two thousand seven, when the

339
00:19:47,560 --> 00:19:51,000
Speaker 1: Stern report came out, the real rate of interest over

340
00:19:51,040 --> 00:19:55,000
Speaker 1: thirty years was something like three And because you're compounding

341
00:19:55,040 --> 00:19:59,199
Speaker 1: this over long periods, I forget the precise arithmetic is

342
00:19:59,240 --> 00:20:01,480
Speaker 1: but you know, a hundred dollars in in fifty years

343
00:20:01,480 --> 00:20:03,720
Speaker 1: at a three percent discount rate, is is worth a

344
00:20:03,720 --> 00:20:07,080
Speaker 1: couple of bucks? Like it? It has a very strong

345
00:20:07,160 --> 00:20:10,639
Speaker 1: effect on the present value of those damages. And if

346
00:20:10,640 --> 00:20:12,320
Speaker 1: you move that to a one percent discount of factor,

347
00:20:12,359 --> 00:20:14,280
Speaker 1: all of a sudden you've got twice as much present

348
00:20:14,320 --> 00:20:17,040
Speaker 1: valure damages or three times as much of a hundred years.

349
00:20:17,080 --> 00:20:19,840
Speaker 1: So the debate was usually that markets are applying a

350
00:20:19,920 --> 00:20:26,320
Speaker 1: higher discount rate then many economists and philosophers, frankly I

351
00:20:26,440 --> 00:20:28,800
Speaker 1: thought was appropriate to the problem. And for all the

352
00:20:28,800 --> 00:20:33,560
Speaker 1: reasons I mentioned. What's happened since then is the long

353
00:20:33,680 --> 00:20:36,360
Speaker 1: term real rate of interest has come down a lot,

354
00:20:36,440 --> 00:20:40,040
Speaker 1: and that's secular stagnation. That's to some extent an endogenous effect.

355
00:20:40,080 --> 00:20:42,320
Speaker 1: In that there could be some climate risk priced into

356
00:20:42,840 --> 00:20:47,240
Speaker 1: that secular stagnation expectation. But the important observation is that

357
00:20:47,640 --> 00:20:50,960
Speaker 1: this argument over whether or not normative discount factors, which

358
00:20:50,960 --> 00:20:53,520
Speaker 1: are relatively low. I think one percent is the is

359
00:20:53,560 --> 00:20:57,120
Speaker 1: the mean of some surveys of economists. Now the real

360
00:20:57,200 --> 00:21:00,600
Speaker 1: rate of interest over fifty years is negative half percent

361
00:21:00,840 --> 00:21:04,479
Speaker 1: or something like that. Like the market versus normative approach

362
00:21:04,480 --> 00:21:07,760
Speaker 1: has flipped in that markets are pricing negative time preference

363
00:21:07,960 --> 00:21:11,479
Speaker 1: over long periods. And when you flip from a negative

364
00:21:11,720 --> 00:21:13,760
Speaker 1: from a positive interest rate to a negative interest rate,

365
00:21:14,160 --> 00:21:17,760
Speaker 1: all of a sudden time inflates the value damages rather

366
00:21:17,800 --> 00:21:20,040
Speaker 1: than discounting them. So discounting is really not the right

367
00:21:20,080 --> 00:21:23,400
Speaker 1: word in a world of negative real rates. And so

368
00:21:23,480 --> 00:21:25,960
Speaker 1: now when when when we look at a hundred years,

369
00:21:26,240 --> 00:21:29,240
Speaker 1: that hundred dollars of of of nominal damage is worth

370
00:21:29,359 --> 00:21:34,080
Speaker 1: three hundred dollars, and so it really it changes the

371
00:21:34,080 --> 00:21:39,800
Speaker 1: way you think about the time distance to these effects.

372
00:21:39,840 --> 00:21:44,000
Speaker 1: So I mean, is the simple is the simple upshot?

373
00:21:44,160 --> 00:21:49,439
Speaker 1: Then that in a period of ultra low real interest rates,

374
00:21:49,480 --> 00:21:52,879
Speaker 1: then setting us that climate people look at a period

375
00:21:52,920 --> 00:21:56,280
Speaker 1: of ultra low interest rates and also a period of

376
00:21:56,800 --> 00:21:59,719
Speaker 1: uh secular stagnation, and they say, Okay, well, then this

377
00:21:59,840 --> 00:22:03,480
Speaker 1: is like a natural period for the government to spend

378
00:22:03,520 --> 00:22:06,920
Speaker 1: a lot of money, invest a lot in research. Generally,

379
00:22:07,119 --> 00:22:11,800
Speaker 1: perhaps um introduce such a sort of spending impulse and

380
00:22:11,920 --> 00:22:15,840
Speaker 1: capital expenditure impulse to break us out of this secular stagnation.

381
00:22:16,680 --> 00:22:20,920
Speaker 1: Is the implication then that that just also applies even

382
00:22:20,920 --> 00:22:23,600
Speaker 1: more so when thinking about addressing the climate threat in

383
00:22:23,640 --> 00:22:28,520
Speaker 1: this economic environment. Yeah, it means that the self corrective

384
00:22:29,520 --> 00:22:32,680
Speaker 1: elements of economic growth just doesn't really apply. So if

385
00:22:32,680 --> 00:22:35,240
Speaker 1: we have negative real interest rates over long periods of time,

386
00:22:35,680 --> 00:22:38,360
Speaker 1: that means that we have basically an expectation of negative

387
00:22:38,800 --> 00:22:43,320
Speaker 1: real growth rates in in equilibrium over long periods of time.

388
00:22:43,920 --> 00:22:47,120
Speaker 1: And so you can sort of like the narrative version

389
00:22:47,160 --> 00:22:49,360
Speaker 1: of this argument is climate change is going to cause

390
00:22:49,359 --> 00:22:52,159
Speaker 1: a lot of damage, but the but the cure is

391
00:22:52,160 --> 00:22:53,959
Speaker 1: worse than the disease, and we're going to spend too

392
00:22:54,000 --> 00:22:57,560
Speaker 1: much money today and do too much economic damage today. Um,

393
00:22:57,640 --> 00:22:59,440
Speaker 1: so we're better off growing our way out of these

394
00:22:59,480 --> 00:23:01,800
Speaker 1: issues like that. That would be one line of argument,

395
00:23:01,800 --> 00:23:04,480
Speaker 1: which is we want to let the economy grow because

396
00:23:04,480 --> 00:23:07,439
Speaker 1: it will we will all be better off over the

397
00:23:07,480 --> 00:23:10,440
Speaker 1: long run. Even in the context of climate change related damage.

398
00:23:10,800 --> 00:23:14,679
Speaker 1: If we allow for for unfettered economic growth and and

399
00:23:14,880 --> 00:23:19,359
Speaker 1: the secular stagnation hypothesis and the consistency of markets with

400
00:23:19,400 --> 00:23:23,280
Speaker 1: that expectation says that's just not true anymore, Like you're

401
00:23:23,320 --> 00:23:26,600
Speaker 1: better off fixing problems today because the longer you wait,

402
00:23:26,720 --> 00:23:29,800
Speaker 1: the bigger the problem is going to be. And um,

403
00:23:29,840 --> 00:23:33,480
Speaker 1: that was not true ten years ago, at least an expectation. Now.

404
00:23:34,000 --> 00:23:36,520
Speaker 1: Secular stagnation, I should say, has been around for a

405
00:23:36,520 --> 00:23:39,560
Speaker 1: long time as a concept. It originates from the late thirties.

406
00:23:39,960 --> 00:23:44,640
Speaker 1: So basically after every major every major recession, there's been

407
00:23:44,680 --> 00:23:49,520
Speaker 1: this wave of secular stagnation prognosticating It kind of reminds

408
00:23:49,560 --> 00:23:52,680
Speaker 1: me of the you know, the kids today, like all

409
00:23:52,720 --> 00:23:54,440
Speaker 1: these kids in their phones, like they're never going to

410
00:23:54,560 --> 00:23:56,879
Speaker 1: learn how to read. Um. But it turns out that

411
00:23:57,000 --> 00:23:59,800
Speaker 1: in the midsu there are a bunch of people who

412
00:23:59,800 --> 00:24:01,760
Speaker 1: are because of the printing press now would learn how

413
00:24:01,800 --> 00:24:03,920
Speaker 1: to write books have I hand anymore. So there's always

414
00:24:04,000 --> 00:24:06,919
Speaker 1: like this this handwringing in the in the in the

415
00:24:06,960 --> 00:24:10,600
Speaker 1: in the presence of change. But this time might actually

416
00:24:10,600 --> 00:24:13,400
Speaker 1: be different. And Larry Summers is obviously sort of associated

417
00:24:13,440 --> 00:24:17,360
Speaker 1: with this argument. Um. But the combination of demographics um,

418
00:24:17,480 --> 00:24:21,320
Speaker 1: the digital transformation, a global savings glad, the rise of

419
00:24:21,760 --> 00:24:25,880
Speaker 1: Asia like all of these, all the globalization trend which

420
00:24:25,920 --> 00:24:28,880
Speaker 1: is still in place. Um. You know, all of these

421
00:24:28,920 --> 00:24:32,720
Speaker 1: things point to lower at least directionally real growth rates

422
00:24:32,760 --> 00:24:36,720
Speaker 1: over time. And and that just means again directionally the

423
00:24:36,760 --> 00:24:39,159
Speaker 1: cost of waiting, the cost of an action is higher.

424
00:24:39,480 --> 00:24:42,200
Speaker 1: And we're using markets because they do a pretty good

425
00:24:42,240 --> 00:24:44,680
Speaker 1: job of finding stuff like that, Like markets are better

426
00:24:44,720 --> 00:24:50,440
Speaker 1: at identifying regime change um than than we are individually.

