1
00:00:00,520 --> 00:00:02,920
Speaker 1: This is Tom Rowland's Reese and you're listening to Switched

2
00:00:02,960 --> 00:00:05,720
Speaker 1: on the B and EF podcast. Copper may be one

3
00:00:05,720 --> 00:00:09,800
Speaker 1: of our most underappreciated metals. Essential for transport, construction and

4
00:00:09,840 --> 00:00:12,559
Speaker 1: the manufacturing of consumer goods. Is also critical for the

5
00:00:12,600 --> 00:00:15,120
Speaker 1: parer and electric vehicle sectors that make up the cornerstone

6
00:00:15,120 --> 00:00:17,759
Speaker 1: of the energy transition. The supply and demand balance of

7
00:00:17,760 --> 00:00:21,160
Speaker 1: the mineral is already worryingly tight, but with increasing demand growth,

8
00:00:21,160 --> 00:00:23,160
Speaker 1: we could see a shortage of copper suppli as soon

9
00:00:23,160 --> 00:00:25,640
Speaker 1: as twenty twenty eight and the six million ton deficit

10
00:00:25,680 --> 00:00:29,320
Speaker 1: by twenty thirty five. Meanwhile, data center rollout is gathering

11
00:00:29,320 --> 00:00:32,200
Speaker 1: pace across the planet, and those huge facilities all require

12
00:00:32,520 --> 00:00:35,240
Speaker 1: you guessed it, lots and lots of copper. With so

13
00:00:35,360 --> 00:00:38,120
Speaker 1: much demand, you'd logically expect that miners would be falling

14
00:00:38,159 --> 00:00:40,199
Speaker 1: over themselves to explore for the mineral and open up

15
00:00:40,200 --> 00:00:43,400
Speaker 1: new mining facilities, but actually the reverse has been happening.

16
00:00:43,640 --> 00:00:46,480
Speaker 1: Economic concerns mean that companies are leaning into keeping old

17
00:00:46,520 --> 00:00:49,000
Speaker 1: minds up and running rather than starting new ones, and

18
00:00:49,080 --> 00:00:52,199
Speaker 1: in some cases minds are even being closed. So just

19
00:00:52,200 --> 00:00:53,960
Speaker 1: what is it about this metal that runs counter to

20
00:00:54,000 --> 00:00:56,800
Speaker 1: common logic and what is its near term demand outlook.

21
00:00:57,080 --> 00:01:00,000
Speaker 1: Today I'm joined by Rosemary Cats, a senior associate from

22
00:01:00,040 --> 00:01:03,040
Speaker 1: BNF's Metals and Minings team. We discussed the findings from

23
00:01:03,040 --> 00:01:05,920
Speaker 1: her note tracking copper and data centers twenty twenty five

24
00:01:05,959 --> 00:01:09,319
Speaker 1: to twenty thirty five supply crunch, which BNF clients can

25
00:01:09,360 --> 00:01:12,120
Speaker 1: find at BNF go on the Bloomberg terminal or on

26
00:01:12,200 --> 00:01:14,720
Speaker 1: BNF dot com. All right, let's get to talking about

27
00:01:14,720 --> 00:01:17,720
Speaker 1: copper supply constraints and the growing demand from data centers.

28
00:01:17,400 --> 00:01:28,399
Speaker 2: With Rose Rose Welcome to the podcast.

29
00:01:28,800 --> 00:01:29,959
Speaker 3: Thank you Tom for having me.

30
00:01:30,560 --> 00:01:34,000
Speaker 1: So copper is what we're talking about today, and we

31
00:01:34,080 --> 00:01:36,399
Speaker 1: have our Metals and Mining team, which you're part of,

32
00:01:36,400 --> 00:01:38,640
Speaker 1: of course, and a lot of the time you're talking

33
00:01:38,720 --> 00:01:43,440
Speaker 1: about sort of metals like lithium and cobble less everyday metals.

34
00:01:43,480 --> 00:01:45,959
Speaker 1: You know, everyone is familiar with copper. In some senses,

35
00:01:46,000 --> 00:01:48,880
Speaker 1: it seems a little bit unsexy because there's nothing new

36
00:01:48,920 --> 00:01:51,800
Speaker 1: about it. So why are we talking so much about copper?

