1
00:00:02,720 --> 00:00:14,080
Speaker 1: Bloomberg Audio Studios, Podcasts, Radio News.

2
00:00:18,360 --> 00:00:21,200
Speaker 2: Hello and welcome to another episode of the ath Thoughts Podcast.

3
00:00:21,320 --> 00:00:22,640
Speaker 2: I'm Tracy Alloway.

4
00:00:22,360 --> 00:00:23,600
Speaker 3: And I'm Joe Wisenthal.

5
00:00:24,000 --> 00:00:27,800
Speaker 2: Joe, imagine you are a utilities analyst.

6
00:00:27,960 --> 00:00:28,200
Speaker 1: Yeah.

7
00:00:28,400 --> 00:00:31,640
Speaker 2: Fun, and for years you are laboring in the utility

8
00:00:31,680 --> 00:00:35,680
Speaker 2: analysis mind. Yeah, and you know, we like talking about utilities,

9
00:00:35,680 --> 00:00:38,960
Speaker 2: We like talking about energy. We find pretty much anything interesting,

10
00:00:39,040 --> 00:00:43,240
Speaker 2: that's right, equal opportunity interest people we are. But yes,

11
00:00:43,520 --> 00:00:46,160
Speaker 2: you got to say utilities. For a while, some people

12
00:00:46,200 --> 00:00:46,760
Speaker 2: would say it was a.

13
00:00:46,720 --> 00:00:49,440
Speaker 3: Little POORI no, that's right. I mean for most of

14
00:00:49,479 --> 00:00:53,600
Speaker 3: our careers. I think if you were utilities analyst, a

15
00:00:53,720 --> 00:00:57,480
Speaker 3: really big part of your job. And maybe I'm wrong,

16
00:00:57,720 --> 00:01:00,600
Speaker 3: but I just didn't the popular discourse was like talking

17
00:01:00,640 --> 00:01:03,480
Speaker 3: about yield relative to treasuries. Right, they were seeing a

18
00:01:03,600 --> 00:01:06,640
Speaker 3: sort of bond like instruments, et cetera. Maybe a little

19
00:01:06,680 --> 00:01:08,119
Speaker 3: bit of growth, but real.

20
00:01:08,120 --> 00:01:11,720
Speaker 2: Reliable, safe haven ish dividend plays.

21
00:01:11,760 --> 00:01:13,840
Speaker 3: I guess totally. And since I know where we're going

22
00:01:13,880 --> 00:01:17,039
Speaker 3: with this conversation, one of the themes of the last

23
00:01:17,120 --> 00:01:21,600
Speaker 3: few years has been what I would say is the

24
00:01:21,640 --> 00:01:25,760
Speaker 3: old industries that were either stable or cyclical becoming secular

25
00:01:25,920 --> 00:01:26,640
Speaker 3: in the way they grow.

26
00:01:27,319 --> 00:01:29,880
Speaker 2: I think that's right. So what is happening now is

27
00:01:30,120 --> 00:01:32,000
Speaker 2: if you were I don't want to say, a lowly

28
00:01:32,200 --> 00:01:35,240
Speaker 2: utilities analyst, but you know, maybe a sort of forgotten

29
00:01:35,360 --> 00:01:39,120
Speaker 2: utilities analyst outside of your sector, suddenly you are very

30
00:01:39,200 --> 00:01:42,640
Speaker 2: in demand, right because all you hear about nowadays is

31
00:01:43,080 --> 00:01:46,679
Speaker 2: the AI build out and energy constraints on that, and

32
00:01:46,720 --> 00:01:48,800
Speaker 2: so obviously a lot of people want to look at

33
00:01:48,800 --> 00:01:50,320
Speaker 2: it from a utilities perspective.

34
00:01:50,720 --> 00:01:54,120
Speaker 3: Totally. I always think, like, what a great luck that

35
00:01:54,200 --> 00:01:56,600
Speaker 3: some people have in their careers. You know, you could

36
00:01:56,600 --> 00:01:59,080
Speaker 3: be an analyst and learn modeling skills and all kinds

37
00:01:59,080 --> 00:02:02,160
Speaker 3: of stuff, and then you get allocated and someone gets

38
00:02:02,240 --> 00:02:07,000
Speaker 3: allocated I don't know, farm equipment, and another person gets

39
00:02:07,000 --> 00:02:10,399
Speaker 3: allocated to they wind up in utilities in twenty twenty two,

40
00:02:10,440 --> 00:02:12,639
Speaker 3: and it's like, man, they're on TV all the time.

41
00:02:12,680 --> 00:02:16,000
Speaker 3: My old boss at Business Insider, Henry Blodge, it it's

42
00:02:16,000 --> 00:02:18,160
Speaker 3: like he was there as an Internet enalyst in like

43
00:02:18,200 --> 00:02:20,640
Speaker 3: the late nineties. What amazing timing and luck and.

44
00:02:20,560 --> 00:02:23,600
Speaker 2: So it's like journalism and the Yeah, it's it's the

45
00:02:23,639 --> 00:02:26,560
Speaker 2: exact same thing. I started out covering airlines, yeah, of

46
00:02:26,600 --> 00:02:29,720
Speaker 2: all things, but those were interesting. Anyway, I'm glad to

47
00:02:29,720 --> 00:02:31,600
Speaker 2: say we do, in fact have the perfect guests. So

48
00:02:31,639 --> 00:02:34,720
Speaker 2: we're going to be speaking to a utilities analyst, someone

49
00:02:34,720 --> 00:02:38,040
Speaker 2: who happens to have a very contrarian take on the

50
00:02:38,120 --> 00:02:41,800
Speaker 2: data center build out and how much energy is actually required.

51
00:02:41,840 --> 00:02:43,760
Speaker 2: We've been hearing a lot from people who are very,

52
00:02:43,840 --> 00:02:47,240
Speaker 2: very bullish on the data center build out, so this

53
00:02:47,280 --> 00:02:49,000
Speaker 2: will be a useful I love their point.

54
00:02:49,200 --> 00:02:49,680
Speaker 3: I love it.

55
00:02:49,800 --> 00:02:53,800
Speaker 2: Okay, So, without further ado, Andy Devreese, head of Investment

56
00:02:53,800 --> 00:02:56,760
Speaker 2: Grade Credit and head of Utilities and power over at

57
00:02:56,760 --> 00:02:59,040
Speaker 2: Credit Sides, thank you so much for coming on all thoughts.

58
00:02:59,320 --> 00:03:00,600
Speaker 4: Thank you the pleasure mine.

59
00:03:00,840 --> 00:03:03,400
Speaker 2: So is it great to be a utilities analyst right now?

60
00:03:03,400 --> 00:03:04,040
Speaker 2: Even better?

61
00:03:04,400 --> 00:03:07,080
Speaker 4: Well your your Bloomberg News reporter Josh Saul wrote an

62
00:03:07,120 --> 00:03:09,320
Speaker 4: article about the how much it's changed for being a

63
00:03:09,400 --> 00:03:11,640
Speaker 4: utilities analyst now the data centers are here. But to

64
00:03:11,680 --> 00:03:14,240
Speaker 4: push back, we did have the largest bankruptcy of all

65
00:03:14,280 --> 00:03:16,960
Speaker 4: time in Enron, Oh yeah, the largest LBO of all

66
00:03:17,000 --> 00:03:20,560
Speaker 4: time in TXU, which then went bankrupt, and the largest

67
00:03:20,600 --> 00:03:23,679
Speaker 4: private equity return ever in Calpine twenty five billion, which

68
00:03:23,680 --> 00:03:27,640
Speaker 4: exceeds Apollos, Lioned, l Trade and Blackstones Hilton Trade. So

69
00:03:27,840 --> 00:03:29,239
Speaker 4: we have had a lot of fun along the way.

70
00:03:29,400 --> 00:03:32,440
Speaker 2: Have you been a utilities analyst throughout that entire timeline?

71
00:03:32,440 --> 00:03:33,560
Speaker 2: How long have you been doing it?

72
00:03:33,760 --> 00:03:36,440
Speaker 4: I started with the first packcast bankruptcy and then went

73
00:03:36,440 --> 00:03:38,200
Speaker 4: through the second and here we are with data centers,

74
00:03:38,200 --> 00:03:39,560
Speaker 4: so twenty five years.

75
00:03:39,600 --> 00:03:42,200
Speaker 3: Wow, So you really have seen it all. It is

76
00:03:42,240 --> 00:03:44,920
Speaker 3: fair to say. So, you're absolutely right. There have been

77
00:03:45,000 --> 00:03:50,440
Speaker 3: some disasters and home runs and utilities. You know, they

78
00:03:50,480 --> 00:03:53,880
Speaker 3: do get central in the news, obviously with the fires

79
00:03:54,000 --> 00:03:57,680
Speaker 3: that we saw for example California several years ago, and

80
00:03:57,720 --> 00:04:02,880
Speaker 3: the court trails about allocation of risk in those situations.

81
00:04:03,040 --> 00:04:06,520
Speaker 3: You mentioned ron et cetera. But it is also fair

82
00:04:06,560 --> 00:04:09,520
Speaker 3: to say that much of the discourse in day to

83
00:04:09,600 --> 00:04:13,080
Speaker 3: day has been like these are sort of bond like instruments. Absolutely,

84
00:04:13,240 --> 00:04:14,720
Speaker 3: before we get into that, just like sort of like

85
00:04:14,760 --> 00:04:17,839
Speaker 3: talk to us about what a normal day is like,

86
00:04:17,960 --> 00:04:19,400
Speaker 3: and when you're thinking about you til.

87
00:04:19,320 --> 00:04:21,600
Speaker 4: It before you know, pre data centers, yeah, pre.

88
00:04:21,560 --> 00:04:23,800
Speaker 3: Data centers, five years ago, whatever, Pre data centers.

89
00:04:23,839 --> 00:04:25,920
Speaker 4: You're looking at a lot of rate cases of studying,

90
00:04:25,960 --> 00:04:28,600
Speaker 4: a lot of local news. You're looking at legislation, you're

91
00:04:28,640 --> 00:04:32,719
Speaker 4: reading you know, dry regulatory documents, and then you're tracking

92
00:04:32,800 --> 00:04:34,839
Speaker 4: natural gas prices because that's setting the price of power.

93
00:04:35,080 --> 00:04:36,960
Speaker 4: And then on the federal level you obviously have the

94
00:04:37,080 --> 00:04:40,880
Speaker 4: renewables displacing coal and that's obviously having a big impact now.

95
00:04:40,920 --> 00:04:42,560
Speaker 4: So it's a lot I actually think it's a lot

96
00:04:42,560 --> 00:04:42,880
Speaker 4: of fun.

97
00:04:43,320 --> 00:04:46,400
Speaker 2: My impression was always the policy aspect of it seemed

98
00:04:46,480 --> 00:04:49,320
Speaker 2: kind of the most important thing to keep track of.

99
00:04:49,400 --> 00:04:52,799
Speaker 4: Is that right, absolutely, and that's down to the state level,

100
00:04:52,839 --> 00:04:54,160
Speaker 4: but also the federal as well.

101
00:04:54,279 --> 00:04:57,200
Speaker 2: Yeah, we've done a few energy episodes, still trying to

102
00:04:57,200 --> 00:05:00,359
Speaker 2: wrap my head around the sort of patchwork of that

103
00:05:00,400 --> 00:05:03,719
Speaker 2: seem to cover our energy infrastructure. But anyway, who should

104
00:05:03,760 --> 00:05:03,919
Speaker 2: be a.

105
00:05:03,920 --> 00:05:07,040
Speaker 3: Little aspy about the way? Like you talk to people

106
00:05:07,080 --> 00:05:08,599
Speaker 3: in this space and you're like, well, how is this

107
00:05:08,680 --> 00:05:11,640
Speaker 3: get priced? And they're like, well, are you talking about

108
00:05:11,680 --> 00:05:14,359
Speaker 3: market or central? It's like, okay, I don't know, you know,

109
00:05:14,400 --> 00:05:15,960
Speaker 3: it's like are you talking about a rate board or

110
00:05:16,040 --> 00:05:17,240
Speaker 3: you talk about market prices.

111
00:05:17,279 --> 00:05:20,320
Speaker 2: It's so hard anyway, especially in a forty minute podcast,

112
00:05:20,440 --> 00:05:21,719
Speaker 2: to try to generalize that.

113
00:05:21,880 --> 00:05:23,680
Speaker 3: But I love the people, don't get me wrong, those

114
00:05:23,680 --> 00:05:24,520
Speaker 3: are my favorite people.

115
00:05:24,600 --> 00:05:27,120
Speaker 2: Anyway, Andy, what is the mood like at the moment?

116
00:05:27,279 --> 00:05:32,960
Speaker 2: Among utilities, people, analysts, investors. Wasn't there a conference recently?

117
00:05:33,080 --> 00:05:36,159
Speaker 4: Everyone's gung ho on this. So the biggest conference of

118
00:05:36,200 --> 00:05:40,720
Speaker 4: the year is EI in November and it was packed.

119
00:05:40,760 --> 00:05:43,680
Speaker 4: I was standing and rolling for some of these presentations. Wow,

120
00:05:44,240 --> 00:05:47,720
Speaker 4: your competitors at CNBC were broadcasting from the floor the

121
00:05:47,760 --> 00:05:49,799
Speaker 4: first time. That has probably a sign of a top.