427
00:24:50,880 --> 00:24:53,639
Speaker 1: And you can see that just because you know the

428
00:24:54,200 --> 00:24:58,440
Speaker 1: secular stagnation hypothesis and the implications for long term interest rates,

429
00:24:58,480 --> 00:25:01,840
Speaker 1: like the growth rates that that's been reflected to some

430
00:25:01,960 --> 00:25:04,720
Speaker 1: extent in surveys of economists, for example. But the transition

431
00:25:04,760 --> 00:25:08,119
Speaker 1: has been very gradual, whereas markets kind of price this

432
00:25:08,160 --> 00:25:11,080
Speaker 1: in in two thousand eight and have largely stayed in

433
00:25:11,119 --> 00:25:14,119
Speaker 1: that range. So um, you know, markets have done a

434
00:25:14,160 --> 00:25:17,560
Speaker 1: decent job at least at a high level of never

435
00:25:17,600 --> 00:25:20,880
Speaker 1: mind the next ten fifteen basis points. You know, categorically

436
00:25:20,920 --> 00:25:23,000
Speaker 1: they're going to be done a good job of spotting

437
00:25:23,040 --> 00:25:26,080
Speaker 1: the shift in the way the world grows and if

438
00:25:26,080 --> 00:25:29,520
Speaker 1: we're shifting to a load of negative growth environment in

439
00:25:29,560 --> 00:25:32,879
Speaker 1: the lead up to the really damaging impact economically of

440
00:25:32,920 --> 00:25:35,960
Speaker 1: climate change. UM, there's a much greater incentive to act now,

441
00:25:35,960 --> 00:25:38,200
Speaker 1: and that's reflected in in the in the social cost

442
00:25:38,200 --> 00:25:40,800
Speaker 1: of carbon like if when you when you calculate even

443
00:25:40,840 --> 00:25:45,880
Speaker 1: taking damages in in sort of local um nominal terms

444
00:25:45,920 --> 00:25:48,479
Speaker 1: as as a given, like let's not argue about how

445
00:25:48,480 --> 00:25:50,800
Speaker 1: we're calculating the dollar value of those damages. That's the

446
00:25:50,880 --> 00:25:54,040
Speaker 1: separate conversation. Even if you take that as a given,

447
00:25:54,280 --> 00:25:56,800
Speaker 1: just the way the environment and expectations have evolved means

448
00:25:56,800 --> 00:26:00,240
Speaker 1: the social cost has gone up six seven times of

449
00:26:00,320 --> 00:26:03,480
Speaker 1: the past ten years, and that really changes your cost benefit.

450
00:26:04,600 --> 00:26:06,639
Speaker 1: That changes the way which you should be willing to

451
00:26:06,680 --> 00:26:10,200
Speaker 1: spend to avoid it. So on that note, I mean

452
00:26:10,320 --> 00:26:13,960
Speaker 1: I was reading a City group note by UM one

453
00:26:14,000 --> 00:26:17,320
Speaker 1: of their strategists recently, Matt King, and apparently City has

454
00:26:17,359 --> 00:26:20,920
Speaker 1: been having a bunch of client calls with Larry Summers,

455
00:26:20,960 --> 00:26:26,640
Speaker 1: and Matt King was strongly UM suggesting that Larry Summers

456
00:26:26,640 --> 00:26:30,600
Speaker 1: had sort of backed off the secular stagnation idea because

457
00:26:30,600 --> 00:26:33,359
Speaker 1: of the recent backup in yields, which seems like I

458
00:26:33,359 --> 00:26:35,199
Speaker 1: don't know a bit of a knee jerk reaction, but

459
00:26:35,760 --> 00:26:40,640
Speaker 1: I'm curious this idea that secular stechnation has multiplied the

460
00:26:40,760 --> 00:26:45,520
Speaker 1: cost of climate change in action. How well does that

461
00:26:45,600 --> 00:26:49,480
Speaker 1: stand up to the recent changes in the treasury market,

462
00:26:49,800 --> 00:26:55,120
Speaker 1: the volatility that we've seen, and UM, the increase in yields. Yeah,

463
00:26:55,200 --> 00:26:58,520
Speaker 1: so most of the move in yields has been an

464
00:26:58,520 --> 00:27:02,640
Speaker 1: inflation expectations. So we're really focused for this purpose on

465
00:27:02,800 --> 00:27:06,520
Speaker 1: the real rate of interest, and over thirty to fifty years,

466
00:27:06,560 --> 00:27:08,800
Speaker 1: the real rate of interest is still negative at least

467
00:27:08,800 --> 00:27:11,000
Speaker 1: if and there's some argument here I should I use

468
00:27:11,040 --> 00:27:14,360
Speaker 1: treasury rate, shall use tips? Shall I use derivative markets?

469
00:27:14,400 --> 00:27:16,919
Speaker 1: And what we do here is we look at expectations

470
00:27:17,400 --> 00:27:20,760
Speaker 1: for the federal funds rate the FEDS target rate over

471
00:27:20,800 --> 00:27:24,760
Speaker 1: fifty years, for example, UM and adjust that for inflation expectations.

472
00:27:24,800 --> 00:27:27,280
Speaker 1: And the reason we do that is UM. When you're

473
00:27:27,280 --> 00:27:30,520
Speaker 1: dealing with securities, and we talked about this back last March,

474
00:27:31,000 --> 00:27:34,320
Speaker 1: securities have sort of different treatment under different circumstances. It's

475
00:27:34,320 --> 00:27:37,960
Speaker 1: harder to hold a bond than a derivative, especially for

476
00:27:38,000 --> 00:27:40,520
Speaker 1: a bank, so you're sort of baking in some of

477
00:27:40,560 --> 00:27:45,400
Speaker 1: these nuances of balancie costs and things, but derivatives specifically

478
00:27:45,560 --> 00:27:49,000
Speaker 1: are UM are sort of an easier proxy. And so

479
00:27:49,080 --> 00:27:51,400
Speaker 1: when we look at fifty year o I S swap rates,

480
00:27:51,440 --> 00:27:54,320
Speaker 1: which is just expectations again for the federal funds rate

481
00:27:54,359 --> 00:27:57,879
Speaker 1: over the next fifty years, it's still negative adjusted for

482
00:27:57,920 --> 00:28:02,560
Speaker 1: inflation expectations. So UM, directionally, you know, it maybe maybe

483
00:28:02,600 --> 00:28:04,359
Speaker 1: more like four or five times over the past ten

484
00:28:04,440 --> 00:28:07,480
Speaker 1: years now rather than six or seven the volatility and

485
00:28:07,480 --> 00:28:10,320
Speaker 1: the social cost of carbon using market discount rates is

486
00:28:10,680 --> 00:28:14,399
Speaker 1: potentially a critique that makes sense, but but directionally it's

487
00:28:14,440 --> 00:28:18,399
Speaker 1: still negative. So markets are pricing in negative real growth

488
00:28:18,480 --> 00:28:21,480
Speaker 1: rates over very long periods of time. UM. The other

489
00:28:21,520 --> 00:28:23,120
Speaker 1: thing that I think is important to keep in mind

490
00:28:23,200 --> 00:28:27,200
Speaker 1: is because we're talking about fifty hundred years UM, the

491
00:28:27,240 --> 00:28:30,160
Speaker 1: bond market stops at a thirty year instrument, but these

492
00:28:30,240 --> 00:28:35,439
Speaker 1: risks go out much further, and derivatives conveniently trade fifty

493
00:28:35,520 --> 00:28:38,920
Speaker 1: sixty in some cases a hundred years out, and the

494
00:28:39,160 --> 00:28:42,960
Speaker 1: discount rate actually starts to decline after that thirty year

495
00:28:43,000 --> 00:28:47,160
Speaker 1: point UM. And there's some technical reasons for that. UM.

496
00:28:47,200 --> 00:28:50,240
Speaker 1: There's a market flow driven reasons for that, but the

497
00:28:50,280 --> 00:28:52,760
Speaker 1: discount rate over sixty years is less than the discount

498
00:28:52,840 --> 00:28:55,640
Speaker 1: rate over thirty years by this measure, for example, So

499
00:28:56,120 --> 00:28:59,360
Speaker 1: you know there is this You can interpret that as

500
00:29:00,000 --> 00:29:03,040
Speaker 1: graaight risk of of negative real growth rates over longer

501
00:29:03,080 --> 00:29:06,960
Speaker 1: horizons and be sort of consistent with the demographic outlook.

502
00:29:07,400 --> 00:29:11,240
Speaker 1: So that means and that's also consistent with more normative arguments,

503
00:29:11,240 --> 00:29:15,360
Speaker 1: which say, I should to balance the scales like these

504
00:29:15,400 --> 00:29:18,479
Speaker 1: market interest rates are set by the current generation and

505
00:29:18,520 --> 00:29:21,240
Speaker 1: the next generation doesn't get a say so I should

506
00:29:21,280 --> 00:29:24,360
Speaker 1: give them. Uh, I should give the current generation a

507
00:29:24,360 --> 00:29:26,600
Speaker 1: haircut when I think about discount rates, and so over

508
00:29:26,720 --> 00:29:29,880
Speaker 1: sixty years, where I'm spanning a bunch of generations, that

509
00:29:29,920 --> 00:29:33,680
Speaker 1: discount rate should be naturally lower because it's just scoping

510
00:29:33,680 --> 00:29:36,560
Speaker 1: in many more people, many most of which haven't been

511
00:29:36,560 --> 00:29:39,120
Speaker 1: born yet, and so they should get a little bit

512
00:29:39,120 --> 00:29:42,760
Speaker 1: of a leg up in the relative value of their utility.

513
00:29:42,960 --> 00:29:45,760
Speaker 1: And and that naturally emerges from a from a market

514
00:29:45,760 --> 00:29:48,840
Speaker 1: based approach. Um, it's kind of part of bond math.

515
00:29:49,400 --> 00:29:53,320
Speaker 1: Does that extent you know, that's also consistent with with

516
00:29:53,360 --> 00:29:57,560
Speaker 1: those arguments. So yeah, I think the are we in

517
00:29:57,560 --> 00:30:01,840
Speaker 1: a secular stagnation world or not? Is a much more

518
00:30:02,920 --> 00:30:06,760
Speaker 1: extensive conversation in certain ways. But but I think at

519
00:30:06,760 --> 00:30:08,640
Speaker 1: the end of the day, like this, moving treasury rates

520
00:30:09,400 --> 00:30:12,880
Speaker 1: and moving interest rates generally UM is much more notable

521
00:30:12,920 --> 00:30:16,800
Speaker 1: for its speed than its magnitude and really hasn't changed

522
00:30:17,720 --> 00:30:20,479
Speaker 1: the underlying narrative that we can extract from markets over

523
00:30:20,640 --> 00:30:24,720
Speaker 1: long periods of time. And that's why I'm saying, basis

524
00:30:24,720 --> 00:30:29,280
Speaker 1: points between friends, you know, over long periods is really

525
00:30:29,280 --> 00:30:32,560
Speaker 1: not that big a deal. And if we were to

526
00:30:32,640 --> 00:30:35,080
Speaker 1: be having this conversation in the mid nineties or the

527
00:30:35,080 --> 00:30:38,959
Speaker 1: early eighties, we would say that's Tuesday. So you know,

528
00:30:39,240 --> 00:30:41,560
Speaker 1: we we've gotten used to a very low voltility world,

529
00:30:42,040 --> 00:30:46,680
Speaker 1: and the outlook really hasn't changed in a fundamental way

530
00:30:47,080 --> 00:30:50,720
Speaker 1: on the basis of these moves UM, and real rates

531
00:30:50,760 --> 00:30:53,400
Speaker 1: themselves only started to adjust for the past few weeks.