37
00:01:51,840 --> 00:01:52,680
Speaker 1: Why does it matter?

38
00:01:53,080 --> 00:01:57,040
Speaker 3: Well, Tom, copper is becoming sexier again, and mainly because

39
00:01:57,040 --> 00:02:00,920
Speaker 3: of the rapid buildout of data center infrastructure across the world,

40
00:02:01,080 --> 00:02:04,520
Speaker 3: and also the other sexier part, as you put it,

41
00:02:04,600 --> 00:02:07,560
Speaker 3: is the fact that there's not enough material that is

42
00:02:07,640 --> 00:02:11,320
Speaker 3: being mined and refined to actually support all this copper

43
00:02:11,320 --> 00:02:15,919
Speaker 3: infrastructure from your traditional sectors like manufacturing, to the energy

44
00:02:15,919 --> 00:02:19,600
Speaker 3: transition itself, and now for data centers. That's why it's

45
00:02:19,639 --> 00:02:21,280
Speaker 3: all a buzz at this point in time.

46
00:02:21,680 --> 00:02:24,600
Speaker 1: So that's kind of a new buzz around it because

47
00:02:24,639 --> 00:02:27,760
Speaker 1: of data centers can just talk about the other ways

48
00:02:27,760 --> 00:02:30,519
Speaker 1: in which copper is important, because I imagine that if maybe

49
00:02:30,600 --> 00:02:33,920
Speaker 1: a shortage of copper supply because of this data center demand,

50
00:02:33,960 --> 00:02:36,680
Speaker 1: there might be other areas that are impacted by that.

51
00:02:37,000 --> 00:02:39,760
Speaker 3: Yeah, so the story is just not about data centers. So,

52
00:02:39,840 --> 00:02:43,200
Speaker 3: as I said, demand from an absolute point of view

53
00:02:43,360 --> 00:02:46,560
Speaker 3: is basically it comes from your traditional sectors, right the

54
00:02:46,680 --> 00:02:52,120
Speaker 3: largest sector of that demand, which includes construction, manufacturing, electronics,

55
00:02:52,280 --> 00:02:56,360
Speaker 3: internal combustion vehicles. They already take up a lot of

56
00:02:56,360 --> 00:02:59,760
Speaker 3: that copper demand and require a lot of supply in turn.

57
00:03:00,040 --> 00:03:03,480
Speaker 3: And then we have as the energy transition that is

58
00:03:03,520 --> 00:03:08,600
Speaker 3: electrification through electric vehicles, through battery storage, through the buildout

59
00:03:08,639 --> 00:03:12,000
Speaker 3: of wind and solar. They also require copper, and the

60
00:03:12,200 --> 00:03:15,160
Speaker 3: energy transition is actually one of the fastest growing markets

61
00:03:15,240 --> 00:03:18,200
Speaker 3: right now. Data centers, in essence, are competing with these

62
00:03:18,240 --> 00:03:21,880
Speaker 3: two massive industries for this limited supply at the end

63
00:03:21,880 --> 00:03:22,200
Speaker 3: of the day.

64
00:03:22,440 --> 00:03:23,520
Speaker 2: See, it's amazing.

65
00:03:23,639 --> 00:03:26,840
Speaker 1: It sounds like copper is both an important part of

66
00:03:26,880 --> 00:03:29,960
Speaker 1: the past and an important part of the future, both

67
00:03:29,960 --> 00:03:34,239
Speaker 1: in terms of the energy transition and areas like artificial intelligence.

68
00:03:34,400 --> 00:03:37,880
Speaker 1: It almost feels like we've been taking copper for granted

69
00:03:38,000 --> 00:03:41,120
Speaker 1: for years, and now we're beginning to find out that

70
00:03:41,120 --> 00:03:43,760
Speaker 1: that was not the right way to think about copper.