122
00:05:50,760 --> 00:05:52,480
Speaker 4: So yeah, people are very happy, and it just to

123
00:05:52,560 --> 00:05:55,640
Speaker 4: quantify it. Utilities have generally grown around four or five

124
00:05:55,760 --> 00:05:58,120
Speaker 4: six percent a year, and then that's moved to five

125
00:05:58,160 --> 00:06:00,680
Speaker 4: to seven percent a year, and now certain names which

126
00:06:00,680 --> 00:06:02,920
Speaker 4: we'll talk about later are up to eight percent a year,

127
00:06:02,920 --> 00:06:04,520
Speaker 4: and that's driven by data center growth.

128
00:06:05,040 --> 00:06:07,760
Speaker 3: Talk to us a little bit more about that. So

129
00:06:08,080 --> 00:06:11,080
Speaker 3: what is the I mean, I'm like looking at a

130
00:06:11,200 --> 00:06:14,680
Speaker 3: chart of the xl U ETF. I don't think it's

131
00:06:14,720 --> 00:06:17,520
Speaker 3: like done insane, But talk to us a little bit about,

132
00:06:17,560 --> 00:06:22,159
Speaker 3: like maybe quantify the exuberance for us. So it's like, okay,

133
00:06:22,160 --> 00:06:24,680
Speaker 3: we we are no longer just in the business of

134
00:06:24,800 --> 00:06:28,440
Speaker 3: measuring bond proxies and looking at policies, et cetera. There

135
00:06:28,520 --> 00:06:31,800
Speaker 3: was a secular growth driver. Talk to us about like

136
00:06:32,040 --> 00:06:35,240
Speaker 3: the bowl case, and then also how we would see

137
00:06:35,480 --> 00:06:38,200
Speaker 3: the ball case sort of like how it's manifesting into

138
00:06:38,240 --> 00:06:39,159
Speaker 3: tradable instruments.

139
00:06:39,200 --> 00:06:41,440
Speaker 4: Sure, and you pulled up the graph of the XLU.

140
00:06:41,560 --> 00:06:44,039
Speaker 4: Obviously it's still a very interest rate sensitive section. Yeah,

141
00:06:44,040 --> 00:06:46,680
Speaker 4: that's dividends. You can find higher yields and other fixed

142
00:06:46,680 --> 00:06:49,080
Speaker 4: income instruments. So you know, maybe didn't do so well

143
00:06:49,160 --> 00:06:52,920
Speaker 4: last year. But the industry argues that as this EPs

144
00:06:52,960 --> 00:06:56,160
Speaker 4: growth rate goes up to the high single digits, mid

145
00:06:56,200 --> 00:06:58,240
Speaker 4: to high single digits, that it shouldn't be as interest

146
00:06:58,279 --> 00:07:00,560
Speaker 4: rate sensitive. So that's the big debate going on with

147
00:07:00,640 --> 00:07:02,240
Speaker 4: investors right now. And this talk.

148
00:07:02,440 --> 00:07:05,000
Speaker 3: I mean, the exo you has done well to ya.

149
00:07:05,200 --> 00:07:07,680
Speaker 2: If you zoom out, it looks pretty good. That's my

150
00:07:07,760 --> 00:07:11,320
Speaker 2: impression of a bitcoin investor, by the way, zoom outoom.

151
00:07:10,840 --> 00:07:12,640
Speaker 3: All right, think about you've gotten the end. You've gotten

152
00:07:12,640 --> 00:07:15,560
Speaker 3: that coupon, right, so like you get so like you

153
00:07:15,640 --> 00:07:17,680
Speaker 3: might in normal time, just be happy with a coupon.

154
00:07:17,840 --> 00:07:19,120
Speaker 3: You're getting coupon plus.

155
00:07:19,240 --> 00:07:21,080
Speaker 4: I mean, it worked out the Fed zero interest rates

156
00:07:21,080 --> 00:07:22,840
Speaker 4: for so long, so utilities are the place to be.

157
00:07:22,960 --> 00:07:25,200
Speaker 4: That's just math. And then as soon as the Fed

158
00:07:25,240 --> 00:07:28,240
Speaker 4: starts jocking uprates, chat TBT comes on the scene and

159
00:07:28,240 --> 00:07:29,960
Speaker 4: all of a sudden there's data centers. So now, all

160
00:07:29,960 --> 00:07:31,679
Speaker 4: of a sudden, you're taking the leg up on growth

161
00:07:31,760 --> 00:07:33,640
Speaker 4: when you don't need to be so interest rate sensitive.

162
00:07:33,920 --> 00:07:36,160
Speaker 2: Interesting. So one of the reasons we wanted to talk

163
00:07:36,200 --> 00:07:39,560
Speaker 2: to you is because you have that contrarian take on

164
00:07:39,760 --> 00:07:42,320
Speaker 2: the data center build out, and we wrote it up

165
00:07:42,360 --> 00:07:45,320
Speaker 2: in the All Thoughts newsletter, which everyone should subscribe to.

166
00:07:45,400 --> 00:07:50,640
Speaker 2: It got a lot of attention. Your analysis, interestingly, is

167
00:07:51,080 --> 00:07:55,000
Speaker 2: just based on some pretty simple math. So why don't

168
00:07:55,040 --> 00:07:57,640
Speaker 2: you just to start out with why don't you walk

169
00:07:57,680 --> 00:08:00,400
Speaker 2: us through the calculations that you're actually making to try

170
00:08:00,400 --> 00:08:04,520
Speaker 2: to analyze how much capacity the utilities are taking on

171
00:08:04,880 --> 00:08:07,239
Speaker 2: to actually power data centers.

172
00:08:07,520 --> 00:08:10,640
Speaker 4: Sure, so, as you said, it's pretty simple math here,

173
00:08:11,080 --> 00:08:13,760
Speaker 4: so utility, it's so data centers now are consuming around

174
00:08:13,800 --> 00:08:16,240
Speaker 4: forty five gigawatts of power, and you can switch between

175
00:08:16,240 --> 00:08:19,400
Speaker 4: capacity and throughput. I'm going to stick with capacity, so

176
00:08:19,440 --> 00:08:22,520
Speaker 4: forty five gigabats of power. And then there's lots and

177
00:08:22,560 --> 00:08:24,680
Speaker 4: lots of third party estimates for where they're going to

178
00:08:24,720 --> 00:08:27,720
Speaker 4: be in twenty thirty and they center around this, you know,

179
00:08:27,840 --> 00:08:31,000
Speaker 4: ninety ninety five gigawatts, so you need to add fifty

180
00:08:31,080 --> 00:08:34,440
Speaker 4: for two thousand and thirty five, there's a lot fewer estimates.

181
00:08:34,480 --> 00:08:38,200
Speaker 4: You come around one sixty. Now, these estimates, they you know,

182
00:08:38,240 --> 00:08:41,319
Speaker 4: they're all over the place. They come from sell side banks,

183
00:08:41,320 --> 00:08:43,560
Speaker 4: they come from consultants, they come from everyone. Bn EF

184
00:08:43,600 --> 00:08:45,360
Speaker 4: has one there, I think, one of the best out there,

185
00:08:46,080 --> 00:08:51,280
Speaker 4: so we use them a lot. So that's on the

186
00:08:51,320 --> 00:08:53,640
Speaker 4: demand side and where you're going to come out of these.

187
00:08:53,679 --> 00:08:55,400
Speaker 4: And then you look at the supply, and everyone talks

188
00:08:55,400 --> 00:08:57,680
Speaker 4: about the demand, right, But then you look at the

189
00:08:57,679 --> 00:09:01,079
Speaker 4: supply and all these tech are too cool to actually

190
00:09:01,160 --> 00:09:04,000
Speaker 4: look at the supply and do utility analysis, right, Who

191
00:09:04,040 --> 00:09:05,800
Speaker 4: wants to be a utility analysts? You were making fun

192
00:09:05,800 --> 00:09:07,959
Speaker 4: of us before, but if you look at this.

193
00:09:09,480 --> 00:09:11,920
Speaker 3: But when we realize our pity is misplaced. But we

194
00:09:11,920 --> 00:09:13,400
Speaker 3: were not making fun anyway, it's great.

195
00:09:13,600 --> 00:09:15,640
Speaker 4: So you look at the supply and these utilities are

196
00:09:15,679 --> 00:09:18,360
Speaker 4: tracking all these data centers connecting to the grid because

197
00:09:18,360 --> 00:09:19,880
Speaker 4: they've got to do a lot of work, spend a

198
00:09:19,880 --> 00:09:23,600
Speaker 4: lot of money and transmission distribution, new substations, transformers. It's

199
00:09:23,640 --> 00:09:25,800
Speaker 4: a lot of work, but a boost the earnings growth,

200
00:09:25,840 --> 00:09:28,320
Speaker 4: so they're happy to talk about this. And so you

201
00:09:28,360 --> 00:09:30,959
Speaker 4: look at where they're at and where they see things coming,

202
00:09:31,360 --> 00:09:34,120
Speaker 4: and they've got around one hundred and forty gigawatts a

203
00:09:34,200 --> 00:09:37,040
Speaker 4: near term supply. Now, kudos to the utilities. They break

204
00:09:37,080 --> 00:09:41,760
Speaker 4: out what's firm committed, signed, contracted versus pipeline behind it,

205
00:09:42,120 --> 00:09:44,360
Speaker 4: because there's a lot of double, triple, quadruple counting. So

206
00:09:44,360 --> 00:09:46,079
Speaker 4: if you're gonna build a data center in the Southeast,

207
00:09:46,120 --> 00:09:47,920
Speaker 4: you're gonna tell Duke, You're gonna tell Southern, You're gonna

208
00:09:47,960 --> 00:09:51,360
Speaker 4: tell Dominion you're gonna build one. So that's the pipeline potential.

209
00:09:52,040 --> 00:09:55,480
Speaker 4: But looking just at the firm committed whatever they want

210
00:09:55,480 --> 00:09:58,000
Speaker 4: to call it, you around one hundred and forty gigawats.

211
00:09:58,480 --> 00:10:00,480
Speaker 4: Now that's you got a pe adjust So when you

212
00:10:00,520 --> 00:10:03,480
Speaker 4: connected data center, yeah, when you connect the data center

213
00:10:03,520 --> 00:10:05,880
Speaker 4: to the grid, you've got lights, you've got cooling. Those

214
00:10:05,920 --> 00:10:08,800
Speaker 4: third party estimates I gave you are just for raw compute.

215
00:10:09,559 --> 00:10:11,319
Speaker 2: Why did you split those out though, because I mean,

216
00:10:11,360 --> 00:10:13,520
Speaker 2: all data centers are going to need to be cooled

217
00:10:13,559 --> 00:10:15,319
Speaker 2: down right, what's the point of splitting it out.

218
00:10:15,559 --> 00:10:17,680
Speaker 4: I'm not splitting out, I'm just adjusting it downward. Because

219
00:10:17,679 --> 00:10:20,000
Speaker 4: the third party estimates are just compute. So if you're

220
00:10:20,000 --> 00:10:22,959
Speaker 4: connecting to the grid. You're going to ask the lights,

221
00:10:22,960 --> 00:10:25,160
Speaker 4: the cooling and everything. So I want to go apples

222
00:10:25,160 --> 00:10:25,960
Speaker 4: to apples.

223
00:10:25,640 --> 00:10:30,280
Speaker 3: Versus the third party, So what power usage effectiveness or

224
00:10:30,320 --> 00:10:31,439
Speaker 3: power usage efficiency?

225
00:10:31,559 --> 00:10:34,040
Speaker 4: So they're at one forty, so that PUE is down

226
00:10:34,040 --> 00:10:37,040
Speaker 4: to one ten on apples to apples. So just to

227
00:10:37,080 --> 00:10:40,520
Speaker 4: go back, you only need fifty on the demand side

228
00:10:40,600 --> 00:10:43,480
Speaker 4: between now in twenty thirty and the utilities are working

229
00:10:43,520 --> 00:10:47,040
Speaker 4: at connecting one ten. So the utilities are working on

230
00:10:47,360 --> 00:10:49,800
Speaker 4: already connecting almost as much as you need by two

231
00:10:49,840 --> 00:10:52,640
Speaker 4: thy thirty five. So again just to make sure, on

232
00:10:52,640 --> 00:10:56,200
Speaker 4: the same page, third party estimates forty five gigabatser data

233
00:10:56,200 --> 00:11:00,280
Speaker 4: centers now going to ninety fives. That's fifty of these

234
00:11:00,280 --> 00:11:02,440
Speaker 4: are working on one to ten. They don't give timing

235
00:11:02,480 --> 00:11:04,640
Speaker 4: for that. Some of it's going to be past twenty thirty.