532
00:30:53,440 --> 00:31:14,960
Speaker 1: It's really been mostly an inflation expectations outlook. So I'm

533
00:31:14,960 --> 00:31:18,800
Speaker 1: still having trouble wrapping my head around something like okay

534
00:31:18,880 --> 00:31:24,800
Speaker 1: because of very low real rates expectations, however you measure it,

535
00:31:25,480 --> 00:31:29,640
Speaker 1: the cost of not doing anything about climate change is

536
00:31:29,680 --> 00:31:32,960
Speaker 1: extremely high right now, and that part I get. But like,

537
00:31:33,040 --> 00:31:36,640
Speaker 1: let's say, like we did enact a series of policies

538
00:31:36,960 --> 00:31:41,600
Speaker 1: right now that really did jolt the economy out of

539
00:31:42,040 --> 00:31:45,800
Speaker 1: something that we call secular stagnation, which seems possible. I mean,

540
00:31:46,360 --> 00:31:48,800
Speaker 1: you know, you mentioned the increase in real rate has

541
00:31:48,800 --> 00:31:50,920
Speaker 1: been very modest, but it's possible. You know. It's like,

542
00:31:50,960 --> 00:31:54,240
Speaker 1: if we already get some sort of sustained fiscal stimulus

543
00:31:54,400 --> 00:31:58,800
Speaker 1: and massive capital expenditures and all kinds of different things, like,

544
00:31:58,840 --> 00:32:01,520
Speaker 1: it seems possible that we could have a set of

545
00:32:01,640 --> 00:32:05,840
Speaker 1: policies that would jolt us out of this whatever we've

546
00:32:05,880 --> 00:32:09,720
Speaker 1: been in for the last decade or longer. Does that

547
00:32:09,960 --> 00:32:14,160
Speaker 1: then imply that the cost of inaction on climate like

548
00:32:14,240 --> 00:32:16,360
Speaker 1: goes down? Like that's like I don't quite sort of

549
00:32:16,360 --> 00:32:18,440
Speaker 1: like get that link. It seems to me it's like, Okay,

550
00:32:18,440 --> 00:32:21,480
Speaker 1: there's like there's like disaster waiting for us at some

551
00:32:21,560 --> 00:32:24,240
Speaker 1: point within action, and either we want to do something

552
00:32:24,240 --> 00:32:27,200
Speaker 1: about it or not. But this idea, that's like, Okay,

553
00:32:27,200 --> 00:32:31,280
Speaker 1: if suddenly they started roaring and companies started spending and investing,

554
00:32:31,320 --> 00:32:34,200
Speaker 1: and we saw wage growth and capital expenditure the likes

555
00:32:34,200 --> 00:32:37,360
Speaker 1: of which we haven't seen, what is that? How does

556
00:32:37,400 --> 00:32:41,160
Speaker 1: that then imply that somehow the cost of inaction goes down?

557
00:32:43,000 --> 00:32:44,680
Speaker 1: I think it depends on a couple of things. The

558
00:32:44,760 --> 00:32:48,160
Speaker 1: first is is inflationary or not? Like is wage growth

559
00:32:48,160 --> 00:32:50,960
Speaker 1: with driving things? And if so, then we're not looking

560
00:32:51,000 --> 00:32:54,840
Speaker 1: at a at a rise in real growth rate expectations

561
00:32:54,840 --> 00:32:56,600
Speaker 1: as much as we're looking for a rise in nominal

562
00:32:56,640 --> 00:32:59,960
Speaker 1: grade birth rate expectations. And that means that an inflation

563
00:33:00,000 --> 00:33:02,719
Speaker 1: adjusted world, which is how we do all of this analysis,

564
00:33:02,880 --> 00:33:05,680
Speaker 1: it won't really show up. But you could argue that

565
00:33:06,240 --> 00:33:09,080
Speaker 1: there is a set of policies, whatever they are, that

566
00:33:09,480 --> 00:33:13,760
Speaker 1: quote unquote fix or help the United States transition to

567
00:33:15,280 --> 00:33:20,120
Speaker 1: a more balanced economy, economy with a higher um equilibrium

568
00:33:20,120 --> 00:33:23,880
Speaker 1: growth rate and and sort of mitigate the the not

569
00:33:24,000 --> 00:33:26,680
Speaker 1: just the deflationary but also the sort of negative growth

570
00:33:26,760 --> 00:33:31,600
Speaker 1: impulse of things like digitalization and aging populations and so forth.

571
00:33:31,640 --> 00:33:34,720
Speaker 1: And so you know, the hypo I guess is like,

572
00:33:35,120 --> 00:33:38,000
Speaker 1: if we can design a set of policies that boost

573
00:33:38,040 --> 00:33:41,120
Speaker 1: real growth rates for a long period of time in

574
00:33:41,160 --> 00:33:44,240
Speaker 1: a fundamental way, restructuring the economy is such that we

575
00:33:44,320 --> 00:33:48,640
Speaker 1: get sustained higher real growth rates, Like, should that change

576
00:33:48,680 --> 00:33:51,920
Speaker 1: the way we think about climate change? Especially because in

577
00:33:51,960 --> 00:33:56,560
Speaker 1: doing so we've we've probably not necessarily made the problem worse. Right,

578
00:33:56,960 --> 00:34:00,400
Speaker 1: economic growth in carbon emissions or somewhat correlated. You certainly

579
00:34:00,480 --> 00:34:02,520
Speaker 1: learned that last year, you know. I think there's a

580
00:34:02,640 --> 00:34:05,200
Speaker 1: version of this where the boom you're describing is the

581
00:34:05,280 --> 00:34:09,320
Speaker 1: conversion of the carbon economy to a to a zero

582
00:34:09,320 --> 00:34:12,360
Speaker 1: mission economy. And so in that context, the cost of

583
00:34:12,360 --> 00:34:14,920
Speaker 1: climate change would be going down because you would be mitigating.

584
00:34:14,920 --> 00:34:17,560
Speaker 1: And so the two things are kind of the same. Um.

585
00:34:17,680 --> 00:34:20,239
Speaker 1: But if you, for example, were to to imagine a

586
00:34:20,280 --> 00:34:24,120
Speaker 1: scenario where all fossil fuel restrictions were lifted and we

587
00:34:24,120 --> 00:34:27,960
Speaker 1: were just prioritizing growth at all costs, independent of the

588
00:34:27,960 --> 00:34:32,680
Speaker 1: impact on the climate, the economic opportunity cost of inaction

589
00:34:32,719 --> 00:34:36,120
Speaker 1: would go down. But that's where the normative arguments come in.

590
00:34:36,200 --> 00:34:39,080
Speaker 1: So that would be uh, something that people wouldn't want

591
00:34:39,120 --> 00:34:42,439
Speaker 1: for other reasons than economics. I think what's convenient right now,

592
00:34:42,920 --> 00:34:47,200
Speaker 1: especially from a policy making standpoint, is the normative and

593
00:34:47,520 --> 00:34:51,040
Speaker 1: descriptive arguments kind of line up in the sense that

594
00:34:51,080 --> 00:34:57,160
Speaker 1: the economics, um, the economics motivate action, the moral imperative

595
00:34:57,200 --> 00:35:00,200
Speaker 1: motivates action, and so the two things are saying the

596
00:35:00,239 --> 00:35:02,400
Speaker 1: same thing. Um, what you'd run into in the scenario

597
00:35:02,480 --> 00:35:06,040
Speaker 1: you describe as is a conflict more along the lines

598
00:35:06,040 --> 00:35:09,560
Speaker 1: of what was true twenty years ago, and you know,

599
00:35:09,600 --> 00:35:12,040
Speaker 1: how the world would behave under the circumstances is kind

600
00:35:12,040 --> 00:35:16,279
Speaker 1: of hard to say, Um, but that's that's an opportunity

601
00:35:16,320 --> 00:35:21,640
Speaker 1: for less economically motivated arguments for climate change action to

602
00:35:22,000 --> 00:35:24,319
Speaker 1: take place. And you know, I guess that it sort

603
00:35:24,320 --> 00:35:29,160
Speaker 1: of brings in the non linear and existential nature of

604
00:35:29,160 --> 00:35:32,000
Speaker 1: the risk and so like how much is survival worth

605
00:35:32,239 --> 00:35:35,320
Speaker 1: to you? Right? It's sort of like what's the value?

606
00:35:35,320 --> 00:35:38,359
Speaker 1: What's the dollar value of the survival of modern civilization?

607
00:35:39,840 --> 00:35:42,520
Speaker 1: I don't think it's just one times global GDP UM.

608
00:35:42,600 --> 00:35:45,280
Speaker 1: So there's you know, the run into this issue where

609
00:35:45,680 --> 00:35:48,560
Speaker 1: economics is probably not the right framework to be thinking

610
00:35:48,600 --> 00:35:53,319
Speaker 1: about that. But from a policy making standpoint, um, the

611
00:35:53,360 --> 00:35:55,919
Speaker 1: amount you should be willing to spend in a world

612
00:35:55,920 --> 00:35:58,799
Speaker 1: of higher growth rates to mitigate the economic impact in

613
00:35:58,880 --> 00:36:02,600
Speaker 1: isolation of climate change would presumably go down. Um. But

614
00:36:02,600 --> 00:36:04,239
Speaker 1: but the there are lots of things we do as

615
00:36:04,239 --> 00:36:09,279
Speaker 1: a country that are not necessarily economic purely economically motivate. Right.