71
00:03:44,040 --> 00:03:46,560
Speaker 3: That's actually quite true, and it's not that, you know,

72
00:03:46,880 --> 00:03:49,280
Speaker 3: we've been taking it for granted, more like it's not

73
00:03:49,440 --> 00:03:53,000
Speaker 3: been kept in the limelight so that we stay abreast

74
00:03:53,080 --> 00:03:56,280
Speaker 3: of this risk that's ever looming. If we look back

75
00:03:56,320 --> 00:04:00,560
Speaker 3: in history, big copper producers have you know, published notes

76
00:04:00,720 --> 00:04:04,800
Speaker 3: of work to say, hey, we're struggling globally and prices

77
00:04:04,880 --> 00:04:06,320
Speaker 3: at the end of the day, I have signaled that

78
00:04:06,680 --> 00:04:09,360
Speaker 3: quite obviously. I mean, copper prices on the first of

79
00:04:09,400 --> 00:04:13,000
Speaker 3: August was at nine six hundred dollars per ton. It's

80
00:04:13,040 --> 00:04:17,160
Speaker 3: gone up to about nine nine hundred recently. So that's

81
00:04:17,240 --> 00:04:20,200
Speaker 3: a clear signal that there's a problem here, and that

82
00:04:20,320 --> 00:04:22,919
Speaker 3: can be attributed to the lack of supply that's coming

83
00:04:22,920 --> 00:04:23,640
Speaker 3: through the system.

84
00:04:24,320 --> 00:04:26,120
Speaker 2: So let's talk about the supply.

85
00:04:26,560 --> 00:04:29,400
Speaker 1: Where is all of this copper that we don't have

86
00:04:29,520 --> 00:04:31,479
Speaker 1: enough of, where is it coming from?

87
00:04:31,560 --> 00:04:36,640
Speaker 3: So supply currently mainly comes from South America, Chile, Peru,

88
00:04:36,920 --> 00:04:40,880
Speaker 3: and then the US and Australia also produce copper. Similarly,

89
00:04:40,880 --> 00:04:44,360
Speaker 3: in Africa, we have the DRC in Zambia also producing

90
00:04:44,440 --> 00:04:48,600
Speaker 3: copper or or copper concentrate, and then refining the material

91
00:04:48,640 --> 00:04:51,440
Speaker 3: into a useful form is mostly done in China. So

92
00:04:51,560 --> 00:04:56,160
Speaker 3: China actually is responsible for about forty percent of refined copper,

93
00:04:56,360 --> 00:04:59,320
Speaker 3: and then the rest comes from South America, the US,

94
00:04:59,520 --> 00:05:01,200
Speaker 3: Australia and the DRC.

95
00:05:01,720 --> 00:05:04,760
Speaker 1: So let me just make sure I've understood you correctly

96
00:05:04,800 --> 00:05:08,320
Speaker 1: here it sounds like a lot of the copper refining

97
00:05:08,400 --> 00:05:13,120
Speaker 1: is happening in China and then everywhere else it's happening where.

98
00:05:12,880 --> 00:05:15,160
Speaker 2: The minerals are being pulled out of the ground. Roughly.

99
00:05:15,440 --> 00:05:18,360
Speaker 1: Yeah, and you might have already said this is China

100
00:05:18,440 --> 00:05:20,600
Speaker 1: got its own source of copper minerals or is that

101
00:05:20,640 --> 00:05:21,920
Speaker 1: something that is lacking.

102
00:05:22,240 --> 00:05:25,400
Speaker 3: China's got its own source, but it does import those

103
00:05:25,440 --> 00:05:28,279
Speaker 3: copper concentrates from these regions for refining.

104
00:05:28,640 --> 00:05:31,320
Speaker 1: So I suppose what's interesting is when I think about it,

105
00:05:31,360 --> 00:05:34,760
Speaker 1: the countries or regions that are missing from that list.

106
00:05:34,920 --> 00:05:38,280
Speaker 1: You know, the United States has big plans around data centers.

107
00:05:38,600 --> 00:05:41,760
Speaker 1: They are neither a producer or a refiner of copper

108
00:05:42,160 --> 00:05:43,000
Speaker 1: in a big way.

109
00:05:43,279 --> 00:05:46,599
Speaker 3: No, So they are actually a producer and refiner. They

110
00:05:46,680 --> 00:05:50,159
Speaker 3: just don't have sufficient domestic supplies to keep up with

111
00:05:50,279 --> 00:05:52,720
Speaker 3: the domestic demand at this point in time.

112
00:05:53,000 --> 00:05:53,320
Speaker 2: Got it?