236
00:11:05,080 --> 00:11:07,040
Speaker 4: What I'm starting to say is there is a lot

237
00:11:07,200 --> 00:11:11,040
Speaker 4: of supply of data centers coming and it's very unclear

238
00:11:11,080 --> 00:11:13,720
Speaker 4: if there's going to be demand for this. So that's

239
00:11:13,800 --> 00:11:16,199
Speaker 4: the issue there, and then it might be worth pausing

240
00:11:16,200 --> 00:11:18,920
Speaker 4: that and just saying how we're tracking these things. Yeah,

241
00:11:19,080 --> 00:11:21,160
Speaker 4: So what we do for the demand side is we

242
00:11:21,280 --> 00:11:25,000
Speaker 4: use the original AI agent, you know, that is a

243
00:11:25,080 --> 00:11:28,920
Speaker 4: Gmail alert the best. So anything that's not on our

244
00:11:28,960 --> 00:11:31,600
Speaker 4: trade pubs, not on Bloomberg News, we get picked up

245
00:11:31,600 --> 00:11:34,079
Speaker 4: by a Gmail alert and so then we get all

246
00:11:34,120 --> 00:11:36,520
Speaker 4: that in a spreadsheet. So that's on the demand side.

247
00:11:36,880 --> 00:11:39,200
Speaker 4: And then on the supply side we use Diego and

248
00:11:39,240 --> 00:11:41,960
Speaker 4: that's not a large language model. That's my junior sitting

249
00:11:42,040 --> 00:11:44,520
Speaker 4: several blocks best of us right now, so he tracks

250
00:11:44,520 --> 00:11:47,120
Speaker 4: all this on utility calls. Just yesterday, next to Era

251
00:11:47,320 --> 00:11:51,000
Speaker 4: moved another two gigawatts from the potential into the committed.

252
00:11:51,559 --> 00:11:53,920
Speaker 4: And these utilities are chopping at the bit to sign

253
00:11:53,960 --> 00:11:55,520
Speaker 4: more and more of these deals, and I think it's

254
00:11:55,520 --> 00:11:58,000
Speaker 4: just going to be oversupply. We're going to overbuild these things.

255
00:11:58,440 --> 00:12:02,079
Speaker 2: Just to be clear, the commit the firm commitments, those

256
00:12:02,080 --> 00:12:06,360
Speaker 2: are signed agreements to actually build this capacity. Yes, okay.

257
00:12:06,440 --> 00:12:09,200
Speaker 4: And so I was talking with the CFO of Encore

258
00:12:09,280 --> 00:12:11,040
Speaker 4: and they made these comments on their call as well.

259
00:12:11,040 --> 00:12:13,800
Speaker 4: They're owned by Sempra, and I said, you know, no

260
00:12:13,880 --> 00:12:16,360
Speaker 4: one really believes these DAMA and estimates. Texas is a

261
00:12:16,559 --> 00:12:19,000
Speaker 4: wald off market, as you guys know, eighty seven gigawatt

262
00:12:19,000 --> 00:12:19,440
Speaker 4: peak market.

263
00:12:19,480 --> 00:12:21,760
Speaker 3: It is the one thing I know about energy Texas.

264
00:12:21,800 --> 00:12:22,880
Speaker 2: Is it market?

265
00:12:23,000 --> 00:12:23,160
Speaker 1: There?

266
00:12:23,160 --> 00:12:25,520
Speaker 4: You go, so eighty seven gigap peak market and the

267
00:12:25,559 --> 00:12:27,680
Speaker 4: demand estimates or they're going to add thirty gigawatts by

268
00:12:27,720 --> 00:12:30,360
Speaker 4: twenty thirty. And I said to the CFO of on cars,

269
00:12:30,440 --> 00:12:32,880
Speaker 4: there's just no way. He said, it might not be thirty,

270
00:12:32,880 --> 00:12:34,560
Speaker 4: but it's going to be closer to thirty than it

271
00:12:34,600 --> 00:12:37,840
Speaker 4: is zero. And I said, I just the forward power

272
00:12:37,880 --> 00:12:40,880
Speaker 4: curves don't reflect that at all, and he said, then

273
00:12:40,880 --> 00:12:44,560
Speaker 4: they're mispriced. So just for the entire market, just for

274
00:12:44,600 --> 00:12:47,600
Speaker 4: the benefit of your users, you cannot trade forward power

275
00:12:47,600 --> 00:12:50,680
Speaker 4: in Texas on interactive brokers. I know that. Oh that's fair, audience,

276
00:12:50,760 --> 00:12:52,640
Speaker 4: That's too bad. Everyone's like looking up there, That's what

277
00:12:52,720 --> 00:12:53,040
Speaker 4: I said.

278
00:12:53,040 --> 00:12:56,240
Speaker 3: Great, And this gives us a bunch of technical questions

279
00:12:56,240 --> 00:13:16,160
Speaker 3: to get into. Let's just keep talking about Texas. Explain

280
00:13:16,320 --> 00:13:19,800
Speaker 3: to us kind of what the forward power curves are

281
00:13:20,200 --> 00:13:24,600
Speaker 3: and how you can back out the implicit assumptions that

282
00:13:24,760 --> 00:13:29,199
Speaker 3: traders are making based on those forward power curves about

283
00:13:29,240 --> 00:13:30,640
Speaker 3: how much demand there's going to pay.

284
00:13:30,960 --> 00:13:33,319
Speaker 4: Sure, So I mean, obviously, if you're gonna go from

285
00:13:33,320 --> 00:13:35,640
Speaker 4: eighty seven and you're gonna add fifteen or thirty whatever,

286
00:13:35,640 --> 00:13:37,360
Speaker 4: it is you'd expect that curve to go higher?

287
00:13:37,920 --> 00:13:40,319
Speaker 3: How much the curve measuring? Okay, you say there's a

288
00:13:40,360 --> 00:13:41,640
Speaker 3: forward power, so the forward.

289
00:13:41,400 --> 00:13:43,800
Speaker 4: Power curve is a round the clock peak or off peak.

290
00:13:43,840 --> 00:13:46,480
Speaker 4: There's three separate curves, and the difference between peak and

291
00:13:46,520 --> 00:13:49,600
Speaker 4: off peak is actually you know, narrowed because data centers

292
00:13:49,679 --> 00:13:51,679
Speaker 4: run twenty four to seven, so it depends on your

293
00:13:51,679 --> 00:13:53,559
Speaker 4: North Texas or South Texas and those are in the

294
00:13:53,640 --> 00:13:54,360
Speaker 4: high fifties.

295
00:13:54,920 --> 00:13:57,760
Speaker 3: But what would we be seeing in the curves for like,

296
00:13:58,080 --> 00:14:02,000
Speaker 3: or is there trading happening every twenty ten.

297
00:14:02,120 --> 00:14:04,800
Speaker 4: Thirty might not be so liquid, but twenty seven and

298
00:14:04,840 --> 00:14:05,680
Speaker 4: twenty eight certainly are.

299
00:14:06,200 --> 00:14:08,719
Speaker 2: But there's other energy markets. So if you look at

300
00:14:08,880 --> 00:14:12,200
Speaker 2: nat gas for instance, although exactly gas traders are really

301
00:14:12,200 --> 00:14:15,320
Speaker 2: weird about the future's curve in gas, which I don't

302
00:14:15,360 --> 00:14:17,400
Speaker 2: really understand. But if you look at that, you point

303
00:14:17,400 --> 00:14:21,000
Speaker 2: out that over the longer term it's downward sloping, which

304
00:14:21,040 --> 00:14:23,600
Speaker 2: suggests that there isn't going to be as much demand

305
00:14:23,720 --> 00:14:25,360
Speaker 2: or maybe there's going to be more supply out in

306
00:14:25,400 --> 00:14:25,840
Speaker 2: the future.

307
00:14:26,160 --> 00:14:28,240
Speaker 4: I love it we're morphing into natural gas because that

308
00:14:28,360 --> 00:14:31,400
Speaker 4: is the main driver of power prices, especially in Texas

309
00:14:31,760 --> 00:14:34,280
Speaker 4: and the forward curve for gas is much more liquid

310
00:14:34,320 --> 00:14:36,480
Speaker 4: than it is for power, and the forward curve for

311
00:14:36,560 --> 00:14:39,200
Speaker 4: gas is inverted. It goes from three seventy to three

312
00:14:39,280 --> 00:14:42,160
Speaker 4: sixty by the end of the decade. So, as my

313
00:14:42,360 --> 00:14:45,600
Speaker 4: energy unlest Charles Johnson points out, the bigger driver there

314
00:14:45,680 --> 00:14:48,360
Speaker 4: isn't data center demand. We're at six pcf day there

315
00:14:48,600 --> 00:14:50,160
Speaker 4: a lot of people are at at ten twelve. We

316
00:14:50,200 --> 00:14:53,520
Speaker 4: can get into that. But LNG exports. We're exporting eighteen

317
00:14:53,560 --> 00:14:55,680
Speaker 4: bcf day. Now we're going to add another twelve. Like

318
00:14:55,720 --> 00:14:58,520
Speaker 4: you'd think that curve would be at least upward sloping

319
00:14:58,560 --> 00:15:01,480
Speaker 4: by twenty five cents. Cents by the way, you can

320
00:15:01,560 --> 00:15:05,400
Speaker 4: trade that and your attractive broker's account. So that goes

321
00:15:05,440 --> 00:15:06,920
Speaker 4: into is there going to be a lot of LNG.

322
00:15:07,080 --> 00:15:09,560
Speaker 4: It starts getting outside of the expertise. But that's what

323
00:15:09,640 --> 00:15:11,320
Speaker 4: we heard from a lot of clients last week when

324
00:15:11,320 --> 00:15:12,880
Speaker 4: we went on the road all over New York City.

325
00:15:13,040 --> 00:15:16,240
Speaker 3: This is very interesting. Actually want to ask another question

326
00:15:16,280 --> 00:15:19,520
Speaker 3: about the pure power curve. But since we are on LNG,

327
00:15:19,720 --> 00:15:21,360
Speaker 3: and then we can get back to the power curve,

328
00:15:21,680 --> 00:15:26,520
Speaker 3: just setting aside data centers, et cetera. Intuitively, you would

329
00:15:26,560 --> 00:15:29,280
Speaker 3: think that what is a growth business in the United

330
00:15:29,360 --> 00:15:32,440
Speaker 3: States LNG exports. And in fact, one of the sort

331
00:15:32,480 --> 00:15:36,080
Speaker 3: of policy debates around the whole question of building out

332
00:15:36,200 --> 00:15:39,960
Speaker 3: LNG terminals is it's going to make gas more expensive

333
00:15:40,000 --> 00:15:42,640
Speaker 3: for American consumers because now we're going to be competing

334
00:15:42,680 --> 00:15:46,600
Speaker 3: with European buyers, Whereas when we didn't have LNG export terminals,

335
00:15:46,680 --> 00:15:48,440
Speaker 3: we were just swimming in it because it had nowhere

336
00:15:48,480 --> 00:15:51,160
Speaker 3: to go, nowhere to go. So it's very interesting to

337
00:15:51,240 --> 00:15:55,720
Speaker 3: hear that, even with everyone acknowledging a booming domestic demand

338
00:15:56,120 --> 00:16:00,440
Speaker 3: and be the expansion of international demand that downward sloping

339
00:16:00,520 --> 00:16:01,080
Speaker 3: gas curve.

340
00:16:01,280 --> 00:16:05,440
Speaker 4: Yes, maybe, I don't know. Maybe there reflecting world peace

341
00:16:05,480 --> 00:16:08,880
Speaker 4: in Europe and Russia, LNG isnunacceptable to the rest of

342
00:16:08,880 --> 00:16:09,880
Speaker 4: the world. That could be a driver.

343
00:16:10,080 --> 00:16:11,640
Speaker 3: Then they'd have to rebuild that pipeline there.

344
00:16:11,720 --> 00:16:11,880
Speaker 1: Yeah.

345
00:16:11,920 --> 00:16:14,760
Speaker 4: But but back to our visual conversation on demand. The

346
00:16:14,800 --> 00:16:17,120
Speaker 4: reason I was talking to the Encore CFO and asking

347
00:16:17,200 --> 00:16:19,040
Speaker 4: him about this is he said he's holding two and

348
00:16:19,080 --> 00:16:22,840
Speaker 4: a half billion dollars of cash collateral postings from some

349
00:16:22,920 --> 00:16:26,120
Speaker 4: of that demand. And he's like, you're not some you know,

350
00:16:26,240 --> 00:16:28,240
Speaker 4: Joe Schno startup. I'm gonna build a data center and

351
00:16:28,280 --> 00:16:30,280
Speaker 4: connect to your grid if you're posting two and a

352
00:16:30,280 --> 00:16:32,120
Speaker 4: half billion dollars, and he says, and this is what

353
00:16:32,120 --> 00:16:34,080
Speaker 4: he said on their earnings call as well. You know

354
00:16:34,160 --> 00:16:37,480
Speaker 4: that's real demand that is coming. It's it's it's material.

355
00:16:37,760 --> 00:16:40,760
Speaker 3: I'm sorry, just now to go back to the power curve,

356
00:16:40,840 --> 00:16:43,080
Speaker 3: which I get is much less liquid out there. But

357
00:16:43,280 --> 00:16:47,760
Speaker 3: there are trades that happened. These are price, like, how

358
00:16:47,800 --> 00:16:51,640
Speaker 3: do you infer volume from price? Because these are price, we.

359
00:16:51,680 --> 00:16:54,240
Speaker 4: Don't get the volume. But it's a it's a yearly curve,

360
00:16:54,280 --> 00:16:55,960
Speaker 4: and then right before the year starts us listen to

361
00:16:56,000 --> 00:16:58,040
Speaker 4: twelve month curves and then it goes into weekly before.