616
00:36:09,320 --> 00:36:11,799
Speaker 1: I guess in the long run, world dead and the

617
00:36:11,840 --> 00:36:14,520
Speaker 1: discount rate doesn't exist. UM. I want to talk a

618
00:36:14,560 --> 00:36:18,120
Speaker 1: little bit more about um. The Treasury market volatility, because

619
00:36:18,120 --> 00:36:21,920
Speaker 1: that's been such a big topic for well anyone in

620
00:36:22,040 --> 00:36:26,400
Speaker 1: markets recently, and it comes into the debate on secularis

621
00:36:26,400 --> 00:36:28,960
Speaker 1: technician and the connection to climate change like we were

622
00:36:29,000 --> 00:36:32,359
Speaker 1: talking about earlier, but more broadly, this is something that

623
00:36:32,400 --> 00:36:36,520
Speaker 1: you've obviously been watching for a long time, and we

624
00:36:36,600 --> 00:36:38,960
Speaker 1: had you on in the aftermath of the big March

625
00:36:39,040 --> 00:36:44,239
Speaker 1: sell off in US treasuries last year. I'm curious, what

626
00:36:44,280 --> 00:36:47,440
Speaker 1: did you observe in the most recent a bout of

627
00:36:47,520 --> 00:36:51,399
Speaker 1: volatility and do you think there's a sort of underlying

628
00:36:51,680 --> 00:36:55,640
Speaker 1: connective tissue between these different dramatic events that we keep

629
00:36:55,640 --> 00:36:58,080
Speaker 1: seeing in the treasury market. So, you know, we had

630
00:36:58,160 --> 00:37:01,880
Speaker 1: the most recent one, we had the sell off, we

631
00:37:01,960 --> 00:37:05,680
Speaker 1: had repo madness. The year before, we had the big

632
00:37:06,360 --> 00:37:09,800
Speaker 1: flash well it wasn't really a flash crash, but flash

633
00:37:09,960 --> 00:37:13,839
Speaker 1: swing in treasury yields a few years before that. Do

634
00:37:13,920 --> 00:37:17,839
Speaker 1: you think there's a sort of underlying cause between all

635
00:37:17,840 --> 00:37:22,120
Speaker 1: of those different events? So what's interesting about this event

636
00:37:22,239 --> 00:37:25,520
Speaker 1: is I think it's in in a weird way, this

637
00:37:25,600 --> 00:37:29,279
Speaker 1: is the healthiest and most like permissible or or least

638
00:37:29,280 --> 00:37:32,920
Speaker 1: worrying version of about a volatility and treasure markets that

639
00:37:32,920 --> 00:37:35,960
Speaker 1: we've had in tenure. And the reason I say that

640
00:37:36,120 --> 00:37:38,640
Speaker 1: is each of the events you just described had some

641
00:37:38,800 --> 00:37:42,279
Speaker 1: underlying weirdness to it that seemed to conflict with the

642
00:37:42,280 --> 00:37:46,200
Speaker 1: intuition of how things are supposed to go. So. Um,

643
00:37:46,280 --> 00:37:49,200
Speaker 1: When when treasuries rallied thirty basis points in an hour,

644
00:37:49,600 --> 00:37:54,480
Speaker 1: that's inherently weird. UM When cash features basis positions that

645
00:37:54,560 --> 00:37:57,320
Speaker 1: the relationship between treasury futures and the cash bonds that

646
00:37:57,360 --> 00:38:01,399
Speaker 1: are deliverable into them completely diverse, like they did in March,

647
00:38:01,760 --> 00:38:06,200
Speaker 1: that's weird. When you have a complete collapse in market

648
00:38:06,239 --> 00:38:09,040
Speaker 1: depth that persists for a long period of time, We've

649
00:38:09,040 --> 00:38:12,719
Speaker 1: had that several on several occasions over the past ten years. Um,

650
00:38:12,760 --> 00:38:15,920
Speaker 1: you know, mark the inability of of dealers and others

651
00:38:15,960 --> 00:38:18,719
Speaker 1: to make markets in a very liquid and safe fast up,

652
00:38:19,600 --> 00:38:22,480
Speaker 1: that's weird. We haven't seen really any of that here.

653
00:38:22,600 --> 00:38:26,120
Speaker 1: And what we've seen is shift in expectations that happens

654
00:38:26,120 --> 00:38:28,640
Speaker 1: all the time from a very low base because there

655
00:38:28,680 --> 00:38:32,680
Speaker 1: was a lot of pessimism early on about the long

656
00:38:32,800 --> 00:38:35,400
Speaker 1: term health of of the economy and the long term

657
00:38:35,400 --> 00:38:37,719
Speaker 1: impacts of COVID. I think it's it's easy to forget now,

658
00:38:37,760 --> 00:38:43,800
Speaker 1: but back last April May, the vaccine prognosis was eighteen

659
00:38:43,800 --> 00:38:46,440
Speaker 1: months at best, and even that's kind of aggressive, right,

660
00:38:46,560 --> 00:38:49,279
Speaker 1: and and and then we and even then, if the

661
00:38:49,360 --> 00:38:52,920
Speaker 1: vaccine six efficacy, that's a great vaccine. And what we

662
00:38:52,960 --> 00:38:59,160
Speaker 1: got instead was efficacy in November and two of them

663
00:38:59,400 --> 00:39:03,960
Speaker 1: and and now how you know, three approved and more coming. So, um,

664
00:39:04,000 --> 00:39:06,080
Speaker 1: we were doing two million shots a day, not one

665
00:39:06,120 --> 00:39:08,000
Speaker 1: million shots to day. I remember when when Biden said

666
00:39:08,000 --> 00:39:10,439
Speaker 1: we want a hundred million people, a hundred million shots

667
00:39:10,480 --> 00:39:12,920
Speaker 1: in a hundred days. Everyone said, well, that's kind of ambitious.

668
00:39:12,920 --> 00:39:15,879
Speaker 1: And now we're gonna we did in sixty um. So

669
00:39:16,520 --> 00:39:19,879
Speaker 1: like there has been truly a shift in the outlook. Um.

670
00:39:19,960 --> 00:39:23,279
Speaker 1: And when you had a massive deflationary shock that was

671
00:39:23,320 --> 00:39:25,960
Speaker 1: expected to persist for eighteen to twenty four months and

672
00:39:25,960 --> 00:39:28,080
Speaker 1: it turned out it's only going to persist for six

673
00:39:28,160 --> 00:39:31,600
Speaker 1: to twelve, there's gonna be a bigger vision and inflation expectations,

674
00:39:31,600 --> 00:39:34,560
Speaker 1: and that's what you saw initially. So the most of

675
00:39:34,600 --> 00:39:37,840
Speaker 1: the move and treasury yields has been on the inflation

676
00:39:37,880 --> 00:39:42,800
Speaker 1: expectations side, even the real rates have started to catch up. Um.

677
00:39:42,840 --> 00:39:46,160
Speaker 1: There has been some positioning effect, but you don't see

678
00:39:46,200 --> 00:39:48,920
Speaker 1: the same sort of I think negative convexity or forced

679
00:39:48,920 --> 00:39:52,000
Speaker 1: hedging is often blamed for a lot of these things,

680
00:39:52,000 --> 00:39:55,239
Speaker 1: and there's certainly been some of that, but um, that's

681
00:39:55,239 --> 00:39:59,319
Speaker 1: all anecdotal, and and the price action suggests that this

682
00:39:59,440 --> 00:40:04,320
Speaker 1: was really changes in in fundamental exposures, not forced activity

683
00:40:04,360 --> 00:40:07,720
Speaker 1: by seeing mortgage headers or there's some c t A effect,

684
00:40:07,840 --> 00:40:11,279
Speaker 1: But you know, you're not seeing the sort of persistent

685
00:40:11,880 --> 00:40:16,360
Speaker 1: relationships and weird relationships between different assets and asset classes

686
00:40:16,400 --> 00:40:18,920
Speaker 1: that you would expect if it was one of these

687
00:40:18,960 --> 00:40:22,520
Speaker 1: sort of forced liquidation or other sort of chasing your

688
00:40:22,560 --> 00:40:25,200
Speaker 1: tail type of events. And the last thing I say,

689
00:40:25,239 --> 00:40:27,600
Speaker 1: which I was frankly somewhat surprised to see, is even

690
00:40:27,600 --> 00:40:30,120
Speaker 1: though we had a twenty basis point move in treasuries

691
00:40:30,160 --> 00:40:34,800
Speaker 1: on a few weeks ago, that the high frequency community

692
00:40:34,840 --> 00:40:39,080
Speaker 1: has been persistently making markets through this volatility, and um,

693
00:40:39,120 --> 00:40:41,640
Speaker 1: that's pretty unusual. Usually they drop off a lot more

694
00:40:41,640 --> 00:40:44,160
Speaker 1: aggressively when ball picks up, and it's a sign of

695
00:40:44,160 --> 00:40:47,719
Speaker 1: a much more healthy market microstructure then we saw, especially

696
00:40:47,920 --> 00:40:51,560
Speaker 1: last March. And so, like I guess, my reaction all

697
00:40:51,560 --> 00:40:53,360
Speaker 1: of this is a bit of a not a yawn,

698
00:40:53,480 --> 00:40:56,400
Speaker 1: but I'm certainly not as worried as I think the

699
00:40:56,520 --> 00:41:00,279
Speaker 1: headline price action would suggest, because when you when you

700
00:41:00,640 --> 00:41:04,640
Speaker 1: revise your expectations dramatically, you should see dramatic changes in prices.

701
00:41:05,120 --> 00:41:09,000
Speaker 1: The question is have we overshot or otherwise exacerbated that

702
00:41:09,160 --> 00:41:13,359
Speaker 1: repricing or are we broadly consistent with the moving fundamental expectations.

703
00:41:13,360 --> 00:41:16,799
Speaker 1: And I think the ladder is mostly true. You know,

704
00:41:16,840 --> 00:41:21,439
Speaker 1: we can point to occasional bouts of of that sort

705
00:41:21,480 --> 00:41:24,440
Speaker 1: of like position squeezes and so forth. Like one of

706
00:41:24,440 --> 00:41:27,000
Speaker 1: the things we were we were watching is there was

707
00:41:27,040 --> 00:41:29,719
Speaker 1: a lot of carry trading that built up over this

708
00:41:29,800 --> 00:41:33,840
Speaker 1: loan for long environment. And when when that happens and

709
00:41:33,920 --> 00:41:37,080
Speaker 1: things start to move and those positions get pressured, you

710
00:41:37,160 --> 00:41:41,640
Speaker 1: usually see the biggest reaction in the highest carry positions.

711
00:41:41,640 --> 00:41:43,640
Speaker 1: So the most attractive carry trades tend to be the

712
00:41:43,640 --> 00:41:46,520
Speaker 1: ones that do the poorest when things start to really reprice.