113
00:05:53,600 --> 00:05:55,800
Speaker 1: So in your research on this, I think one of

114
00:05:55,800 --> 00:05:58,680
Speaker 1: the things that came up is that copper minds have

115
00:05:58,839 --> 00:06:03,239
Speaker 1: been shutting down, which naively doesn't make any sense given

116
00:06:03,279 --> 00:06:03,920
Speaker 1: what we're saying.

117
00:06:04,040 --> 00:06:05,840
Speaker 2: So what's the story with that?

118
00:06:06,400 --> 00:06:09,840
Speaker 3: So it's a two pronged story. So we have one

119
00:06:09,960 --> 00:06:13,440
Speaker 3: issue from an economic standpoint. Currently, a lot of like

120
00:06:13,520 --> 00:06:17,040
Speaker 3: the good copper, high grade oxide copper has been mined

121
00:06:17,040 --> 00:06:19,800
Speaker 3: out quite extensively and we're left with the more lower

122
00:06:19,839 --> 00:06:20,919
Speaker 3: grade material.

123
00:06:21,120 --> 00:06:23,280
Speaker 1: This is a new concept to me, good copper and

124
00:06:23,320 --> 00:06:25,719
Speaker 1: bad copper. Can you just talk through what good copper

125
00:06:25,760 --> 00:06:26,640
Speaker 1: and bad copper is?

126
00:06:27,480 --> 00:06:31,360
Speaker 3: So good copper is basically a material that comes from

127
00:06:31,400 --> 00:06:35,119
Speaker 3: what we call oxide ore or oxide that's mined out,

128
00:06:35,400 --> 00:06:39,320
Speaker 3: and those are higher grade material, whereas bad copper are

129
00:06:39,360 --> 00:06:42,520
Speaker 3: your lower grade materials. So high grade stuff is easy

130
00:06:42,560 --> 00:06:45,880
Speaker 3: to process, to refine, to move through the system, whereas

131
00:06:45,920 --> 00:06:48,800
Speaker 3: your lower grade material it's harder to mine. One you

132
00:06:48,880 --> 00:06:51,279
Speaker 3: have to remove more waste to get to the ore,

133
00:06:51,440 --> 00:06:53,760
Speaker 3: and then it's harder to refine because you have all

134
00:06:53,800 --> 00:06:57,560
Speaker 3: this what they call deleterious material or gang waste that

135
00:06:57,640 --> 00:06:59,560
Speaker 3: you need to remove so it can go through the

136
00:06:59,600 --> 00:07:04,160
Speaker 3: same finding process, and that makes for higher operational costs.

137
00:07:04,200 --> 00:07:07,000
Speaker 3: So most of what we're finding out is the bad copper.

138
00:07:07,320 --> 00:07:09,720
Speaker 1: Just to make sure I've understood this correctly, the end

139
00:07:09,760 --> 00:07:12,880
Speaker 1: product is the same in terms of quality. It's just

140
00:07:13,360 --> 00:07:15,200
Speaker 1: I suppose you could say good copper is the low

141
00:07:15,240 --> 00:07:17,680
Speaker 1: hanging fruit and bad copper is a high hanging fruit,

142
00:07:17,720 --> 00:07:20,400
Speaker 1: and most of the low hanging fruit has already been picked.

143
00:07:20,520 --> 00:07:23,320
Speaker 3: Yes exactly. That's that's a great way to put it

144
00:07:23,320 --> 00:07:25,520
Speaker 3: for the listeners. And because of that, you have to

145
00:07:25,520 --> 00:07:27,880
Speaker 3: climb higher up the tree, right, That takes a lot

146
00:07:27,920 --> 00:07:31,640
Speaker 3: more material, And because of the operational costs are going up,

147
00:07:32,080 --> 00:07:35,640
Speaker 3: minds are kind of moving away from these operations with

148
00:07:35,800 --> 00:07:38,640
Speaker 3: economics just don't make sense. And then the second issue

149
00:07:38,840 --> 00:07:43,280
Speaker 3: on top of that is ESG. So from an ESG perspective.