362
00:16:58,160 --> 00:16:59,880
Speaker 3: What I'm saying is, how do you infer what it

363
00:17:00,080 --> 00:17:03,480
Speaker 3: expected volume in twenty twenty eighty. It goes from a price.

364
00:17:03,400 --> 00:17:06,040
Speaker 4: Curve, Sure it's flat, it goes up a dollar from

365
00:17:06,040 --> 00:17:08,280
Speaker 4: here to twenty thirty. Whereas if you're going to add

366
00:17:08,359 --> 00:17:10,560
Speaker 4: twenty percent to your group demand and you'd expect it

367
00:17:10,600 --> 00:17:13,360
Speaker 4: to go up, okay, several dollars a right gas side,

368
00:17:13,359 --> 00:17:15,160
Speaker 4: I'd want to see forty cents fifty cents.

369
00:17:15,200 --> 00:17:16,880
Speaker 3: I assue what you're saying, okay with pat.

370
00:17:17,040 --> 00:17:18,719
Speaker 4: And then just to go back to the fore your

371
00:17:18,800 --> 00:17:22,679
Speaker 4: upper fifties in Texas, low sixties for peak, and the

372
00:17:22,760 --> 00:17:25,560
Speaker 4: data center companies are paying ninety five. So Vistra just

373
00:17:25,640 --> 00:17:28,320
Speaker 4: did a deal off tenny five dollars a megab an hour, okay,

374
00:17:28,520 --> 00:17:31,400
Speaker 4: for round the clock. So Vistra contracted out its commandche

375
00:17:31,400 --> 00:17:33,640
Speaker 4: peak plant Texas ninety five dollars a megab an hour.

376
00:17:33,840 --> 00:17:36,919
Speaker 4: So big Tech is paying a very pretty penny. You

377
00:17:36,920 --> 00:17:39,040
Speaker 4: can argue some of that's for the CO two free

378
00:17:39,080 --> 00:17:41,320
Speaker 4: aspect of it, and some of it's just a lock

379
00:17:41,359 --> 00:17:41,879
Speaker 4: and the supply.

380
00:17:42,800 --> 00:17:44,720
Speaker 2: So just to go back to the math and your

381
00:17:44,760 --> 00:17:47,679
Speaker 2: overall argument, I mean, you're basically saying that utilities are

382
00:17:47,720 --> 00:17:51,520
Speaker 2: already committed to building out. I guess twice as much

383
00:17:51,560 --> 00:17:55,760
Speaker 2: capacity as is forecast to be needed by twenty thirty.

384
00:17:56,040 --> 00:17:59,760
Speaker 2: So wildcard to me seems to be the demand forecast, right,

385
00:18:00,119 --> 00:18:04,479
Speaker 2: and we're already seeing those change pretty wildly. I know

386
00:18:04,520 --> 00:18:07,520
Speaker 2: you mentioned Bloomberg NEF, but you know they've raised their

387
00:18:07,560 --> 00:18:10,720
Speaker 2: forecast because of the data center build out, so they've

388
00:18:10,760 --> 00:18:13,080
Speaker 2: raised their forecast of how much energy is actually needed.

389
00:18:13,640 --> 00:18:18,240
Speaker 2: How much confidence do you have in those demand numbers

390
00:18:18,240 --> 00:18:20,119
Speaker 2: and how could they change over time?

391
00:18:21,160 --> 00:18:23,520
Speaker 4: Moderate confidence. But like, look where we're at now, Like

392
00:18:24,520 --> 00:18:28,120
Speaker 4: open Aye built all of CHATCHYBT using two gigawatts. All

393
00:18:28,160 --> 00:18:30,560
Speaker 4: the big tech hyperscalers they haven't given their twenty twenty

394
00:18:30,560 --> 00:18:32,680
Speaker 4: five volumes yet, but if you take their twenty twenty

395
00:18:32,680 --> 00:18:35,479
Speaker 4: four volumes and then double it, and this is output.

396
00:18:35,520 --> 00:18:37,320
Speaker 4: So I'm going to transfer it back to capacity and

397
00:18:37,359 --> 00:18:40,879
Speaker 4: you assumer sixty percent capacity factor. All the hyperscalers combined

398
00:18:40,880 --> 00:18:43,640
Speaker 4: around fifteen gigawatts, and that's got to be over half

399
00:18:43,680 --> 00:18:47,560
Speaker 4: the data center demand. So to talk about ninety five gigawatts,

400
00:18:47,640 --> 00:18:50,600
Speaker 4: I mean it's a staggering number. And then you get

401
00:18:50,600 --> 00:18:54,119
Speaker 4: more advances in you know, Navidia chip efficiency. Yeah, obviously

402
00:18:54,200 --> 00:18:56,960
Speaker 4: Jevins paradox kicks in. You've had numerous guests talk about that.

403
00:18:57,560 --> 00:19:00,000
Speaker 4: It's just a lot of power, a lot of power.

404
00:19:00,200 --> 00:19:02,760
Speaker 2: Can you just remind us one gigawatt is enough to

405
00:19:02,840 --> 00:19:04,960
Speaker 2: power what I like these comparisons.

406
00:19:05,000 --> 00:19:07,000
Speaker 4: There's a million homes, but it depends if you're in

407
00:19:07,040 --> 00:19:09,639
Speaker 4: Florida or the Northeast, but generally speaking that's.

408
00:19:09,480 --> 00:19:13,800
Speaker 3: Where you're at. Not only do I find electricity markets,

409
00:19:13,840 --> 00:19:17,720
Speaker 3: so in market structure and electricity very difficulty at my head, around.

410
00:19:18,240 --> 00:19:21,720
Speaker 3: Even after all of these conversations, I have built no

411
00:19:22,440 --> 00:19:25,760
Speaker 3: heuristics or intuitions for what these A gigawat kill a

412
00:19:25,760 --> 00:19:27,919
Speaker 3: lot mega like you say these things, and I know

413
00:19:27,960 --> 00:19:30,200
Speaker 3: gigawats bigger than a kill a lot like what does

414
00:19:30,280 --> 00:19:32,760
Speaker 3: actually mean? And then the fact that even there we're

415
00:19:32,800 --> 00:19:35,000
Speaker 3: talking about the difference between a gigawatt and a gigwot

416
00:19:35,000 --> 00:19:37,760
Speaker 3: hour and the like, I've yet to develop the sort

417
00:19:37,800 --> 00:19:39,639
Speaker 3: of intuitions that I haven't.

418
00:19:39,720 --> 00:19:43,760
Speaker 4: Haven't you seen back to the future, Yeah, oh, there

419
00:19:43,760 --> 00:19:44,040
Speaker 4: you go.

420
00:19:44,320 --> 00:19:47,359
Speaker 2: We got to print out, you know, a little tape

421
00:19:47,520 --> 00:19:49,439
Speaker 2: like the way we used to do for credit ratings

422
00:19:49,600 --> 00:19:51,879
Speaker 2: the financial crisis. We need that up there. And actually

423
00:19:51,880 --> 00:19:54,679
Speaker 2: credit ratings are going to be interesting from a utilities

424
00:19:54,720 --> 00:19:55,560
Speaker 2: perspective as well.

425
00:19:55,720 --> 00:19:57,840
Speaker 3: Could I just ask, you know, obviously one of the

426
00:19:57,920 --> 00:20:01,000
Speaker 3: sensitivity is in general with all of thics, data center

427
00:20:01,040 --> 00:20:03,640
Speaker 3: and utilities, is this view, and I think it's kind

428
00:20:03,640 --> 00:20:06,760
Speaker 3: of overstated. Is the average ratepayer going to end up

429
00:20:07,240 --> 00:20:10,280
Speaker 3: paying for a lot of data centers or we'll raise

430
00:20:10,280 --> 00:20:13,280
Speaker 3: our electricity bill? And I understand, like these are complex

431
00:20:13,400 --> 00:20:15,920
Speaker 3: questions and the math isn't so clear. And also from

432
00:20:15,920 --> 00:20:18,399
Speaker 3: what I understand, the emergence of a data center can

433
00:20:18,440 --> 00:20:22,040
Speaker 3: actually lower the consumer's electricity bill because there's just that

434
00:20:22,200 --> 00:20:25,439
Speaker 3: simple math, which is, if there's more buyers splitting the

435
00:20:25,480 --> 00:20:28,640
Speaker 3: cost of the buildout, than actually your price tag can

436
00:20:28,800 --> 00:20:32,119
Speaker 3: go down. But in the scenario you're laying out in

437
00:20:32,160 --> 00:20:35,639
Speaker 3: which there's a bunch of upfront capital investments and everyone

438
00:20:35,680 --> 00:20:37,760
Speaker 3: is very excited to build it out and at cart

439
00:20:37,800 --> 00:20:40,199
Speaker 3: wires and you have to buy transformers and gear and

440
00:20:40,280 --> 00:20:44,040
Speaker 3: all this stuff, if the demand does not materialize is expected,

441
00:20:44,359 --> 00:20:47,400
Speaker 3: that does sound like conditions in which we could see

442
00:20:47,680 --> 00:20:49,280
Speaker 3: consumer rates.

443
00:20:48,960 --> 00:20:52,119
Speaker 4: Go up absolutely. So that's what we're spending all our

444
00:20:52,160 --> 00:20:54,520
Speaker 4: time on. And it's state by state, and even within

445
00:20:54,560 --> 00:20:57,639
Speaker 4: the same state, you've got numerous jurisdictions. So is it

446
00:20:57,720 --> 00:21:00,920
Speaker 4: legislatively mandated or is it done by a rate case

447
00:21:01,560 --> 00:21:04,160
Speaker 4: or in the case of Northern Indiana, have the companies

448
00:21:04,160 --> 00:21:07,080
Speaker 4: themselves data center companies themselves gotten ahead of it and

449
00:21:07,119 --> 00:21:08,760
Speaker 4: said we're going to put in a solution where rate

450
00:21:08,800 --> 00:21:12,679
Speaker 4: payers are absolutely protected and get money back. So you

451
00:21:12,680 --> 00:21:15,920
Speaker 4: look at nice source with Midwestern Utility. They own Northern

452
00:21:15,920 --> 00:21:19,800
Speaker 4: Indiana Public Service NIPSCOE, and they've got a deal where

453
00:21:19,800 --> 00:21:23,080
Speaker 4: they've got an inside rate base they've got a separate

454
00:21:23,160 --> 00:21:26,439
Speaker 4: gen Co. So they and that Genco is doing a

455
00:21:26,480 --> 00:21:29,320
Speaker 4: deal with Amazon and they're going to kick back a

456
00:21:29,400 --> 00:21:33,520
Speaker 4: billion dollars over fifteen years two rate payers. So rather

457
00:21:33,560 --> 00:21:35,960
Speaker 4: than have a debate, oh who's funding what it's like

458
00:21:36,600 --> 00:21:39,040
Speaker 4: done and you get sixty seven million dollars a year

459
00:21:39,480 --> 00:21:41,800
Speaker 4: and that's the blueprint, that's the gold standard.

460
00:21:41,920 --> 00:21:42,120
Speaker 3: Yeah.

461
00:21:42,200 --> 00:21:45,400
Speaker 4: I keep in mind, six months before that Genco was launched,

462
00:21:46,160 --> 00:21:49,480
Speaker 4: nice Or sold twenty percent of Nipsco to Blackstone. So

463
00:21:49,520 --> 00:21:51,600
Speaker 4: you could argue Blackstone said, hey, let's go ahead and

464
00:21:51,640 --> 00:21:54,480
Speaker 4: do this. And then the utility right north of Indiana

465
00:21:54,600 --> 00:21:57,120
Speaker 4: or northeast of the Indiana is Ohio, and the CEO

466
00:21:57,440 --> 00:22:00,639
Speaker 4: of First Energy is a next black Stone guy, So

467
00:22:00,680 --> 00:22:02,800
Speaker 4: maybe they look at doing a GENDO or something like that.

468
00:22:02,800 --> 00:22:05,960
Speaker 4: That's pure speculation. I have no idea pack gas specific

469
00:22:05,920 --> 00:22:08,080
Speaker 4: gas electric. They've done a deal where they've got rates

470
00:22:08,119 --> 00:22:11,639
Speaker 4: in place that protect presidential rate payers Amerin has, but

471
00:22:11,680 --> 00:22:13,439
Speaker 4: a lot of utilities don't. They don't have these protection.

472
00:22:13,520 --> 00:22:15,960
Speaker 3: The point is someone if it turns out that there's

473
00:22:15,960 --> 00:22:18,520
Speaker 3: an overbuild and there's not as much demand for it,

474
00:22:18,560 --> 00:22:20,320
Speaker 3: someone's paying for it, and it either it's going to

475
00:22:20,359 --> 00:22:22,359
Speaker 3: be the customers or perhaps the utility shareholders.