713
00:41:46,560 --> 00:41:49,240
Speaker 1: And you definitely saw that at times. That's why, for example,

714
00:41:49,320 --> 00:41:52,400
Speaker 1: five your treasures moved the most across the curve on

715
00:41:52,480 --> 00:41:55,120
Speaker 1: that one big day a few weeks ago. But that's

716
00:41:55,120 --> 00:41:58,400
Speaker 1: not a fundamental thing. That's a position squeeze. But you know,

717
00:41:58,480 --> 00:42:00,719
Speaker 1: at the end of the day, that's a day or

718
00:42:00,760 --> 00:42:03,239
Speaker 1: two here there. Um I don't even mean to be

719
00:42:03,239 --> 00:42:05,600
Speaker 1: too sangue about it, because you know, at some point

720
00:42:05,600 --> 00:42:08,680
Speaker 1: everybody's a convex the heager, meaning like we are. There's

721
00:42:08,680 --> 00:42:10,960
Speaker 1: a there's a large community of levere holders who are

722
00:42:11,040 --> 00:42:12,520
Speaker 1: very p and L sensitive, and a lot of them

723
00:42:12,520 --> 00:42:15,680
Speaker 1: were long durations, so you know, you rates back up enough,

724
00:42:15,719 --> 00:42:19,759
Speaker 1: and if if the levered community is chasing that from

725
00:42:19,760 --> 00:42:22,319
Speaker 1: a positioning standpoint, it can end up exacerbating things. But

726
00:42:22,719 --> 00:42:26,120
Speaker 1: but that only goes so far, and I think at

727
00:42:26,120 --> 00:42:29,000
Speaker 1: this point price action is just a lot healthier then

728
00:42:29,120 --> 00:42:35,080
Speaker 1: then say the Tampa tantrum, or certainly March, or after

729
00:42:35,120 --> 00:42:37,880
Speaker 1: the US downgrade, or the European sovereign deet crisis, all

730
00:42:37,880 --> 00:42:41,120
Speaker 1: of these more existential and seeming events like this is

731
00:42:41,200 --> 00:42:44,440
Speaker 1: much more benign in a lot of ways. What, um so,

732
00:42:44,520 --> 00:42:47,200
Speaker 1: I should just note here we are recording this mark.

733
00:42:48,800 --> 00:42:53,480
Speaker 1: It's actually a day before a federal reserve decision, so

734
00:42:54,440 --> 00:42:57,840
Speaker 1: you know, listeners should note that. I'm curious about something

735
00:42:57,880 --> 00:43:02,640
Speaker 1: you mentioned. Looking at the frequency market maker community and

736
00:43:03,000 --> 00:43:07,080
Speaker 1: there ongoing making of such markets even during the worst

737
00:43:07,120 --> 00:43:10,680
Speaker 1: of the volatility a few weeks ago. What are the

738
00:43:10,760 --> 00:43:13,480
Speaker 1: specific data points that you look like when you talk

739
00:43:13,560 --> 00:43:17,759
Speaker 1: about treasury market micro structure, such that we're beyond just

740
00:43:17,840 --> 00:43:20,440
Speaker 1: looking at price, but how well the price is actually working?

741
00:43:20,840 --> 00:43:24,279
Speaker 1: What are the things that tell an observer that, yes,

742
00:43:24,360 --> 00:43:27,720
Speaker 1: price society was very volatile and historic level, but actually

743
00:43:28,080 --> 00:43:32,040
Speaker 1: it still looks like a just a functioning market. Yees.

744
00:43:32,160 --> 00:43:35,560
Speaker 1: So we don't have all the information because a lot

745
00:43:35,600 --> 00:43:40,440
Speaker 1: of these systems obscure information in various ways to protect privacy,

746
00:43:40,560 --> 00:43:43,640
Speaker 1: So there's anonymity concerns and so forth. But um, when

747
00:43:43,640 --> 00:43:46,680
Speaker 1: we think about the treasury market in particular, Um, there's

748
00:43:46,719 --> 00:43:50,239
Speaker 1: kind of two categories of trading. The first is what

749
00:43:50,320 --> 00:43:52,560
Speaker 1: you would call in the old world voice trading or

750
00:43:52,960 --> 00:43:56,360
Speaker 1: dealer to client trading, where uh say, a large central

751
00:43:56,400 --> 00:43:58,759
Speaker 1: bank has a few billion dollars worth of treasuries to

752
00:43:58,800 --> 00:44:00,839
Speaker 1: move and they call up their dealer and they say,

753
00:44:00,840 --> 00:44:03,960
Speaker 1: I'd like to sell you, you know, two billion triple

754
00:44:04,000 --> 00:44:06,319
Speaker 1: old fives, and dealer says, okay, I'll pay you this

755
00:44:06,360 --> 00:44:10,000
Speaker 1: many basis points for from the hot run and markets

756
00:44:10,000 --> 00:44:11,960
Speaker 1: this wide, and they agree in a price. It's it's

757
00:44:12,040 --> 00:44:15,080
Speaker 1: very like it's a much more familiar kind of human

758
00:44:15,080 --> 00:44:19,080
Speaker 1: interaction kind of thing, and it happens in very chunky fashion.

759
00:44:19,280 --> 00:44:22,640
Speaker 1: We don't see that really. Um. There is some reporting

760
00:44:22,640 --> 00:44:25,200
Speaker 1: of that to regulators, but uh, and certainly if you

761
00:44:25,239 --> 00:44:27,799
Speaker 1: work at a broker dealer, you you can you hear

762
00:44:27,840 --> 00:44:32,920
Speaker 1: about generically transactions like that, UM, but there's no systematic

763
00:44:32,960 --> 00:44:35,960
Speaker 1: source of data that's available to public that lets us

764
00:44:36,040 --> 00:44:39,640
Speaker 1: track that activity. But what's useful is that you know, no,

765
00:44:39,640 --> 00:44:41,480
Speaker 1: no single dealer is going to want to hold two

766
00:44:41,520 --> 00:44:44,480
Speaker 1: billion worth of triple old fives. So what tends to

767
00:44:44,480 --> 00:44:47,879
Speaker 1: happen is that transaction gets broken up into pieces and

768
00:44:47,880 --> 00:44:51,200
Speaker 1: and socialized across the dealer community through a variety of transactions,

769
00:44:51,680 --> 00:44:55,360
Speaker 1: some through the futures markets, some through cash markets. But

770
00:44:55,440 --> 00:44:59,040
Speaker 1: there are these inter dealer brokers that facilitate the distribution

771
00:44:59,080 --> 00:45:01,640
Speaker 1: of that risk across the dealer community. And you know

772
00:45:01,680 --> 00:45:04,759
Speaker 1: how that happens in detail is sort of a much

773
00:45:04,840 --> 00:45:07,719
Speaker 1: more technical conversation, but sufficed to say, the risks broken

774
00:45:07,760 --> 00:45:11,960
Speaker 1: up and distributed, and we can observe much more granular

775
00:45:12,000 --> 00:45:16,320
Speaker 1: and rich data on that process. So UM, that's useful

776
00:45:16,320 --> 00:45:18,200
Speaker 1: in the sense that the liquidity of the inter dealer

777
00:45:18,239 --> 00:45:21,560
Speaker 1: market will determine the liquidity those dealers can offer to

778
00:45:21,600 --> 00:45:24,120
Speaker 1: their clients, because your ability to to get out of

779
00:45:24,239 --> 00:45:27,680
Speaker 1: risk is going to be directly related to your willingness

780
00:45:27,680 --> 00:45:29,719
Speaker 1: to put on risk and how much you willing to

781
00:45:29,800 --> 00:45:32,640
Speaker 1: charge for that. So so we we do spend a

782
00:45:32,719 --> 00:45:35,480
Speaker 1: lot of time looking at those transactions. And what we

783
00:45:35,560 --> 00:45:39,080
Speaker 1: have is basically every order in the inter dealer broker

784
00:45:39,160 --> 00:45:43,799
Speaker 1: market going back fifteen years, you know, buy sell, change

785
00:45:43,840 --> 00:45:47,040
Speaker 1: the price, change the amount, change the level you cancel

786
00:45:47,120 --> 00:45:48,840
Speaker 1: the order. We we have all that. It's basically just

787
00:45:49,239 --> 00:45:53,120
Speaker 1: a debug output. It's been Piper context file and we

788
00:45:53,160 --> 00:45:54,640
Speaker 1: can we can sift through that and try to make

789
00:45:54,640 --> 00:45:57,279
Speaker 1: as much use of it as we can. And one

790
00:45:57,320 --> 00:46:00,160
Speaker 1: of the things that we've done, as we said, how

791
00:46:00,239 --> 00:46:03,040
Speaker 1: quickly did this order react to anything else that happened

792
00:46:03,040 --> 00:46:04,960
Speaker 1: in the market, Because we don't know who's placing it,

793
00:46:05,000 --> 00:46:07,640
Speaker 1: but we know some fraction of the participants in that

794
00:46:07,640 --> 00:46:10,919
Speaker 1: inter dealing market are so called high frequency, So how

795
00:46:11,040 --> 00:46:16,640
Speaker 1: fast was this thing? And we find that about of

796
00:46:16,960 --> 00:46:20,560
Speaker 1: orders now are reacting very quickly to something else that happened,

797
00:46:21,239 --> 00:46:25,240
Speaker 1: and we know that's very likely to be high frequency

798
00:46:25,280 --> 00:46:28,680
Speaker 1: because when we plot up a histogram, if we look

799
00:46:28,680 --> 00:46:31,360
Speaker 1: at the distribution of timing of those events, there's a

800
00:46:31,440 --> 00:46:33,880
Speaker 1: very strong peak at eight mili The seconds and and

801
00:46:33,920 --> 00:46:36,960
Speaker 1: the questions, what's what's significant about eight milli seconds, Well,

802
00:46:36,960 --> 00:46:38,440
Speaker 1: that's the time it takes for an email to get

803
00:46:38,440 --> 00:46:41,080
Speaker 1: from New York to Chicago. So that's people checking the

804
00:46:41,160 --> 00:46:43,800
Speaker 1: checking the pit, coming back to New York, trading the

805
00:46:43,840 --> 00:46:46,040
Speaker 1: cash market. So you can see these little features that

806
00:46:46,080 --> 00:46:49,600
Speaker 1: are indicative of what you think you're seeing because we

807
00:46:49,600 --> 00:46:51,840
Speaker 1: again don't know who's trading. We just know that they're trading.