150
00:07:43,560 --> 00:07:45,840
Speaker 3: In South America, a lot of the issues that minds

151
00:07:45,880 --> 00:07:49,040
Speaker 3: have been facing is water scarcity. And even though the

152
00:07:49,080 --> 00:07:51,920
Speaker 3: companies like bhp I've come up with solutions like using

153
00:07:52,000 --> 00:07:55,640
Speaker 3: desalination to fill that gap of water scarcity, it's just

154
00:07:55,680 --> 00:08:00,240
Speaker 3: not being enough. Desalinating seawater is very expensive, so that's

155
00:08:00,240 --> 00:08:04,640
Speaker 3: why we've seen BHPS Cero Colorado mine in Chilling twenty

156
00:08:04,640 --> 00:08:07,280
Speaker 3: twenty three shutdown because it was just too expensive to

157
00:08:07,360 --> 00:08:10,600
Speaker 3: d salnate. And then we also saw in Panama It's

158
00:08:10,720 --> 00:08:15,040
Speaker 3: Cobra Panama mine closed down because of these environmental issues

159
00:08:15,040 --> 00:08:18,040
Speaker 3: and protests, and this trend is quite pervasive. You see

160
00:08:18,040 --> 00:08:20,880
Speaker 3: it in the US right now. There's a copper project

161
00:08:20,960 --> 00:08:24,080
Speaker 3: that has been in process for twenty years now, but

162
00:08:24,160 --> 00:08:27,680
Speaker 3: because the communities are not comfortable with the fact that,

163
00:08:27,800 --> 00:08:30,880
Speaker 3: you know, these minds have put together strategies to protect

164
00:08:30,920 --> 00:08:33,360
Speaker 3: the environment, to protect them at the end of the day,

165
00:08:33,559 --> 00:08:36,240
Speaker 3: it has led to this long lead time of getting

166
00:08:36,240 --> 00:08:37,400
Speaker 3: that mine operational.

167
00:08:37,800 --> 00:08:41,559
Speaker 1: It almost feels like we're in a moment of transition

168
00:08:42,040 --> 00:08:45,080
Speaker 1: in the copper world, because you know, if the good copper,

169
00:08:45,400 --> 00:08:48,200
Speaker 1: that low hanging fruit has mainly been plucked and you know,

170
00:08:48,200 --> 00:08:50,640
Speaker 1: we mentioned prices going up for copper, but the price

171
00:08:50,720 --> 00:08:54,760
Speaker 1: is currently on enough to really make doing bad copper well,

172
00:08:54,880 --> 00:08:58,080
Speaker 1: including you know, keeping local communities happy, complying with the

173
00:08:58,200 --> 00:09:02,240
Speaker 1: SG you know whatever waters by limitations that there's still

174
00:09:02,520 --> 00:09:04,720
Speaker 1: more upward pressure on prices to come because it's just

175
00:09:04,760 --> 00:09:08,280
Speaker 1: going to be more expensive to produce in the future.

176
00:09:08,520 --> 00:09:10,800
Speaker 2: Let's alone the scarcity exactly.

177
00:09:11,080 --> 00:09:15,400
Speaker 3: So on Bloomberg Terminal's or commodity price forecasts, we see

178
00:09:15,440 --> 00:09:19,560
Speaker 3: copper price forecasts going up to about twelve thousand and

179
00:09:19,640 --> 00:09:23,959
Speaker 3: some and the speculate highs about thirteen five hundred because

180
00:09:24,000 --> 00:09:28,120
Speaker 3: of this constraint, this structural constraint that's coming in terms

181
00:09:28,160 --> 00:09:31,520
Speaker 3: of a lack of supply coming online. High prices just

182
00:09:31,640 --> 00:09:37,080
Speaker 3: makes efficiency gains more accessible. For example, we had there's

183
00:09:37,120 --> 00:09:39,920
Speaker 3: a company called Jetty and then there's a company called

184
00:09:40,000 --> 00:09:43,720
Speaker 3: Rio Tinto. We're trying to leach waste heaps just to

185
00:09:43,760 --> 00:09:48,120
Speaker 3: get additional incremental supply online. But that, unfortunately is not

186
00:09:48,160 --> 00:09:51,240
Speaker 3: a silver bullet in terms of dealing with this deficit

187
00:09:51,280 --> 00:09:51,720
Speaker 3: in supply.