476
00:22:22,440 --> 00:22:24,440
Speaker 4: I mean, you just the political risk of having mom

477
00:22:24,480 --> 00:22:27,320
Speaker 4: and pop bail out, you know, Mark Szuckenberg, Jeff Bezos

478
00:22:27,400 --> 00:22:30,520
Speaker 4: is just you can't have that happen. But again, six

479
00:22:30,560 --> 00:22:32,600
Speaker 4: months ago this was coming up on the tail end

480
00:22:32,600 --> 00:22:36,040
Speaker 4: of conference calls, and now these utility CEOs are having

481
00:22:36,080 --> 00:22:38,359
Speaker 4: in the prepared remarks, so I'm pretty confident they're going

482
00:22:38,400 --> 00:22:39,360
Speaker 4: to figure it out.

483
00:22:40,040 --> 00:22:42,200
Speaker 2: You mentioned Blackstone just then. I do want to talk

484
00:22:42,240 --> 00:22:45,480
Speaker 2: about who is currently making a lot of money from

485
00:22:45,760 --> 00:22:48,120
Speaker 2: the data center build out, but just to stress test

486
00:22:48,200 --> 00:22:50,719
Speaker 2: the thesis a little bit more, because it is a

487
00:22:50,760 --> 00:22:52,920
Speaker 2: contrarian take, and so I think we should ask a

488
00:22:52,920 --> 00:22:56,080
Speaker 2: bunch of questions about it. But does it take into

489
00:22:56,160 --> 00:23:00,280
Speaker 2: account time lags for projects? So I think, you know,

490
00:23:00,640 --> 00:23:04,760
Speaker 2: capacity build out in the energy sector is notoriously bureaucratic.

491
00:23:04,840 --> 00:23:06,800
Speaker 2: That is one thing that Joe and I do actually

492
00:23:06,800 --> 00:23:09,520
Speaker 2: know about the sector. Is it possible that a lot

493
00:23:09,520 --> 00:23:14,040
Speaker 2: of these committed projects actually take much longer to get

494
00:23:14,280 --> 00:23:16,560
Speaker 2: working on the ground than currently forecast.

495
00:23:17,359 --> 00:23:19,840
Speaker 4: I think the delays will be on building the new generation,

496
00:23:20,040 --> 00:23:22,440
Speaker 4: not the data center. So data center takes two three years,

497
00:23:22,520 --> 00:23:24,439
Speaker 4: Even if that slips to four or five years. The

498
00:23:24,480 --> 00:23:26,879
Speaker 4: power plants take six seven years, and as you know,

499
00:23:27,200 --> 00:23:30,120
Speaker 4: you can't get a g Rnova gas turbine for years

500
00:23:30,119 --> 00:23:32,920
Speaker 4: and years, which is obviously a bullish backshop here.

501
00:23:33,119 --> 00:23:36,320
Speaker 3: Yeah, talk to us more about that element of it

502
00:23:36,480 --> 00:23:41,120
Speaker 3: all because building out that let's if you overshoot on production,

503
00:23:41,640 --> 00:23:44,119
Speaker 3: then that's a problem in itself. If you overshoot on

504
00:23:44,200 --> 00:23:47,120
Speaker 3: production at a time when it's gotten really expensive because

505
00:23:47,119 --> 00:23:49,840
Speaker 3: there's massive inflation in the construction sector, that's an even

506
00:23:49,880 --> 00:23:53,040
Speaker 3: greater problem. Talk to us just about like, per any

507
00:23:53,200 --> 00:23:57,280
Speaker 3: giving unit of productive capacity on the utility side, how

508
00:23:57,359 --> 00:23:59,520
Speaker 3: much more expensive is it gotten and what do you

509
00:23:59,800 --> 00:24:00,879
Speaker 3: think we're casting for that?

510
00:24:01,520 --> 00:24:03,840
Speaker 4: Sure, so to build a combined cycle gas plant ten

511
00:24:03,920 --> 00:24:06,560
Speaker 4: years ago is one thousand and twelve hundred kW to build.

512
00:24:06,600 --> 00:24:09,000
Speaker 4: Then it got to two thousand. Then utility analysts like myself,

513
00:24:09,000 --> 00:24:11,119
Speaker 4: I was like, WHOA, that's insane. Now we're up to

514
00:24:11,200 --> 00:24:13,639
Speaker 4: three thousand, and then it's like, who's actually spending this?

515
00:24:14,440 --> 00:24:16,920
Speaker 4: But that three thousand dollars a kW for a new

516
00:24:16,960 --> 00:24:20,720
Speaker 4: gas plant compares to the data center itself that cost

517
00:24:21,040 --> 00:24:24,560
Speaker 4: forty thousand dollars. So for Big Tech to spend another

518
00:24:24,680 --> 00:24:27,639
Speaker 4: three to lock in their gas price, its their fuel source.

519
00:24:27,680 --> 00:24:29,960
Speaker 4: It's like it's nothing. Oh, it's to Minimus, which goes

520
00:24:30,000 --> 00:24:32,679
Speaker 4: back to the output four powers fifty five sixty and

521
00:24:32,680 --> 00:24:35,600
Speaker 4: pick TeX's paying ninety five in the grand scheme of things.

522
00:24:35,640 --> 00:24:37,920
Speaker 4: So the cost of the data center, it's nothing. So that's

523
00:24:37,920 --> 00:24:42,000
Speaker 4: why our utility and just jaw dropping on how shocking

524
00:24:42,040 --> 00:24:43,960
Speaker 4: it is Big Tech's willing to pay these amounts.

525
00:24:44,280 --> 00:24:48,920
Speaker 3: What about if you don't measure production of new plants

526
00:24:48,960 --> 00:24:52,080
Speaker 3: and dollars, but new plants in time and again, if

527
00:24:52,080 --> 00:24:54,840
Speaker 3: you're talking about okay, well we can't get this turbine

528
00:24:55,200 --> 00:24:57,240
Speaker 3: that is several years ago we could have got delivered

529
00:24:57,280 --> 00:25:00,040
Speaker 3: next month. How much longer are these projects taken?

530
00:25:00,200 --> 00:25:02,360
Speaker 4: So if you can figure this out, I think you're

531
00:25:02,440 --> 00:25:04,040
Speaker 4: lusing this. If you figure this out, you can make

532
00:25:04,080 --> 00:25:06,960
Speaker 4: a lot of money. Because obviously Inflation Production Act had

533
00:25:07,320 --> 00:25:08,520
Speaker 4: enormous I'm going to vibe.

534
00:25:08,400 --> 00:25:10,200
Speaker 3: Code a thing to figure it out.

535
00:25:10,280 --> 00:25:14,719
Speaker 4: Keep going an inflation Production Act, enormous tax credits for renewables,

536
00:25:15,240 --> 00:25:17,520
Speaker 4: and then the one big beautiful bill. Obviously clip those

537
00:25:17,520 --> 00:25:21,080
Speaker 4: if you're not aligned by a certain date. So all

538
00:25:21,160 --> 00:25:23,600
Speaker 4: this data center numbers are weighted towards the end of

539
00:25:23,600 --> 00:25:27,040
Speaker 4: the decade, whereas the new solar is right here, right

540
00:25:27,080 --> 00:25:30,800
Speaker 4: now rushing power prices. So you actually want to be

541
00:25:30,840 --> 00:25:32,879
Speaker 4: a little short power for the next few years and

542
00:25:32,880 --> 00:25:35,680
Speaker 4: then flip to being long. And if you can figure

543
00:25:35,680 --> 00:25:37,959
Speaker 4: out when that flip is, you can make a lot

544
00:25:38,000 --> 00:25:39,879
Speaker 4: of money in either the forward power curves or the

545
00:25:39,960 --> 00:25:42,879
Speaker 4: natural gas curves. But as far as your original question,

546
00:25:42,920 --> 00:25:45,160
Speaker 4: as I said, data centers two three years to build

547
00:25:45,400 --> 00:25:49,159
Speaker 4: new power plants, four five, But then you don't need

548
00:25:49,200 --> 00:25:51,840
Speaker 4: as many new power plants as everyone's saying. So Constellation

549
00:25:52,000 --> 00:25:53,720
Speaker 4: CEO set on a call the other day he said

550
00:25:53,840 --> 00:25:56,720
Speaker 4: he used the Texas market. He said, eighty seven gigawat

551
00:25:56,720 --> 00:25:59,520
Speaker 4: peep market. You could add ten gigawatts to Texas tomorrow,

552
00:25:59,520 --> 00:26:02,480
Speaker 4: which would be the equivalent of sending every single Navidia

553
00:26:02,600 --> 00:26:05,439
Speaker 4: chip for an entire year to Texas and running them

554
00:26:05,440 --> 00:26:08,000
Speaker 4: twenty four to seven. That's ten gigawatts because you can

555
00:26:08,119 --> 00:26:10,720
Speaker 4: run it right now, existing grid, existing plants, for all

556
00:26:10,720 --> 00:26:13,960
Speaker 4: but forty fifty hours a year. We stress tested it.

557
00:26:14,000 --> 00:26:16,280
Speaker 4: There are some coal plants that can ramp up capacity factor.

558
00:26:16,320 --> 00:26:18,080
Speaker 4: There's plenty of gas plants that can. So I don't

559
00:26:18,080 --> 00:26:20,000
Speaker 4: know if it's forty hours, one hundred hours, one hundred

560
00:26:20,000 --> 00:26:22,720
Speaker 4: and fifty hours, But it makes more sense to pay

561
00:26:23,040 --> 00:26:26,720
Speaker 4: someone else not to run their chemical company, their refinery

562
00:26:26,720 --> 00:26:29,480
Speaker 4: company for forty fifty hours a year, rather than have

563
00:26:29,560 --> 00:26:32,680
Speaker 4: the utilities go out and spend ten billion dollars connecting

564
00:26:32,720 --> 00:26:35,639
Speaker 4: far away wind farms. That's the argument. We're sort of

565
00:26:35,640 --> 00:26:37,880
Speaker 4: come in the middle of it. But there is plenty

566
00:26:37,920 --> 00:26:40,800
Speaker 4: of existing capacity on the grid that could ramp up

567
00:26:40,800 --> 00:26:43,119
Speaker 4: to meet it and then have other guests to point

568
00:26:43,119 --> 00:26:45,280
Speaker 4: it out on the odd bots, you know, the peak

569
00:26:45,320 --> 00:26:47,840
Speaker 4: demand of the grid is eight hundred and fifty gigawatts,

570
00:26:48,520 --> 00:26:51,080
Speaker 4: the overall size of the grid is are twelve hundred gigawatts,

571
00:26:51,960 --> 00:26:54,600
Speaker 4: and then you're adding fifty gigawatts a year of solar,

572
00:26:54,880 --> 00:26:57,119
Speaker 4: and then you're gonna start adding twenty gigawatts of gas.

573
00:26:57,160 --> 00:26:59,160
Speaker 4: I mean, we're gonna handle it. I'm not really worried

574
00:26:59,160 --> 00:27:00,879
Speaker 4: about any rochesters or anything.

575
00:27:00,920 --> 00:27:03,720
Speaker 2: Oh yeah, talk to us about regional transmission, because this

576
00:27:03,840 --> 00:27:05,919
Speaker 2: is something that we hear a lot. It's not necessarily

577
00:27:05,960 --> 00:27:09,560
Speaker 2: the power generation that's an issue here. It's the transmission,

578
00:27:09,560 --> 00:27:12,000
Speaker 2: which the US you know, seems to struggle with to

579
00:27:12,000 --> 00:27:12,720
Speaker 2: put it mildly.

580
00:27:14,080 --> 00:27:19,080
Speaker 4: So there's regional markets MISO, Midwest, the Mid Continent.

581
00:27:19,920 --> 00:27:20,320
Speaker 3: ISO.

582
00:27:21,320 --> 00:27:23,240
Speaker 4: These guys are tired the most amount of coal, so

583
00:27:23,280 --> 00:27:25,240
Speaker 4: I think they're going to be in the worst shape.

584
00:27:25,280 --> 00:27:27,879
Speaker 4: And then Texas, and then it depends if anyone builds

585
00:27:27,920 --> 00:27:30,320
Speaker 4: anything in New England. New England's got the faraway more

586
00:27:30,400 --> 00:27:32,880
Speaker 4: expensive power prices seventy dollars the rest of the countries.

587
00:27:33,119 --> 00:27:34,280
Speaker 2: You know, I'm well aware.

588
00:27:34,840 --> 00:27:37,800
Speaker 4: Yes I have two because I live in Connecticut. So

589
00:27:38,080 --> 00:27:40,040
Speaker 4: if anyone builds a data center in New England, they're

590
00:27:40,040 --> 00:27:41,560
Speaker 4: going to be the tightest. But after that it's really

591
00:27:41,600 --> 00:27:42,120
Speaker 4: MISO and.

592
00:27:42,119 --> 00:27:44,840
Speaker 3: No one's building data centers in Vermont where they occasionally

593
00:27:44,840 --> 00:27:46,880
Speaker 3: have to switch over to oil and wood, right.

594
00:27:47,200 --> 00:27:50,000
Speaker 4: I mean they ISO New England app, which we all

595
00:27:50,000 --> 00:27:53,080
Speaker 4: have on our phone. Right, they were getting forty percent

596
00:27:53,080 --> 00:27:54,280
Speaker 4: of their power from oil.

597
00:27:54,440 --> 00:27:54,600
Speaker 1: Yeah.