808
00:46:52,200 --> 00:46:54,560
Speaker 1: High frequency is one of those things that like relies

809
00:46:54,680 --> 00:46:59,840
Speaker 1: on on strong and resilient market infrast market micro structure

810
00:47:00,200 --> 00:47:03,360
Speaker 1: to be profitable, and that's the liquidity that's on the

811
00:47:03,360 --> 00:47:05,520
Speaker 1: screen until you need it and then it's all. And

812
00:47:05,719 --> 00:47:07,919
Speaker 1: so when we track that, you know that goes from

813
00:47:08,440 --> 00:47:12,359
Speaker 1: the market. The market under periods of real stress. This

814
00:47:12,480 --> 00:47:14,839
Speaker 1: past few weeks, it really hasn't dipped that much. It's

815
00:47:14,880 --> 00:47:19,200
Speaker 1: gone down to maybe, but it stayed pretty resilient. And

816
00:47:19,280 --> 00:47:21,920
Speaker 1: that means that your on screen liquidity, what you see

817
00:47:22,080 --> 00:47:25,239
Speaker 1: in the market is available to transact, is really there

818
00:47:25,840 --> 00:47:28,520
Speaker 1: and that has not always been true, and so like

819
00:47:28,560 --> 00:47:47,759
Speaker 1: that's a sign of healthy market infrastruction. So we've been

820
00:47:47,800 --> 00:47:52,000
Speaker 1: focused on treasury market micro structure and the plumbing of

821
00:47:52,000 --> 00:47:55,200
Speaker 1: the actual system. But if we look at it from

822
00:47:55,200 --> 00:48:00,160
Speaker 1: a more macro perspective, I'm wondering, at what point do

823
00:48:00,400 --> 00:48:04,359
Speaker 1: you see the rise and yields or I don't want

824
00:48:04,400 --> 00:48:08,640
Speaker 1: to see, say, treasury market um dysfunction because you know,

825
00:48:08,719 --> 00:48:11,479
Speaker 1: as you put it, the recent bout of volatility wasn't

826
00:48:11,480 --> 00:48:16,400
Speaker 1: nearly as bad as we've seen in some previous UM instances.

827
00:48:16,440 --> 00:48:19,440
Speaker 1: But at what point do the current events in the

828
00:48:19,440 --> 00:48:23,520
Speaker 1: treasury market become a problem for the FED or at

829
00:48:23,600 --> 00:48:27,640
Speaker 1: least something that it feels it has to push back against.

830
00:48:29,320 --> 00:48:31,520
Speaker 1: So I think that the memory of last March is

831
00:48:31,560 --> 00:48:34,000
Speaker 1: that not just that rates were moving a lot, but

832
00:48:34,080 --> 00:48:37,440
Speaker 1: that price discovery was broken. So you know, something that

833
00:48:37,520 --> 00:48:40,040
Speaker 1: was issued three months earlier than another bond that were

834
00:48:40,040 --> 00:48:42,320
Speaker 1: functionally the same instrument with the same credit, we're trading

835
00:48:42,320 --> 00:48:45,120
Speaker 1: at very different prices. And that's only if you thought

836
00:48:45,160 --> 00:48:46,880
Speaker 1: the price you could get was really the price. And

837
00:48:46,920 --> 00:48:51,360
Speaker 1: so the FED is not specifically concerned with rates rising,

838
00:48:51,960 --> 00:48:54,560
Speaker 1: and especially if it's inflation expectations moving up, I think

839
00:48:54,560 --> 00:48:59,480
Speaker 1: that's that's a sign of healthy recovery UM. Their concern

840
00:48:59,680 --> 00:49:02,600
Speaker 1: from when we talk about market functioning we're not talking

841
00:49:02,600 --> 00:49:06,040
Speaker 1: about volatility in isolation. We're talking about price discovery really,

842
00:49:06,520 --> 00:49:11,960
Speaker 1: and the questions, can markets assign realistic prices to securities

843
00:49:12,400 --> 00:49:14,239
Speaker 1: in a realistic period of time and it really and

844
00:49:14,320 --> 00:49:18,719
Speaker 1: a decent enough size to facilitate the broader functioning of

845
00:49:18,719 --> 00:49:21,239
Speaker 1: the market. Are they introducing sort of the risk of

846
00:49:21,280 --> 00:49:24,680
Speaker 1: a market failure in which there just is not enough

847
00:49:24,719 --> 00:49:30,799
Speaker 1: intermediation capacity two to keep things flowing. Um, So you know,

848
00:49:30,960 --> 00:49:37,720
Speaker 1: that's that's more that's much more existential, that's much more concerning. UM.

849
00:49:37,760 --> 00:49:41,480
Speaker 1: There's not a real sign that that's imminent in any respect.

850
00:49:41,680 --> 00:49:45,560
Speaker 1: Right now, you see some some dispersion in the pricing

851
00:49:45,600 --> 00:49:49,239
Speaker 1: of securities, but we think that's more attributable to the

852
00:49:49,320 --> 00:49:52,720
Speaker 1: details of how banks invest with their portfolio and potential

853
00:49:52,760 --> 00:49:56,440
Speaker 1: regulatory changes, which are important. But like what to have

854
00:49:56,719 --> 00:49:58,759
Speaker 1: an event like last March, you need two things. You

855
00:49:58,760 --> 00:50:03,040
Speaker 1: need an underlying vulner ability and you need an event

856
00:50:03,120 --> 00:50:06,279
Speaker 1: of sufficient magnitude of shock that's big enough kind of

857
00:50:06,320 --> 00:50:11,000
Speaker 1: exposed and widen that crack. And so the COVID shock

858
00:50:11,160 --> 00:50:14,640
Speaker 1: in the context of balance she constraints on bank market making,

859
00:50:14,920 --> 00:50:17,560
Speaker 1: was precisely that. So the risk had been there all

860
00:50:17,600 --> 00:50:20,520
Speaker 1: along it just took a big enough shock to expose

861
00:50:20,600 --> 00:50:23,600
Speaker 1: it and widen it and turn it into something much more,

862
00:50:25,080 --> 00:50:29,719
Speaker 1: much more concerning. Now, I think, on the one hand,

863
00:50:29,760 --> 00:50:32,279
Speaker 1: we don't really have that vulnerability, although we'll see what

864
00:50:32,320 --> 00:50:35,960
Speaker 1: happens with with the regulatory outlook. We're likely to get

865
00:50:36,000 --> 00:50:38,640
Speaker 1: some color on that or some clarity on that pretty soon.

866
00:50:39,400 --> 00:50:41,640
Speaker 1: But more importantly, this just doesn't seem like the kind

867
00:50:41,640 --> 00:50:47,080
Speaker 1: of shock that's a sufficient magnitude to drive a wedge

868
00:50:47,120 --> 00:50:50,319
Speaker 1: through that vulnerability in the first place. So you just

869
00:50:50,360 --> 00:50:52,839
Speaker 1: don't have the same set of weakness and you don't

870
00:50:52,840 --> 00:50:55,719
Speaker 1: have the same magnitude of shock. Um. So I hate

871
00:50:55,719 --> 00:50:58,200
Speaker 1: to say everything's gonna be fine, because that's their famous

872
00:50:58,280 --> 00:51:01,759
Speaker 1: last words, um, But we could have a big move

873
00:51:01,840 --> 00:51:04,640
Speaker 1: in rates, and I think it's important to say again,

874
00:51:04,680 --> 00:51:07,920
Speaker 1: like that's happened before, it will happen again. Um. Markets

875
00:51:07,960 --> 00:51:11,359
Speaker 1: re price as the view changes, as the expectations change,

876
00:51:11,360 --> 00:51:14,440
Speaker 1: because the world changes, and sometimes dramatically. But so long

877
00:51:14,480 --> 00:51:20,080
Speaker 1: as the market is reflecting accurately that underlying fundamental shifting expectations,

878
00:51:20,680 --> 00:51:23,319
Speaker 1: there's nothing to be quote unquote concerned about, at least

879
00:51:23,320 --> 00:51:25,759
Speaker 1: from a market functioning standpoint. Now, we may not like

880
00:51:25,840 --> 00:51:29,680
Speaker 1: the way the world's going, and you know, inflation running

881
00:51:29,680 --> 00:51:32,120
Speaker 1: away is clearly a problem that the FED would have

882
00:51:32,160 --> 00:51:35,200
Speaker 1: to deal with, but that's a macroeconomic issue, that's not

883
00:51:35,239 --> 00:51:39,680
Speaker 1: a markets issue. So you you did kind of allude

884
00:51:39,719 --> 00:51:43,040
Speaker 1: to this, this sort of upcoming sort of regulatory questions.

885
00:51:43,160 --> 00:51:46,239
Speaker 1: Of course after um, you know, in the in the

886
00:51:46,280 --> 00:51:50,880
Speaker 1: wake of the crash, the market crisis whatever, um, last

887
00:51:50,920 --> 00:51:54,400
Speaker 1: March um, one of the things the FED did was

888
00:51:54,480 --> 00:51:58,120
Speaker 1: basically make it I guess you describe it such that

889
00:51:58,400 --> 00:52:01,600
Speaker 1: the banks could hold unlimited amounts of treasuries without any

890
00:52:01,680 --> 00:52:05,080
Speaker 1: sort of like penalty on their total assets. So again

891
00:52:05,160 --> 00:52:08,560
Speaker 1: mention the proviso when we're recording this, we don't know

892
00:52:08,600 --> 00:52:12,080
Speaker 1: exactly what we'll here by the time uh this comes out.

893
00:52:12,320 --> 00:52:15,080
Speaker 1: Can you just walk us through the tension here? And

894
00:52:15,120 --> 00:52:17,279
Speaker 1: it's sort of like how we should think about this

895
00:52:17,360 --> 00:52:21,320
Speaker 1: question of the SLR and how big it could be regard,

896
00:52:21,400 --> 00:52:24,720
Speaker 1: you know, depending on how the FED acts with extending

897
00:52:24,760 --> 00:52:28,480
Speaker 1: this program. Yeah. So so sl ARE the supplementary leverage

898
00:52:28,560 --> 00:52:33,640
Speaker 1: ratio is a fairly blunt regulatory instrument. So it basically

899
00:52:33,680 --> 00:52:37,440
Speaker 1: says you, as a bank need to hold capital relative

900
00:52:37,480 --> 00:52:39,920
Speaker 1: to the overall size of your institution, So it doesn't

901
00:52:39,920 --> 00:52:42,480
Speaker 1: matter what you have. However much you have of it,

902
00:52:42,800 --> 00:52:45,759
Speaker 1: make sure you have five percent in capital um at

903
00:52:45,760 --> 00:52:48,160
Speaker 1: the holding company level and in six percent for your

904
00:52:48,320 --> 00:52:51,839
Speaker 1: your bank operating companies. So the issue there was that

905
00:52:51,920 --> 00:52:55,759
Speaker 1: back last March um there was just a ton of

906
00:52:55,800 --> 00:52:58,480
Speaker 1: demand for liquidity and people have been holding treasuries as

907
00:52:58,480 --> 00:53:01,320
Speaker 1: a cash surrogain, so that the thought it was treasuries

908
00:53:01,360 --> 00:53:04,920
Speaker 1: are risk free, they are highly liquid, they pay a yield,

909
00:53:05,160 --> 00:53:08,040
Speaker 1: so instead of just earning cash returns, I'm going to

910
00:53:08,239 --> 00:53:11,080
Speaker 1: buy a five year treasury and if I need to

911
00:53:11,080 --> 00:53:13,440
Speaker 1: sell it, I can sell it at low cost. And

912
00:53:13,960 --> 00:53:16,880
Speaker 1: the issue is that banks facilitate that sale because the

913
00:53:16,920 --> 00:53:20,000
Speaker 1: vast majority of trading happens with bank affiliated dealers, and

914
00:53:20,000 --> 00:53:23,440
Speaker 1: those dealers is subject to these regulations. Um. I think

915
00:53:23,480 --> 00:53:25,160
Speaker 1: it's a little bit of a misnomber to say that

916
00:53:25,239 --> 00:53:29,399
Speaker 1: balance sheet constraints were strictly binding last March, meaning banks

917
00:53:29,440 --> 00:53:33,200
Speaker 1: simply ran out of capacity. Um, they had plenty of capacity. Actually,

918
00:53:33,280 --> 00:53:36,000
Speaker 1: it's kind of a misnomer, like the SLR was never

919
00:53:36,440 --> 00:53:41,320
Speaker 1: explicitly binding last March um. So if it wasn't binding,

920
00:53:41,320 --> 00:53:43,680
Speaker 1: then what was the problem. And you know, I've been

921
00:53:43,680 --> 00:53:47,040
Speaker 1: out there talking about how balance shee constraints were a problem.