188
00:09:52,160 --> 00:09:54,679
Speaker 1: It's so interesting hearing all of this, and we're going

189
00:09:54,679 --> 00:09:56,480
Speaker 1: to get onto data centers in a moment. But this

190
00:09:56,600 --> 00:09:59,760
Speaker 1: framing is so interesting because I think that, you know,

191
00:10:00,040 --> 00:10:02,880
Speaker 1: as a child growing up, when people were first starting

192
00:10:02,880 --> 00:10:05,560
Speaker 1: to talk about climate change in school and you know,

193
00:10:05,600 --> 00:10:07,680
Speaker 1: the burning of fossil fuels, the other thing they would

194
00:10:07,720 --> 00:10:09,720
Speaker 1: say is, oh, and of course the fossil fuels are

195
00:10:09,720 --> 00:10:12,200
Speaker 1: going to run out eventually, you know, during your lifetime.

196
00:10:12,240 --> 00:10:15,240
Speaker 1: And you know, I'm forty four now and there's no

197
00:10:15,360 --> 00:10:18,080
Speaker 1: sign of the fossil fuels running out. But I always wondered,

198
00:10:18,200 --> 00:10:20,360
Speaker 1: you know, what does that look like? But it almost

199
00:10:20,400 --> 00:10:23,480
Speaker 1: seems like actually we're seeing what that looks like in

200
00:10:23,600 --> 00:10:27,040
Speaker 1: terms of copper, which maybe is just as essential a

201
00:10:27,120 --> 00:10:31,680
Speaker 1: material as to modern society as fossil fuels have been.

202
00:10:31,880 --> 00:10:35,480
Speaker 1: And it really seems like there's a reckoning going to happen, yeah,

203
00:10:35,520 --> 00:10:38,360
Speaker 1: because there's just simply not going to be enough, and

204
00:10:38,400 --> 00:10:41,080
Speaker 1: there's it sounds like there's no getting around that.

205
00:10:41,800 --> 00:10:43,960
Speaker 3: So right now, so if we look at what supply

206
00:10:44,120 --> 00:10:46,679
Speaker 3: looks like at this point in time, it comes from

207
00:10:46,760 --> 00:10:49,920
Speaker 3: a few sources, right so, mind or supply, it comes

208
00:10:49,920 --> 00:10:52,559
Speaker 3: from the circular economy in terms of scrap and then

209
00:10:52,600 --> 00:10:56,000
Speaker 3: obviously these incremental gains that we're trying to get by

210
00:10:56,240 --> 00:11:00,280
Speaker 3: making processing more efficient to extend mind lives. We go

211
00:11:00,360 --> 00:11:03,280
Speaker 3: back to fundamentals from a mind perspective. If we look

212
00:11:03,280 --> 00:11:08,360
Speaker 3: at the geology globally, there are sufficient reserves to close

213
00:11:08,400 --> 00:11:13,120
Speaker 3: this gap, but the issue is exploration. To ascertain which

214
00:11:13,200 --> 00:11:16,439
Speaker 3: reserves are great, which reserves or medium, and which reserves

215
00:11:16,440 --> 00:11:20,000
Speaker 3: are the bad coppers and more economically challenged takes time.

216
00:11:20,280 --> 00:11:23,640
Speaker 3: It takes a lot of KPEX, So exploration KPEX can

217
00:11:23,640 --> 00:11:26,920
Speaker 3: be about like sixty percent of a mind's budget in

218
00:11:26,960 --> 00:11:29,360
Speaker 3: some years, and that's a lot of money for minds

219
00:11:29,400 --> 00:11:33,080
Speaker 3: to invest in these The fact is, yes, it's a reckoning,

220
00:11:33,160 --> 00:11:36,080
Speaker 3: but a reckoning that can be solved if we become

221
00:11:36,120 --> 00:11:40,280
Speaker 3: more aggressive about exploration mining and also try and find

222
00:11:40,280 --> 00:11:44,120
Speaker 3: ways to work around externalities that are currently challenging getting

223
00:11:44,120 --> 00:11:47,120
Speaker 3: these minds from exploration to first poor and some of

224
00:11:47,160 --> 00:11:52,199
Speaker 3: these externalities outside of ESG can also be permitting from governments.