598
00:27:54,680 --> 00:27:57,680
Speaker 4: Yeah, the cold snap the other day and it's tough

599
00:27:57,720 --> 00:27:59,760
Speaker 4: to talk about New England power without talking politics, and

600
00:27:59,760 --> 00:28:03,439
Speaker 4: we're not going to go down that. So transmission is

601
00:28:03,480 --> 00:28:05,200
Speaker 4: very important because you've got to connect all these far

602
00:28:05,200 --> 00:28:06,520
Speaker 4: away renewables to the grid.

603
00:28:06,680 --> 00:28:09,000
Speaker 3: You said something that I think is actually kind of important.

604
00:28:09,040 --> 00:28:12,480
Speaker 3: There's this narrative meme you know, people talking about the

605
00:28:12,520 --> 00:28:16,119
Speaker 3: AI race US versus China, and that one of the

606
00:28:16,160 --> 00:28:19,120
Speaker 3: things I've seen people say China is going to win

607
00:28:19,240 --> 00:28:22,280
Speaker 3: because they could just build out power more easily than

608
00:28:22,320 --> 00:28:24,840
Speaker 3: we can. It sounds like, I know you're not an

609
00:28:24,880 --> 00:28:28,600
Speaker 3: ai an list, but sounds like from your perspective, we

610
00:28:28,680 --> 00:28:30,440
Speaker 3: don't know like what it means or who's going to

611
00:28:30,520 --> 00:28:33,840
Speaker 3: win US versus China. But then from your perspective, power

612
00:28:33,920 --> 00:28:35,640
Speaker 3: is not going to be the decider.

613
00:28:35,240 --> 00:28:38,160
Speaker 4: Here, not in China. It's not, but it doesn't.

614
00:28:38,200 --> 00:28:40,320
Speaker 3: You said, like, you know that we could ship every

615
00:28:40,600 --> 00:28:43,600
Speaker 3: current with existing capacity. We could put every in video

616
00:28:43,640 --> 00:28:46,360
Speaker 3: chip in Texas today and we could run them for

617
00:28:46,400 --> 00:28:48,840
Speaker 3: fifty hours for all the all exception of a few

618
00:28:48,880 --> 00:28:50,480
Speaker 3: really hot hours in the summer.

619
00:28:50,520 --> 00:28:53,920
Speaker 2: I like the imagery of all the invidio chips going

620
00:28:53,960 --> 00:28:55,720
Speaker 2: on a field trip to Texas, Texas.

621
00:28:55,760 --> 00:28:58,840
Speaker 3: But it sounds like to your view, that really isn't

622
00:28:58,880 --> 00:29:00,840
Speaker 3: going to be from the US perspective, that won't be

623
00:29:00,880 --> 00:29:01,840
Speaker 3: the binding constraint.

624
00:29:02,160 --> 00:29:04,400
Speaker 4: It's it's going to be a little tight. But I'm

625
00:29:04,440 --> 00:29:06,640
Speaker 4: not one of these dooms theors. Oh, it's the absolute

626
00:29:06,720 --> 00:29:09,400
Speaker 4: gating factor. It's all going to stop. I was in Shenzhen,

627
00:29:09,520 --> 00:29:13,160
Speaker 4: China last year, and a robot got in one floor

628
00:29:13,200 --> 00:29:15,400
Speaker 4: and the elevator went up and got off another one, Like,

629
00:29:15,440 --> 00:29:16,400
Speaker 4: what is going on here?

630
00:29:31,720 --> 00:29:34,400
Speaker 2: There's another reason we wanted to talk to you, aside

631
00:29:34,520 --> 00:29:39,040
Speaker 2: from your capacity analysis, which is one of the interesting

632
00:29:39,080 --> 00:29:41,960
Speaker 2: things that's been happening in the credit market is obviously

633
00:29:42,440 --> 00:29:45,920
Speaker 2: private credit has been a big story for the past

634
00:29:45,920 --> 00:29:48,640
Speaker 2: few years, but now private credit is getting in on

635
00:29:49,080 --> 00:29:52,640
Speaker 2: the data center build out as well. There's sort of,

636
00:29:52,680 --> 00:29:55,760
Speaker 2: I guess, getting on your turf a little bit in

637
00:29:55,800 --> 00:29:58,640
Speaker 2: the public bond market, But what sort of activity have

638
00:29:58,720 --> 00:29:59,320
Speaker 2: you seen there?

639
00:30:00,120 --> 00:30:01,840
Speaker 4: Sure, so we think that's where the risk is going

640
00:30:01,880 --> 00:30:04,880
Speaker 4: to is going to happen. And frankly, Bloomberg News broke

641
00:30:04,920 --> 00:30:08,440
Speaker 4: the story and Pimco made two billion dollars on day

642
00:30:08,520 --> 00:30:12,040
Speaker 4: one loaning to the Meta data center in Louisiana. So

643
00:30:12,080 --> 00:30:14,680
Speaker 4: they priced twenty five billion dollars debt debt at two

644
00:30:14,800 --> 00:30:17,160
Speaker 4: twenty over treasuries and it immediately started trading at one

645
00:30:17,280 --> 00:30:20,760
Speaker 4: forty and handed Pimco two billion. Great for Pimco, but

646
00:30:20,800 --> 00:30:23,200
Speaker 4: then everyone else nice to be Pimco. Is nice to

647
00:30:23,200 --> 00:30:27,240
Speaker 4: be Pimco, especially the weather in Newport Beach. But everyone

648
00:30:27,240 --> 00:30:29,000
Speaker 4: else in private credit is like, oh, these guys just

649
00:30:29,040 --> 00:30:31,120
Speaker 4: made two billion dollars, we need to start lending to

650
00:30:31,200 --> 00:30:33,600
Speaker 4: data centers. And we all know how this ends. Covenants

651
00:30:33,640 --> 00:30:37,600
Speaker 4: start falling, great start falling. And again, if you're big tech,

652
00:30:38,000 --> 00:30:40,320
Speaker 4: who cares if you overspend? Like you think AI would

653
00:30:40,320 --> 00:30:42,080
Speaker 4: be all end all, you've been overspend. It's when you

654
00:30:42,080 --> 00:30:44,560
Speaker 4: get down to the second tier. The QTS is the

655
00:30:44,640 --> 00:30:46,880
Speaker 4: vantages of the world, and then you get down to

656
00:30:46,920 --> 00:30:48,840
Speaker 4: sort of the ones below that, and you get like,

657
00:30:48,880 --> 00:30:51,240
Speaker 4: you know, the core weaves and the nebiuses of the world,

658
00:30:51,240 --> 00:30:53,200
Speaker 4: and you know there's a lot of shorts going out

659
00:30:53,240 --> 00:30:56,000
Speaker 4: on Equinix, and obviously your guest Jim Chanos, and it's

660
00:30:56,000 --> 00:30:57,360
Speaker 4: all about the chips. I'm not going to get into

661
00:30:57,360 --> 00:30:59,640
Speaker 4: the chip debate, but it's interesting you look at a

662
00:30:59,680 --> 00:31:02,200
Speaker 4: core even they got a fifty billion dollar market cap.

663
00:31:02,720 --> 00:31:04,360
Speaker 4: That's a real company. You're gonna be around for a

664
00:31:04,360 --> 00:31:06,959
Speaker 4: long time. But the bond market is saying we want

665
00:31:07,000 --> 00:31:09,640
Speaker 4: a ten percent yield to win you twenty and thirty paper.

666
00:31:09,840 --> 00:31:12,000
Speaker 4: You might not be a real company. And if you

667
00:31:12,000 --> 00:31:14,400
Speaker 4: look at our supply demand outlooks, we're kind of in

668
00:31:14,440 --> 00:31:17,040
Speaker 4: the camp of the bond market. But timing, which you

669
00:31:17,080 --> 00:31:20,360
Speaker 4: mentioned earlier, Joe, is so key because this data center

670
00:31:20,440 --> 00:31:22,280
Speaker 4: is you're gonna ramp for a couple of years, and

671
00:31:22,280 --> 00:31:24,760
Speaker 4: the oversupply is really a twenty thirty event. So good

672
00:31:24,800 --> 00:31:25,680
Speaker 4: luck timing that one.

673
00:31:25,840 --> 00:31:28,760
Speaker 3: You said something you talked about that Pimco Meta deal,

674
00:31:28,920 --> 00:31:31,640
Speaker 3: and this question has come up, and I still don't

675
00:31:31,680 --> 00:31:36,560
Speaker 3: think got a totally satisfactory answer to it. Meta is

676
00:31:36,600 --> 00:31:39,800
Speaker 3: a very highly rated company as you see it as

677
00:31:39,800 --> 00:31:42,960
Speaker 3: a credit analyst. What is it about the private credit

678
00:31:43,000 --> 00:31:44,960
Speaker 3: you know to talk about, Oh, it's flexible, et cetera.

679
00:31:45,120 --> 00:31:47,840
Speaker 3: But your two twenty spread over treasuries is not nothing

680
00:31:47,880 --> 00:31:51,000
Speaker 3: at all and is real? Is that two twenty spread

681
00:31:51,320 --> 00:31:54,160
Speaker 3: really like worth it for like a little bit more flexibility,

682
00:31:54,160 --> 00:31:57,840
Speaker 3: et cetera? Like what are they paying for exactly in

683
00:31:57,920 --> 00:32:00,640
Speaker 3: the private credit market that they couldn't get it cheaper?

684
00:32:00,720 --> 00:32:02,760
Speaker 3: I would think in the public bond market.

685
00:32:03,240 --> 00:32:06,040
Speaker 4: I don't know, but I could speculate. There's okay, couple

686
00:32:06,520 --> 00:32:08,160
Speaker 4: if the metal question is why did you put this

687
00:32:08,240 --> 00:32:10,720
Speaker 4: off balance sheet? Yeah, so you've got the state of

688
00:32:10,720 --> 00:32:14,120
Speaker 4: the art data center with the best Navidia chips out there,

689
00:32:14,640 --> 00:32:16,920
Speaker 4: and you're a tech company in AI is the be all,

690
00:32:17,000 --> 00:32:19,320
Speaker 4: end all for everything. Did you kick it off your

691
00:32:19,320 --> 00:32:22,680
Speaker 4: balance sheet because you didn't want to damage your balance sheet?

692
00:32:22,720 --> 00:32:25,080
Speaker 4: But the agencies are imputing it, but maybe quant funds

693
00:32:25,160 --> 00:32:27,880
Speaker 4: running their screens, they don't impute that, So maybe that helps.

694
00:32:28,640 --> 00:32:31,200
Speaker 4: Or maybe you didn't want the depreciation running through your

695
00:32:31,240 --> 00:32:34,560
Speaker 4: income statement, maybe that helps, or maybe you want to

696
00:32:34,600 --> 00:32:36,760
Speaker 4: walk from this thing in five years. I don't know,

697
00:32:37,120 --> 00:32:39,320
Speaker 4: but one of those is definitely the reasons, because why

698
00:32:39,320 --> 00:32:41,760
Speaker 4: else would you pay that much bigger spread one hundred

699
00:32:41,760 --> 00:32:43,360
Speaker 4: and fifty BIPs over their borrowing costs.

700
00:32:43,400 --> 00:32:45,440
Speaker 3: But so the key thing is here when you talk

701
00:32:45,480 --> 00:32:49,880
Speaker 3: about that pimcoat meta deal, technically this is not meta debt.

702
00:32:50,440 --> 00:32:53,560
Speaker 3: It is not it's okay, right, okay, So they create

703
00:32:53,640 --> 00:32:57,000
Speaker 3: a vehicle, they're not gonna okay, that's I think that's

704
00:32:57,040 --> 00:32:59,800
Speaker 3: an important element that they're not just arbitrarily paying a

705
00:32:59,800 --> 00:33:02,240
Speaker 3: lot more for like a sort.

706
00:33:02,080 --> 00:33:06,760
Speaker 4: Of and if you read the credit docs, they've guaranteed

707
00:33:06,760 --> 00:33:09,760
Speaker 4: this debt even if the data center shuts down. But

708
00:33:10,000 --> 00:33:12,280
Speaker 4: our understanding of the docks is if they sell it,

709
00:33:12,320 --> 00:33:14,880
Speaker 4: then the guarantee goes away, and so that would create

710
00:33:14,880 --> 00:33:17,840
Speaker 4: a little risk. But back to my utility routes. What

711
00:33:17,920 --> 00:33:20,320
Speaker 4: happens that the data center shuts down for rate payers,

712
00:33:20,760 --> 00:33:23,680
Speaker 4: and they actually have an explicit guarantee from META to

713
00:33:23,720 --> 00:33:26,960
Speaker 4: protect rate payers, So they have that a lot of

714
00:33:26,960 --> 00:33:32,000
Speaker 4: other utilities don't, so a lot of states Louisiana, Mississippi, Tennessee, Texas,

715
00:33:32,000 --> 00:33:34,320
Speaker 4: they need to do better job protecting their rate payers.

716
00:33:34,360 --> 00:33:35,840
Speaker 4: And by the way, that's just one line in the

717
00:33:35,920 --> 00:33:39,160
Speaker 4: dock that could fall away in other new data centers.

718
00:33:39,160 --> 00:33:40,720
Speaker 4: And that's what we spend our time looking at.

719
00:33:40,840 --> 00:33:44,240
Speaker 2: You mentioned the credit ratings just then, So the rating

720
00:33:44,280 --> 00:33:48,200
Speaker 2: agencies they look at the off balance sheet vehicles, even

721
00:33:48,240 --> 00:33:51,360
Speaker 2: though it's not officially part of the company's.