922
00:53:47,040 --> 00:53:49,080
Speaker 1: But if they weren't strictly binding them, then where did

923
00:53:49,080 --> 00:53:51,759
Speaker 1: the stress come from? And the answer is it's less

924
00:53:51,800 --> 00:53:54,960
Speaker 1: about your institutional constraints and more about your local constraints.

925
00:53:54,960 --> 00:53:58,080
Speaker 1: So if you're assigned a commodity and asked to distributed

926
00:53:58,120 --> 00:54:00,400
Speaker 1: among different businesses, so bank has a hunt of dollars

927
00:54:00,400 --> 00:54:03,160
Speaker 1: worth of balance sheet and some of it goes to

928
00:54:03,160 --> 00:54:04,959
Speaker 1: the treasury desks, so it goes to the credit desks,

929
00:54:04,960 --> 00:54:06,719
Speaker 1: and it goes to the equities desks, and it goes

930
00:54:06,760 --> 00:54:09,080
Speaker 1: to the repo desks, and it goes to the f

931
00:54:09,320 --> 00:54:11,160
Speaker 1: X desk. You have to you have to allocate this

932
00:54:11,239 --> 00:54:13,960
Speaker 1: scarce commodity, but you need to make sure that at

933
00:54:13,960 --> 00:54:17,400
Speaker 1: an institutional level you have enough or you don't go

934
00:54:17,480 --> 00:54:20,800
Speaker 1: over budget. So that means those assignments are pretty rigid

935
00:54:21,320 --> 00:54:24,200
Speaker 1: and it's hard to reallocate. So I think there's also

936
00:54:24,280 --> 00:54:26,879
Speaker 1: a sense of banks are highly efficient in their use

937
00:54:26,880 --> 00:54:29,520
Speaker 1: of their balance sheet, and they are uh, like we're

938
00:54:29,520 --> 00:54:32,560
Speaker 1: talking about over long periods, but not necessarily over short periods,

939
00:54:32,600 --> 00:54:35,640
Speaker 1: And so you can run into a situation where a

940
00:54:35,719 --> 00:54:38,799
Speaker 1: treasury trading desk or repo desk, which uses a ton

941
00:54:38,800 --> 00:54:41,040
Speaker 1: of balance sheet and is very volatile, and its use

942
00:54:41,040 --> 00:54:44,319
Speaker 1: of balance sheet can run into local limits that are

943
00:54:44,400 --> 00:54:48,760
Speaker 1: part of the business management and planet capital planning process. Um.

944
00:54:48,800 --> 00:54:52,160
Speaker 1: But the institution's got plenty of excess, but they are constrained,

945
00:54:52,320 --> 00:54:55,480
Speaker 1: and so they start to act like the institution is

946
00:54:55,520 --> 00:54:58,200
Speaker 1: balance sheet constraint, even though it's really just a local effect.

947
00:54:58,520 --> 00:55:01,120
Speaker 1: What the SLR car that's did. As they said, those

948
00:55:01,160 --> 00:55:06,480
Speaker 1: treasuries don't contribute to your leverage exposure, So treasury desk

949
00:55:06,600 --> 00:55:09,279
Speaker 1: used as much as you want, basically like this is

950
00:55:09,280 --> 00:55:11,759
Speaker 1: not going to be a problem. Uh, you know, don't

951
00:55:11,760 --> 00:55:13,759
Speaker 1: go crazy and don't and make sure you know how

952
00:55:13,840 --> 00:55:15,520
Speaker 1: much you got. But like if you need balance sheet,

953
00:55:15,880 --> 00:55:17,920
Speaker 1: you know your need for additional balance sheet is not

954
00:55:17,960 --> 00:55:21,279
Speaker 1: going to affect my capital planning at an institutional level

955
00:55:21,480 --> 00:55:23,280
Speaker 1: at least as long as these carve outs are in place,

956
00:55:23,320 --> 00:55:26,200
Speaker 1: and therefore you can be much more flexible with how

957
00:55:26,239 --> 00:55:28,600
Speaker 1: that desk operates and so they don't run into these constraints.

958
00:55:28,600 --> 00:55:32,120
Speaker 1: Is frequently that's about mitigating market failure risk. It's not

959
00:55:32,200 --> 00:55:36,400
Speaker 1: about reducing leverage constraints in real time. It's about clipping

960
00:55:36,400 --> 00:55:40,640
Speaker 1: the tails of these more problematic outcomes, especially at a

961
00:55:40,640 --> 00:55:44,879
Speaker 1: period of time when the trajectory was towards an acceleration

962
00:55:44,880 --> 00:55:46,520
Speaker 1: in the use of balance sheet So it was the

963
00:55:46,640 --> 00:55:48,560
Speaker 1: it wasn't the level, it was the path of growth

964
00:55:49,239 --> 00:55:53,720
Speaker 1: and the way that that was affecting local behavior. Um.

965
00:55:53,760 --> 00:55:57,520
Speaker 1: So those rules were put in place to avoid those

966
00:55:58,080 --> 00:56:01,080
Speaker 1: those can to avoid those market failure risks. They've been

967
00:56:01,120 --> 00:56:04,640
Speaker 1: largely effective at doing so in combination with fit purchases. Um.

968
00:56:04,760 --> 00:56:06,279
Speaker 1: But they're going to expire at the end of this month,

969
00:56:06,440 --> 00:56:08,520
Speaker 1: so the question is, you know, are they going to

970
00:56:08,560 --> 00:56:11,600
Speaker 1: get extended, if so, for how long and what form?

971
00:56:11,680 --> 00:56:14,000
Speaker 1: And they're going to be extended in a format. And

972
00:56:14,040 --> 00:56:17,919
Speaker 1: there's a distinction between how the bank operating company which

973
00:56:17,920 --> 00:56:20,960
Speaker 1: is the deposit taking institution, and the dealer, which is

974
00:56:20,960 --> 00:56:23,640
Speaker 1: part of the holding company, how they are separately affected

975
00:56:23,640 --> 00:56:26,680
Speaker 1: by these rules. Um. But but I think that this

976
00:56:26,719 --> 00:56:30,080
Speaker 1: specific issue of market functioning risk is a dealer issue,

977
00:56:30,320 --> 00:56:32,880
Speaker 1: which means it's a holding company issue, which means the

978
00:56:32,960 --> 00:56:36,959
Speaker 1: question is, on the one hand, you know, how how

979
00:56:37,040 --> 00:56:39,759
Speaker 1: is this going to be treated after March three one?

980
00:56:40,320 --> 00:56:42,040
Speaker 1: And on the other how's that going to affect the

981
00:56:42,080 --> 00:56:46,160
Speaker 1: way that these institutions behave internally and manage their balante

982
00:56:46,160 --> 00:56:49,200
Speaker 1: exposure because because that's where the sort of real risk

983
00:56:49,239 --> 00:56:52,960
Speaker 1: factors lie in practice. The issue here is that we're

984
00:56:52,960 --> 00:56:57,080
Speaker 1: coming into the end of this period with less buffer

985
00:56:57,120 --> 00:56:59,680
Speaker 1: at an institutional level than was true early last year.

986
00:57:00,080 --> 00:57:03,359
Speaker 1: So banks would be entering a balance sheet constrained world,

987
00:57:03,440 --> 00:57:05,000
Speaker 1: or at least a world in which SR is in

988
00:57:05,080 --> 00:57:09,280
Speaker 1: principal binding with less buffer. That means more rigid internal

989
00:57:09,400 --> 00:57:12,800
Speaker 1: allocations and economies a balance sheet, which means greater market

990
00:57:12,840 --> 00:57:17,520
Speaker 1: functioning ris And so you know that's a scenario where

991
00:57:17,760 --> 00:57:20,439
Speaker 1: in principle, if if this keeps going, you could run

992
00:57:20,480 --> 00:57:23,280
Speaker 1: into world where there is a need to sell and

993
00:57:23,360 --> 00:57:25,480
Speaker 1: raise liquidity on the part of the real economy, the

994
00:57:25,560 --> 00:57:28,800
Speaker 1: non banks, and it would be much harder for for

995
00:57:28,840 --> 00:57:32,040
Speaker 1: dealers to intermediate that under the circumstances. Again, you need

996
00:57:32,080 --> 00:57:35,240
Speaker 1: a shock of sufficient magnitude to to really exacerbate that

997
00:57:35,280 --> 00:57:38,800
Speaker 1: and turn that into a a real issue. But the

998
00:57:38,920 --> 00:57:42,959
Speaker 1: risk is greater all outs equal if banks have less

999
00:57:42,960 --> 00:57:47,760
Speaker 1: flexibility in the way that they allocate that balance sheet. Okay, well,

1000
00:57:47,920 --> 00:57:50,880
Speaker 1: clearly a lot going on in the treasury market um,

1001
00:57:50,880 --> 00:57:52,880
Speaker 1: and we do have that FED meeting coming up. We

1002
00:57:52,920 --> 00:57:55,600
Speaker 1: are recording before it, so it'll be interesting to see

1003
00:57:55,640 --> 00:57:58,640
Speaker 1: how everything shakes out. So thank you so much Josh