225
00:11:52,520 --> 00:11:56,599
Speaker 3: We need to have more seamless permitting for various licenses

226
00:11:56,679 --> 00:11:58,600
Speaker 3: for these minds to come in. Tract the fact that

227
00:11:58,679 --> 00:12:01,640
Speaker 3: copper minds from x operation to first port take about

228
00:12:01,679 --> 00:12:04,760
Speaker 3: seven to ten years. It's too long. We already seeing

229
00:12:04,800 --> 00:12:07,240
Speaker 3: today that there's a little bit of a rumble in

230
00:12:07,240 --> 00:12:08,679
Speaker 3: the jungle, already.

231
00:12:08,679 --> 00:12:11,200
Speaker 1: That's so interesting, and from what you're saying, it seems

232
00:12:11,240 --> 00:12:15,240
Speaker 1: like the real issue here is not the fundamental scarcity

233
00:12:15,360 --> 00:12:19,320
Speaker 1: of copper. It's that we've all collectively as a global

234
00:12:19,360 --> 00:12:22,360
Speaker 1: economy maybe caught off guard by the change of pace

235
00:12:22,520 --> 00:12:25,520
Speaker 1: in demand, you know, for the energy transition, and then

236
00:12:25,559 --> 00:12:28,320
Speaker 1: now data centers ramping it up even more. Whereas, like

237
00:12:28,440 --> 00:12:30,800
Speaker 1: you know, copper supply was just this kind of thing

238
00:12:30,840 --> 00:12:33,240
Speaker 1: maybe bumbling along at its own pace, and that was

239
00:12:33,280 --> 00:12:37,920
Speaker 1: fine for everyone. Suddenly this increase in urgency means that

240
00:12:38,000 --> 00:12:39,760
Speaker 1: these are issues like permitting delays.

241
00:12:40,000 --> 00:12:40,600
Speaker 2: And it's so funny.

242
00:12:40,600 --> 00:12:42,240
Speaker 1: I mean, when we're talking about building the grid or

243
00:12:42,240 --> 00:12:43,839
Speaker 1: building renewables, permitting.

244
00:12:43,440 --> 00:12:46,320
Speaker 2: Delays, it's the same thing. It just crops up everywhere.

245
00:12:46,320 --> 00:12:49,400
Speaker 1: When we talk about this is you know, the energy transition,

246
00:12:49,640 --> 00:12:52,400
Speaker 1: build out of data centers. It's all about something that

247
00:12:52,480 --> 00:12:55,440
Speaker 1: is changing rapidly at a large infrastructure scale, and we

248
00:12:55,440 --> 00:12:57,320
Speaker 1: don't we're not set up well to do that at

249
00:12:57,360 --> 00:12:59,760
Speaker 1: the moment because maybe for most of our lifetimes the

250
00:12:59,760 --> 00:13:02,920
Speaker 1: world has been in a state of kind of gradual transition.

251
00:13:03,400 --> 00:13:05,120
Speaker 2: Yeah, in relative terms.

252
00:13:05,000 --> 00:13:08,480
Speaker 3: I think with the advent of the energy transition particularly,

253
00:13:08,800 --> 00:13:11,440
Speaker 3: I think it caught people of god, a lot of people.

254
00:13:11,720 --> 00:13:14,800
Speaker 3: We were anticipating it to an extent, but when it happened,

255
00:13:14,800 --> 00:13:18,280
Speaker 3: it happened very quickly, and just that added demand is

256
00:13:18,280 --> 00:13:22,040
Speaker 3: what kind of shocked the initial supplied demand balance. Now

257
00:13:22,080 --> 00:13:24,960
Speaker 3: you have data centers wanting this material, and now the

258
00:13:25,080 --> 00:13:28,199
Speaker 3: US are thinking about, you know, military spending because it's

259
00:13:28,200 --> 00:13:31,240
Speaker 3: spent so much. Copper is like the second top two

260
00:13:31,360 --> 00:13:34,480
Speaker 3: material that it uses. So if you keep adding more

261
00:13:34,520 --> 00:13:39,000
Speaker 3: demand layers with the supply pipeline that is so weak

262
00:13:39,160 --> 00:13:42,240
Speaker 3: and fragile already, Yeah, it's definitely in a course some

263
00:13:42,400 --> 00:13:45,480
Speaker 3: form of reckoning or shock in the market at some point.