722
00:33:50,880 --> 00:33:54,440
Speaker 4: Debt, they impute the least payments and include that as debt.

723
00:33:54,960 --> 00:33:56,880
Speaker 3: I see, and.

724
00:33:56,840 --> 00:33:59,040
Speaker 4: For Meta specifically, they won't do that until the least

725
00:33:59,040 --> 00:34:01,479
Speaker 4: starts when it comes online, but everyone's doing it.

726
00:34:02,360 --> 00:34:04,040
Speaker 2: And then I was just thinking, I don't mean to

727
00:34:04,240 --> 00:34:07,520
Speaker 2: labor this analogy too much, but you know, you started

728
00:34:07,520 --> 00:34:10,839
Speaker 2: out by talking about all the exciting moments in the

729
00:34:10,880 --> 00:34:13,279
Speaker 2: history of being a utilities analyst, and one of those

730
00:34:13,400 --> 00:34:16,120
Speaker 2: was Enron, which I assume means you know, you have

731
00:34:16,400 --> 00:34:19,960
Speaker 2: some experience with circular deals. But what do you think

732
00:34:19,960 --> 00:34:25,399
Speaker 2: about all the sort of incestuous financing deals that seem

733
00:34:25,480 --> 00:34:28,080
Speaker 2: to be happening between all the various players in the

734
00:34:28,160 --> 00:34:29,400
Speaker 2: data center industry.

735
00:34:29,600 --> 00:34:31,399
Speaker 4: You mean, we'll buy your equity so you can buy

736
00:34:31,400 --> 00:34:31,920
Speaker 4: our chips.

737
00:34:32,000 --> 00:34:32,719
Speaker 3: Yeah.

738
00:34:32,960 --> 00:34:35,040
Speaker 4: Again, I'm on the side of bondholders in that one.

739
00:34:35,160 --> 00:34:37,239
Speaker 4: Just look at the market caps, look at the bond yields, and.

740
00:34:38,880 --> 00:34:41,399
Speaker 3: Explain what you mean by that. For so, like people

741
00:34:41,400 --> 00:34:44,080
Speaker 3: who don't have a Bloomberg, so we've goes people like

742
00:34:44,120 --> 00:34:46,120
Speaker 3: me who have a Bloomberg but are too lazy.

743
00:34:46,960 --> 00:34:49,439
Speaker 4: So these names are going out and yeah in either

744
00:34:49,480 --> 00:34:52,840
Speaker 4: open Ai or Navidia is going out and buying equity

745
00:34:52,880 --> 00:34:55,759
Speaker 4: in these neocloud companies. So then they can go out

746
00:34:55,800 --> 00:34:58,080
Speaker 4: and either supply the compute to open Ai and buy

747
00:34:58,080 --> 00:35:00,160
Speaker 4: the chips from Navidia. So it's all very circular. And

748
00:35:00,200 --> 00:35:02,680
Speaker 4: I think the example people used twenty years ago is

749
00:35:02,760 --> 00:35:06,160
Speaker 4: Nortel was doing this called the vendor financing. So there's

750
00:35:06,200 --> 00:35:08,040
Speaker 4: a little lot of skepticism on that.

751
00:35:08,360 --> 00:35:11,160
Speaker 2: Okay, so we can't do a utilities episode. I know

752
00:35:11,160 --> 00:35:13,239
Speaker 2: we've been focused on data centers, but we can't do

753
00:35:13,280 --> 00:35:17,799
Speaker 2: a utilities episode without mentioning nuclear power. What's it going

754
00:35:17,840 --> 00:35:22,920
Speaker 2: to take to actually get some capacity from nuclear sure.

755
00:35:23,320 --> 00:35:26,560
Speaker 4: So obviously the Vogel plant was the last big nuclear

756
00:35:26,600 --> 00:35:28,640
Speaker 4: plant in the line. Was supposed to cost fourteen billion.

757
00:35:28,640 --> 00:35:30,840
Speaker 4: It ended up costing thirty two billion. It came online

758
00:35:30,840 --> 00:35:33,120
Speaker 4: ten years late. No utility wants to take that risk.

759
00:35:33,520 --> 00:35:36,640
Speaker 4: Now everyone's talking about these small modular reactors, and I

760
00:35:36,640 --> 00:35:38,520
Speaker 4: think that's what you're going to start seeing is more

761
00:35:38,560 --> 00:35:40,959
Speaker 4: talk of these. The only way we think a small

762
00:35:40,960 --> 00:35:44,800
Speaker 4: module actor goes final investment decision FID is if big

763
00:35:44,840 --> 00:35:47,759
Speaker 4: tech agrees to do two things. They agree to buy

764
00:35:47,840 --> 00:35:52,240
Speaker 4: some SMRs and they invest equity in those SMR manufacturers

765
00:35:52,360 --> 00:35:54,080
Speaker 4: to give them the capex to build.

766
00:35:53,840 --> 00:35:55,800
Speaker 3: Which sounds like something they would do to be honest

767
00:35:56,200 --> 00:35:56,680
Speaker 3: for sure.

768
00:35:57,080 --> 00:35:58,839
Speaker 4: And I think that's the only way you get one

769
00:35:58,880 --> 00:36:00,920
Speaker 4: of these off the ground. And I think if those

770
00:36:00,960 --> 00:36:03,719
Speaker 4: stocks rally on that deal, they're all shorts because they're

771
00:36:03,719 --> 00:36:06,200
Speaker 4: already reflecting several of those deals happening. So the big

772
00:36:06,200 --> 00:36:10,040
Speaker 4: ones are our new Scale and Oaklow and Sam Outman

773
00:36:10,120 --> 00:36:12,480
Speaker 4: of open AI used to be the chairman of Oaklow

774
00:36:12,520 --> 00:36:14,160
Speaker 4: and then he stepped down so they could do a deal.

775
00:36:14,520 --> 00:36:18,480
Speaker 4: So something along those lines would happen. That being said,

776
00:36:18,640 --> 00:36:21,759
Speaker 4: Donald Trump has talked about doing work with Westinghouse and

777
00:36:21,800 --> 00:36:24,680
Speaker 4: taking equity ownership to build another AP one thousand and

778
00:36:24,719 --> 00:36:28,000
Speaker 4: Obviously President Trump is all about taking equity, but none

779
00:36:28,040 --> 00:36:30,120
Speaker 4: of the utilities in my coverage are going to build

780
00:36:30,160 --> 00:36:31,560
Speaker 4: something without some sort of backstock.

781
00:36:31,680 --> 00:36:35,920
Speaker 3: We did an episode recently with an infrastructure investor and I,

782
00:36:36,239 --> 00:36:38,520
Speaker 3: you know, I'm a journalist. I look at the past.

783
00:36:38,560 --> 00:36:40,640
Speaker 3: I don't talk about the future. But I was put

784
00:36:40,680 --> 00:36:42,640
Speaker 3: on the spot and I said, I think off the

785
00:36:42,760 --> 00:36:45,879
Speaker 3: next twenty years. Gun to my head, we will never

786
00:36:45,960 --> 00:36:49,680
Speaker 3: have another local We're not going to have another project

787
00:36:49,719 --> 00:36:52,800
Speaker 3: like that in America. And sounds like you agree.

788
00:36:53,160 --> 00:36:56,240
Speaker 4: I agree. I do think you see some SMRs. I mean, frankly,

789
00:36:56,280 --> 00:36:58,959
Speaker 4: our country has been making nuclear submarines for sixty seven years.

790
00:36:58,960 --> 00:37:02,120
Speaker 4: That's an SMR there. So I think that's the way

791
00:37:02,120 --> 00:37:04,040
Speaker 4: it happens, is big tech goes in and does that.

792
00:37:04,360 --> 00:37:06,840
Speaker 2: For sure, in twenty years, we'll have you back on

793
00:37:06,960 --> 00:37:08,680
Speaker 2: to see whether or not both of you.

794
00:37:08,880 --> 00:37:09,600
Speaker 4: This is my only writer.

795
00:37:09,719 --> 00:37:11,640
Speaker 3: Wrong, I don't nothing about the future. This is my

796
00:37:11,719 --> 00:37:14,080
Speaker 3: only one call. And I just don't think we're going

797
00:37:14,160 --> 00:37:16,000
Speaker 3: to ever get them right. That's we're not going to

798
00:37:16,000 --> 00:37:16,959
Speaker 3: get a bunch of those times.

799
00:37:17,600 --> 00:37:19,480
Speaker 4: I'm with you on that, and in twenty years hopefully

800
00:37:19,520 --> 00:37:20,560
Speaker 4: I can dial in from the beach.

801
00:37:20,719 --> 00:37:23,160
Speaker 2: All right, we see you in twenty years then, no,

802
00:37:23,320 --> 00:37:26,279
Speaker 2: probably before because this was a fantastic, this is great conversation.

803
00:37:26,640 --> 00:37:28,560
Speaker 4: Thank you so much for coming on off of Thanks

804
00:37:28,600 --> 00:37:29,080
Speaker 4: for having.

805
00:37:28,960 --> 00:37:43,200
Speaker 2: Me, Joe. That was a really fun conversation.

806
00:37:43,280 --> 00:37:44,400
Speaker 3: That's super fun. I love that.

807
00:37:44,760 --> 00:37:49,319
Speaker 2: My the opinion or framing that I'm sort of coalescing

808
00:37:49,360 --> 00:37:55,120
Speaker 2: around is that AI can be simultaneously under hyped and overvalued,

809
00:37:55,320 --> 00:37:58,160
Speaker 2: right and actually throughout history, that's kind of what we've

810
00:37:58,160 --> 00:38:02,400
Speaker 2: seen with transformative technology, right, Like think about the Internet bubble.

811
00:38:02,640 --> 00:38:05,240
Speaker 2: The Internet changed the world, but it was a bubble.

812
00:38:05,400 --> 00:38:08,880
Speaker 2: Think about railroads. Railroads in the eighteen hundreds changed the world,

813
00:38:09,040 --> 00:38:11,920
Speaker 2: but also a bubble. So I think that's kind of

814
00:38:12,239 --> 00:38:14,479
Speaker 2: it's kind of what I think. The key issue, which

815
00:38:14,520 --> 00:38:17,520
Speaker 2: Andy and you both touched on is the timing, right.

816
00:38:17,680 --> 00:38:20,799
Speaker 3: The timing, And yeah, I mean I think it's it's

817
00:38:20,960 --> 00:38:25,400
Speaker 3: very interesting because of course his argument doesn't even you know,

818
00:38:25,920 --> 00:38:29,319
Speaker 3: doesn't even rest on any valuations right now. It's like

819
00:38:29,600 --> 00:38:32,759
Speaker 3: there's all of this expectation for build out there is

820
00:38:32,920 --> 00:38:34,680
Speaker 3: you know, as he put it, there is a number

821
00:38:34,760 --> 00:38:37,840
Speaker 3: for the amount of the volume of data centered demand.

822
00:38:38,520 --> 00:38:39,000
Speaker 4: There is an.

823
00:38:38,960 --> 00:38:41,920
Speaker 3: Amount that's being built up, and he's like, the second

824
00:38:42,000 --> 00:38:45,680
Speaker 3: number looks bigger, and that's going to be a that's

825
00:38:45,719 --> 00:38:48,359
Speaker 3: going to be a problem. Yeah, and the time to

826
00:38:48,400 --> 00:38:51,840
Speaker 3: your point, like I thought, you know, really interesting observation

827
00:38:51,880 --> 00:38:54,600
Speaker 3: he had was a little bit not tangential to his

828
00:38:54,680 --> 00:38:57,680
Speaker 3: core idea, but this idea that some of our energy

829
00:38:57,680 --> 00:39:00,960
Speaker 3: policies are encouraging a lot of production right now, particularly

830
00:39:01,000 --> 00:39:04,040
Speaker 3: the expiring solar credits. At the same time, a lot

831
00:39:04,040 --> 00:39:07,000
Speaker 3: of this demand is going to come online in the

832
00:39:07,040 --> 00:39:10,360
Speaker 3: back end, et cetera. I do think, you know, it

833
00:39:10,400 --> 00:39:13,160
Speaker 3: seems like a really definitely seems like a fun space.

834
00:39:13,280 --> 00:39:16,640
Speaker 3: It's very far from when we were just talking about

835
00:39:16,719 --> 00:39:19,960
Speaker 3: like utilities as raat proxy. So I think for old

836
00:39:19,960 --> 00:39:21,440
Speaker 3: people to get income.

837
00:39:21,680 --> 00:39:23,880
Speaker 2: Well, the other thing I was thinking about on the

838
00:39:23,880 --> 00:39:28,879
Speaker 2: demand side is I think there's a tendency among AI

839
00:39:29,080 --> 00:39:32,200
Speaker 2: bulls vibecoders such as yourself that's right to think that

840
00:39:32,440 --> 00:39:34,719
Speaker 2: demand is just going to go one way, right, So

841
00:39:35,239 --> 00:39:38,360
Speaker 2: there's going to be more demand for AI because I

842
00:39:38,400 --> 00:39:40,680
Speaker 2: don't know every piece of software is going to be

843
00:39:40,800 --> 00:39:45,320
Speaker 2: replicated through clod code or whatever, and so power demand

844
00:39:45,360 --> 00:39:48,080
Speaker 2: is going to go up as well. But what we've

845
00:39:48,120 --> 00:39:51,160
Speaker 2: seen so far is that these things are getting more

846
00:39:51,200 --> 00:39:51,920
Speaker 2: and more efficient.