1004
00:57:58,720 --> 00:58:02,080
Speaker 1: for coming on for your fourth All Thoughts appearance, and

1005
00:58:02,160 --> 00:58:05,960
Speaker 1: also for connecting the world of real yields with climate change,

1006
00:58:06,000 --> 00:58:12,320
Speaker 1: which not many people do. Thanks Spanning of Magnitude. That

1007
00:58:12,440 --> 00:58:14,560
Speaker 1: was great. Thank you so much, jos Yeah, thanks very

1008
00:58:14,640 --> 00:58:35,480
Speaker 1: much so, Joe. I found that conversation very interesting, a

1009
00:58:35,480 --> 00:58:38,680
Speaker 1: little bit technical, but I do think this broad idea

1010
00:58:39,720 --> 00:58:42,800
Speaker 1: that the way we think about climate change, or the

1011
00:58:42,840 --> 00:58:46,680
Speaker 1: way we make these cost benefit calculations is tied to

1012
00:58:47,640 --> 00:58:50,840
Speaker 1: estimates of the time value of money the discount rate

1013
00:58:50,920 --> 00:58:54,720
Speaker 1: as much as our actual estimates of how climate change

1014
00:58:54,880 --> 00:58:57,760
Speaker 1: is proceeding and climate change damage. I think that's really

1015
00:58:57,760 --> 00:59:01,840
Speaker 1: interesting because I don't know, like the climate change questions

1016
00:59:01,880 --> 00:59:05,680
Speaker 1: seems to be framed so much in science, which you know,

1017
00:59:05,720 --> 00:59:09,080
Speaker 1: for obvious reasons, people talk about it from a scientific perspective,

1018
00:59:09,160 --> 00:59:12,960
Speaker 1: but people don't really talk about it from an economic perspective,

1019
00:59:13,080 --> 00:59:15,800
Speaker 1: except in a broad sense that well, you know, you

1020
00:59:15,840 --> 00:59:19,400
Speaker 1: have to trade off economic growth for restrictions on pollution

1021
00:59:19,520 --> 00:59:21,920
Speaker 1: and things like that. They don't actually get into things

1022
00:59:22,520 --> 00:59:26,240
Speaker 1: like interest rates, um all that much. I think. No,

1023
00:59:26,400 --> 00:59:30,080
Speaker 1: it's super interesting because it does feel like there is

1024
00:59:30,120 --> 00:59:32,320
Speaker 1: this you know, I think a lot of people would

1025
00:59:32,360 --> 00:59:37,120
Speaker 1: accept this premise that something like climate change, which is

1026
00:59:37,600 --> 00:59:43,360
Speaker 1: slow moving global in nature, affects everyone but also at

1027
00:59:43,520 --> 00:59:46,520
Speaker 1: very sort of like indeterminate times in the future is

1028
00:59:46,520 --> 00:59:50,040
Speaker 1: like a very like hard thing for like market to price.

1029
00:59:50,840 --> 00:59:54,600
Speaker 1: But the idea of still using market pricing concepts to

1030
00:59:54,680 --> 00:59:57,960
Speaker 1: try to get an estimate of the cost of inaction,

1031
00:59:58,040 --> 00:59:59,760
Speaker 1: which is I guess basically what we're talking about the

1032
00:59:59,760 --> 01:00:03,320
Speaker 1: cause to inaction is a really sort of like, um,

1033
01:00:03,320 --> 01:00:06,320
Speaker 1: it's a really interesting exercise. Yeah. And also just the

1034
01:00:06,360 --> 01:00:09,320
Speaker 1: broad point about how the last time we did a

1035
01:00:09,520 --> 01:00:12,960
Speaker 1: really really big climate change report was the Stern Report

1036
01:00:13,040 --> 01:00:15,760
Speaker 1: back in gosh, I can't even remember when that was.

1037
01:00:15,760 --> 01:00:18,680
Speaker 1: Was that like two thousand six or something like that.

1038
01:00:19,680 --> 01:00:23,880
Speaker 1: I think I think it said two okay, yeah, like

1039
01:00:24,200 --> 01:00:27,840
Speaker 1: and back then interest rates were completely different to what

1040
01:00:27,880 --> 01:00:30,400
Speaker 1: they are now, and you know, his estimate of how

1041
01:00:30,480 --> 01:00:33,400
Speaker 1: much that changes the calculation is just something that well

1042
01:00:33,600 --> 01:00:36,720
Speaker 1: I certainly hadn't thought of it before, but I guess

1043
01:00:36,720 --> 01:00:40,080
Speaker 1: like I still have some issue with this idea that's like, okay,

1044
01:00:40,160 --> 01:00:45,040
Speaker 1: like there is this like looming humanitarian catastrophe. Maybe we're

1045
01:00:45,040 --> 01:00:49,320
Speaker 1: already seeing aspects of it played out playing out, and

1046
01:00:50,480 --> 01:00:52,200
Speaker 1: it's like either we're gonna we want to do something

1047
01:00:52,240 --> 01:00:54,400
Speaker 1: about it or not. And so it's kind of weird

1048
01:00:54,480 --> 01:00:56,680
Speaker 1: still for me to wrap my head around this idea

1049
01:00:56,760 --> 01:01:00,960
Speaker 1: that well, that would have been you know, a certain

1050
01:01:01,000 --> 01:01:03,520
Speaker 1: costs in two thousand seven, but the real interest rate

1051
01:01:03,560 --> 01:01:06,640
Speaker 1: environment is changed from two thousand seven to two thowy one,

1052
01:01:07,040 --> 01:01:09,400
Speaker 1: and so now there's like a total um sort of

1053
01:01:09,400 --> 01:01:12,760
Speaker 1: like different cost benefit analysis. But I guess still just

1054
01:01:13,280 --> 01:01:15,760
Speaker 1: this idea that's like, all right, if we accept certain

1055
01:01:15,760 --> 01:01:18,880
Speaker 1: premises about the cost of capital, how much is it

1056
01:01:18,960 --> 01:01:23,960
Speaker 1: costing us to wait? Is sort of a a useful

1057
01:01:24,640 --> 01:01:28,040
Speaker 1: a useful frame in terms of thinking about policy. Yeah,

1058
01:01:28,120 --> 01:01:30,800
Speaker 1: you can see there's something distinctly awkward about, you know,

1059
01:01:31,000 --> 01:01:34,120
Speaker 1: telling someone we're sorry that your house is underwater from

1060
01:01:34,200 --> 01:01:37,280
Speaker 1: climate change. But it's because the interest rate change between

1061
01:01:37,280 --> 01:01:39,680
Speaker 1: the last time that we made an estimate of how

1062
01:01:39,760 --> 01:01:42,040
Speaker 1: much it would cost to fix this problem. And now

1063
01:01:42,400 --> 01:01:44,840
Speaker 1: like there's something yeah, yeah, and you hinted it that

1064
01:01:45,040 --> 01:01:47,600
Speaker 1: like future generations or people who aren't born yet don't

1065
01:01:47,600 --> 01:01:49,440
Speaker 1: get a vote. But it's like, oh, we're going to

1066
01:01:49,560 --> 01:01:53,120
Speaker 1: do something about climate. But the real interest real interest

1067
01:01:53,280 --> 01:01:56,040
Speaker 1: rates were sharply positive at the time, and things were growing,

1068
01:01:56,080 --> 01:01:59,280
Speaker 1: and so the time value of money discounted fifty years

1069
01:01:59,280 --> 01:02:03,480
Speaker 1: from now, like it wasn't really that much. It does

1070
01:02:03,560 --> 01:02:07,400
Speaker 1: sort of like raise some like I guess kind of awkward. Yeah,

1071
01:02:07,480 --> 01:02:09,960
Speaker 1: I mean, I guess Josh mentioned this, But there's the

1072
01:02:10,000 --> 01:02:13,560
Speaker 1: sort of financial imperative and the moral imperative as well, right,

1073
01:02:13,560 --> 01:02:16,320
Speaker 1: but maybe I don't know, maybe by focusing on the

1074
01:02:16,320 --> 01:02:20,560
Speaker 1: financial imperative, that's a way to get more people involved

1075
01:02:20,880 --> 01:02:24,480
Speaker 1: in the whole project, and you know, maybe the ends

1076
01:02:24,560 --> 01:02:28,120
Speaker 1: sort of justify the means, right, Yeah, and again like

1077
01:02:28,600 --> 01:02:30,400
Speaker 1: the rules are going to come from somewhere, right, Like

1078
01:02:30,480 --> 01:02:32,560
Speaker 1: that's sort of what I took away. It's like someone

1079
01:02:32,760 --> 01:02:36,400
Speaker 1: how politicians around the world addressed this, Like they're still

1080
01:02:36,440 --> 01:02:40,400
Speaker 1: built up things like carbon pricing or other regulations that

1081
01:02:40,560 --> 01:02:43,680
Speaker 1: may have some impediment to slow the economy or maybe

1082
01:02:44,040 --> 01:02:47,160
Speaker 1: you know, needs some uh guess of like okay, how

1083
01:02:47,240 --> 01:02:49,240
Speaker 1: much do we invest in X or Y? Like the

1084
01:02:49,320 --> 01:02:51,720
Speaker 1: rules have to come from somewhere as they're trying to

1085
01:02:51,880 --> 01:02:54,080
Speaker 1: get ahead of this by like thinking about how these

1086
01:02:54,120 --> 01:02:56,800
Speaker 1: models might get shaped as a I use future that. Yeah,

1087
01:02:57,080 --> 01:03:00,400
Speaker 1: absolutely all right? Um, shall we leave it there? Yeah?

1088
01:03:00,600 --> 01:03:04,320
Speaker 1: Leave it there? Okay. This has been another episode of

1089
01:03:04,360 --> 01:03:07,200
Speaker 1: the All Thoughts Podcast. I'm Tracy Alloway. You can follow

1090
01:03:07,200 --> 01:03:10,360
Speaker 1: me on Twitter at Tracy Alloway and I'm Joe wi

1091
01:03:10,400 --> 01:03:13,240
Speaker 1: Isn't Thal. You could follow me on Twitter at the Stalwart.

1092
01:03:13,600 --> 01:03:17,520
Speaker 1: Follow our producer Laura Carlson. She's at Laura M. Carlson.

1093
01:03:17,840 --> 01:03:21,920
Speaker 1: Followed the Bloomberg head of podcast, Francesco Levi at Francesco Today,

1094
01:03:22,360 --> 01:03:25,320
Speaker 1: and check out all of our podcasts under the handle

1095
01:03:25,600 --> 01:03:27,320
Speaker 1: at podcast. Thanks for listening.