847
00:39:51,960 --> 00:39:54,360
Speaker 3: They're definitely getting right, more and more efficient.

848
00:39:54,239 --> 00:39:56,399
Speaker 2: Like faster than anyone expected.

849
00:39:57,200 --> 00:40:00,520
Speaker 3: Yeah, you know, this is a little bit tangent to

850
00:40:00,560 --> 00:40:02,759
Speaker 3: the point, but I do think like one of the

851
00:40:02,800 --> 00:40:07,319
Speaker 3: recurring phenomenons that we're seeing across this industry is that

852
00:40:08,080 --> 00:40:11,560
Speaker 3: every I mean and this is I guess it's a

853
00:40:11,560 --> 00:40:15,440
Speaker 3: bowl case, which is that you know, even the optimist

854
00:40:15,520 --> 00:40:19,040
Speaker 3: keep getting turned out to be too pessimistic. The pace

855
00:40:19,120 --> 00:40:23,280
Speaker 3: of say like efficiency gains for the cost of processing

856
00:40:23,280 --> 00:40:27,560
Speaker 3: a token dropping faster than people expected this morning, recording

857
00:40:27,560 --> 00:40:31,360
Speaker 3: to this January twenty eighth ASML the big chip equipment company,

858
00:40:31,640 --> 00:40:35,640
Speaker 3: way better than expected. The emails, the benchmarks for the

859
00:40:35,680 --> 00:40:38,520
Speaker 3: models where it's like the optimists say, like maybe it

860
00:40:38,560 --> 00:40:41,280
Speaker 3: could code at this level by twenty twenty seven, turns

861
00:40:41,320 --> 00:40:43,760
Speaker 3: out it hits there by like you know, early twenty

862
00:40:43,840 --> 00:40:46,320
Speaker 3: twenty six, etc. So like if you want to just

863
00:40:46,400 --> 00:40:48,600
Speaker 3: make I'm not making any case here, but if you

864
00:40:48,680 --> 00:40:50,920
Speaker 3: just want to make a bowl case, it's like even

865
00:40:50,960 --> 00:40:54,719
Speaker 3: the optimists keep getting surprised to the upside. On the

866
00:40:54,760 --> 00:40:58,880
Speaker 3: other hand, it's fascinating to hear him say, look at

867
00:40:58,880 --> 00:41:02,000
Speaker 3: what the markets are saying. They're not pricing in any

868
00:41:02,040 --> 00:41:05,080
Speaker 3: of these expectations. And I was particularly surprised because I

869
00:41:05,080 --> 00:41:08,399
Speaker 3: didn't realize this that even like you know, for all

870
00:41:08,400 --> 00:41:11,120
Speaker 3: of the talk of LERG export terminals, et cetera, that

871
00:41:11,719 --> 00:41:13,879
Speaker 3: gas is expected to be cheaper a few years. Yeah,

872
00:41:13,960 --> 00:41:18,160
Speaker 3: no than it is right now. Very interesting dissonance between

873
00:41:18,200 --> 00:41:19,520
Speaker 3: that and the popular narrative.

874
00:41:19,880 --> 00:41:22,680
Speaker 2: Maybe gas traders just aren't vibe coders yet.

875
00:41:22,920 --> 00:41:26,200
Speaker 3: Then they would understand exactly how much more of this

876
00:41:26,400 --> 00:41:27,680
Speaker 3: UH plan going.

877
00:41:27,600 --> 00:41:29,719
Speaker 2: To Just from an energy perspective, though, there is a

878
00:41:29,719 --> 00:41:32,080
Speaker 2: push and pull factor here, right, So on the one hand,

879
00:41:32,120 --> 00:41:34,600
Speaker 2: everyone could use AI and demand goes up, But on

880
00:41:34,640 --> 00:41:37,480
Speaker 2: the other hand, maybe it gets super super efficient and

881
00:41:37,480 --> 00:41:41,879
Speaker 2: then demand goes down. I think that's that's the difficulty, or.

882
00:41:41,800 --> 00:41:44,600
Speaker 3: It's just there is a number that's out there. It's

883
00:41:44,640 --> 00:41:48,160
Speaker 3: like some sort of like some reasonable inferences about where

884
00:41:48,200 --> 00:41:50,759
Speaker 3: it's going to go, and it's very high. And it

885
00:41:50,840 --> 00:41:53,480
Speaker 3: was actually very striking listening to him talk about some

886
00:41:53,520 --> 00:41:58,879
Speaker 3: of those super the hyperscaler numbers, because it's like where

887
00:41:58,920 --> 00:42:01,280
Speaker 3: is it at up right? As he pointed out, Okay,

888
00:42:01,400 --> 00:42:04,280
Speaker 3: chet GPT like came out two gig loots, et cetera,

889
00:42:05,080 --> 00:42:08,640
Speaker 3: Like there are a ton of chet GPTs out there, right,

890
00:42:08,719 --> 00:42:12,160
Speaker 3: and that's one of the most computationally intensive things. Like

891
00:42:12,480 --> 00:42:14,640
Speaker 3: maybe there are reasons to think this is all going

892
00:42:14,680 --> 00:42:16,359
Speaker 3: to go great, there's gonna be a ton of money

893
00:42:16,400 --> 00:42:19,359
Speaker 3: made in AI, but you can't really just like get

894
00:42:19,400 --> 00:42:21,879
Speaker 3: to the number. I don't know, I think he it

895
00:42:21,920 --> 00:42:24,719
Speaker 3: was this was a very useful perspective, just sort of

896
00:42:24,760 --> 00:42:27,320
Speaker 3: on some of this simple math, and the math sounds

897
00:42:27,360 --> 00:42:29,399
Speaker 3: like a subtraction, like this sounds like what the sun

898
00:42:29,480 --> 00:42:30,000
Speaker 3: is learning about.

899
00:42:30,040 --> 00:42:30,200
Speaker 4: Well.

900
00:42:30,400 --> 00:42:32,319
Speaker 2: The other thing I thought was really interesting was the

901
00:42:32,360 --> 00:42:36,440
Speaker 2: response to your question about why would you finance these

902
00:42:36,440 --> 00:42:39,320
Speaker 2: things off balance sheet if you're you know, this massive

903
00:42:39,840 --> 00:42:44,240
Speaker 2: cash rich technology giant, and the suggestion there was, well,

904
00:42:44,280 --> 00:42:47,880
Speaker 2: maybe at some point in the future, like five years

905
00:42:47,920 --> 00:42:50,640
Speaker 2: down the line, you need to get rid of this liability.

906
00:42:50,680 --> 00:42:51,680
Speaker 2: You don't want to deal with it.

907
00:42:51,840 --> 00:42:55,160
Speaker 3: This is why we did that episode with the guy

908
00:42:55,239 --> 00:42:58,080
Speaker 3: who has you know, the company doing the legal docs right,

909
00:42:58,080 --> 00:43:02,000
Speaker 3: et cetera. This is why it's pretty crucial to understand

910
00:43:02,160 --> 00:43:05,520
Speaker 3: some of these things, because some of the questions sound

911
00:43:05,800 --> 00:43:09,520
Speaker 3: like Facebook's or Meta's option to walk away. Right, there's

912
00:43:09,520 --> 00:43:14,920
Speaker 3: some call option implicitly to walk away, et cetera. And

913
00:43:13,760 --> 00:43:18,560
Speaker 3: they how they what they could do, what scenarios, and

914
00:43:18,600 --> 00:43:20,520
Speaker 3: what they would allow it to be do is obviously

915
00:43:20,880 --> 00:43:23,840
Speaker 3: going to be pretty crucial for any investors in this

916
00:43:23,880 --> 00:43:26,400
Speaker 3: off balance sheet paper. I did not know until you

917
00:43:26,480 --> 00:43:30,520
Speaker 3: asked this whole idea that the ratings agency is, well,

918
00:43:30,560 --> 00:43:32,640
Speaker 3: they don't look at his debt, they do back out

919
00:43:32,680 --> 00:43:36,440
Speaker 3: a lease cost, and therefore it can inform their overall

920
00:43:37,000 --> 00:43:38,120
Speaker 3: credit sustainability.

921
00:43:38,360 --> 00:43:41,640
Speaker 2: Well, the other thing, you know, we touched on this,

922
00:43:41,760 --> 00:43:45,680
Speaker 2: but there's more and more demand from investors for data

923
00:43:45,719 --> 00:43:48,279
Speaker 2: center debt, right, like the space is getting more for

924
00:43:48,320 --> 00:43:51,000
Speaker 2: some reason, the space is getting more competitive, and so

925
00:43:51,360 --> 00:43:54,800
Speaker 2: naturally what you see in any other credit cycle throughout

926
00:43:54,840 --> 00:43:58,360
Speaker 2: history is as demand pros and people are competing for deals,

927
00:43:58,960 --> 00:44:02,240
Speaker 2: the documentation and the protections tend to be weird.

928
00:44:02,400 --> 00:44:02,600
Speaker 1: You know.

929
00:44:03,120 --> 00:44:06,200
Speaker 3: I find the existence of hype cycles for debt to

930
00:44:06,239 --> 00:44:09,040
Speaker 3: be a little bit weird because I get like, oh,

931
00:44:09,080 --> 00:44:11,600
Speaker 3: I really want to get into AI equity, right, because

932
00:44:11,600 --> 00:44:14,399
Speaker 3: that could one hundred x next year, right, And it's

933
00:44:14,440 --> 00:44:17,120
Speaker 3: like I'm really excited about getting into data centered debt

934
00:44:17,280 --> 00:44:20,360
Speaker 3: because it might maaybe fifty BIPs more I find to

935
00:44:20,400 --> 00:44:22,520
Speaker 3: be very strange, or one hundred BIPs more. It's like,

936
00:44:22,800 --> 00:44:25,879
Speaker 3: if I haven't fixed look, I'm a simple, simple guy.

937
00:44:25,880 --> 00:44:28,520
Speaker 3: But if I have a fixed income allocation, all I

938
00:44:28,640 --> 00:44:32,600
Speaker 3: care about is minimizing downside and I don't really care,

939
00:44:32,640 --> 00:44:35,080
Speaker 3: like what sector it is. I'm not participating in the upside.

940
00:44:35,239 --> 00:44:36,719
Speaker 2: You're not going to get greedy, You're not going to

941
00:44:36,800 --> 00:44:37,160
Speaker 2: get rich.

942
00:44:38,160 --> 00:44:40,239
Speaker 3: I gotta get super rich. I'm like, I just don't

943
00:44:40,239 --> 00:44:41,920
Speaker 3: want to lose my like for fixing COF. I just

944
00:44:41,920 --> 00:44:43,440
Speaker 3: don't want to lose my money.

945
00:44:43,480 --> 00:44:45,319
Speaker 2: Fair enough, all right, shall we leave it there?

946
00:44:45,360 --> 00:44:46,040
Speaker 3: Let's save it there.

947
00:44:46,080 --> 00:44:48,520
Speaker 2: This has been another episode of the All Thoughts podcast.

948
00:44:48,640 --> 00:44:51,360
Speaker 2: I'm Tracy Alloway. You can follow me at Tracy Alloway

949
00:44:51,520 --> 00:44:52,759
Speaker 2: and I'm Joe Wisenthal.

950
00:44:52,840 --> 00:44:55,840
Speaker 3: You can follow me at The Stalwart. Follow our producers

951
00:44:55,880 --> 00:44:59,319
Speaker 3: Carmen Rodriguez at Carmen armandsh Oll Bennett at dashbod and

952
00:44:59,400 --> 00:45:02,360
Speaker 3: kill Brooks at Kelbrooks Now. For more odd Laws content,

953
00:45:02,400 --> 00:45:04,719
Speaker 3: go to Bloomberg dot com slash odd Lots for the

954
00:45:04,800 --> 00:45:07,319
Speaker 3: daily newsletter and all of our episodes and you can

955
00:45:07,400 --> 00:45:09,640
Speaker 3: chat about all of these topics twenty four seven in

956
00:45:09,800 --> 00:45:12,959
Speaker 3: our discord discord dot gg, slash.

957
00:45:12,640 --> 00:45:15,200
Speaker 2: Od lots And if you enjoy all lots, if you

958
00:45:15,440 --> 00:45:17,719
Speaker 2: like it when we talk about the data center build out,

959
00:45:17,719 --> 00:45:20,280
Speaker 2: then please leave us a positive review on your favorite

960
00:45:20,280 --> 00:45:23,880
Speaker 2: podcast platform. And remember, if you are a Bloomberg subscriber,

961
00:45:23,960 --> 00:45:27,080
Speaker 2: you can listen to all of our episodes absolutely ad free.

962
00:45:27,520 --> 00:45:30,080
Speaker 2: All you need to do is find the Bloomberg channel

963
00:45:30,080 --> 00:45:33,800
Speaker 2: on Apple Podcasts and follow the instructions there. Thanks for listening.