264
00:13:46,080 --> 00:13:48,760
Speaker 1: So let's get to talking about copper as it relates

265
00:13:48,800 --> 00:13:51,360
Speaker 1: to data centers, because that is the sort of, I suppose,

266
00:13:51,360 --> 00:13:55,320
Speaker 1: the thing that's the newest in this story of copper,

267
00:13:55,360 --> 00:13:57,920
Speaker 1: the unlikely hero that has sort of risen to the fore.

268
00:13:58,200 --> 00:14:00,520
Speaker 2: How integral is copper to the building of data centers.

269
00:14:01,080 --> 00:14:04,800
Speaker 3: It is quite essential. So you can find copper everywhere

270
00:14:04,960 --> 00:14:09,040
Speaker 3: from like your graphic processing units to your servers, to

271
00:14:09,080 --> 00:14:12,400
Speaker 3: your electrical systems. Within the data centers, you find them

272
00:14:12,520 --> 00:14:14,839
Speaker 3: in the cooling systems, find them in the lights, in

273
00:14:14,880 --> 00:14:18,600
Speaker 3: the ceiling, you find them in redundancy systems that a

274
00:14:18,600 --> 00:14:22,000
Speaker 3: lot of hyperscalers use to guarantee the uptime at the

275
00:14:22,080 --> 00:14:22,640
Speaker 3: end of the day.

276
00:14:22,760 --> 00:14:23,600
Speaker 2: So, just like.

277
00:14:23,720 --> 00:14:26,680
Speaker 3: Blood is important for the circular tree system of the body,

278
00:14:26,840 --> 00:14:30,760
Speaker 3: That's how I view copper. Copper carries like blood carries oxygen,

279
00:14:30,800 --> 00:14:35,479
Speaker 3: which is power and electricity to these integral components and infrastructure.

280
00:14:35,760 --> 00:14:37,480
Speaker 1: And so you know, there's a bit of a note

281
00:14:37,480 --> 00:14:39,600
Speaker 1: here about that. You know, we talk a lot about

282
00:14:39,600 --> 00:14:42,080
Speaker 1: the different types of data centers. I mean, do you

283
00:14:42,120 --> 00:14:44,160
Speaker 1: see that? Is it from a copper point of view?

284
00:14:44,160 --> 00:14:47,520
Speaker 1: All data centers are created the same and AI is

285
00:14:47,600 --> 00:14:50,320
Speaker 1: just more data centers. There is there anything specific about

286
00:14:50,360 --> 00:14:54,200
Speaker 1: some of these applications that impact the necessity around copper.

287
00:14:54,480 --> 00:14:59,479
Speaker 3: So not all data centers use copper in the same intensity,

288
00:14:59,520 --> 00:15:02,120
Speaker 3: if I can say, or at the same rate. So

289
00:15:02,200 --> 00:15:05,560
Speaker 3: if we look at AI in terms of training and inference,

290
00:15:05,600 --> 00:15:07,960
Speaker 3: and I guess the podcast you did with Natalie and

291
00:15:08,040 --> 00:15:10,400
Speaker 3: Helen gave us a great overview in terms of what

292
00:15:10,480 --> 00:15:13,680
Speaker 3: these two workload types are. So in terms of training

293
00:15:14,400 --> 00:15:17,360
Speaker 3: workloads themselves, they're very dense. And what I mean by

294
00:15:17,400 --> 00:15:21,840
Speaker 3: densities that they require a lot more copper but megawatts

295
00:15:21,920 --> 00:15:26,440
Speaker 3: than your inference data centers would, so that basically means

296
00:15:26,480 --> 00:15:29,720
Speaker 3: that you would require copper, more copper in your cooling,

297
00:15:29,880 --> 00:15:34,680
Speaker 3: in your electrical systems, in your GPUs graphical processing units

298
00:15:34,840 --> 00:15:38,320
Speaker 3: within a training data center. Inference, on the other hand,

299
00:15:38,600 --> 00:15:42,400
Speaker 3: is more decentralized, it's more widespread. It doesn't use the

300
00:15:42,520 --> 00:15:45,120
Speaker 3: type of GPUs or the number of GPUs that a

301
00:15:45,200 --> 00:15:48,560
Speaker 3: training center would use, so it's less specialized compared to

302
00:15:48,760 --> 00:15:52,720
Speaker 3: a training data center. Hence it uses less copper per

303
00:15:52,760 --> 00:15:57,160
Speaker 3: square meter. But because inference is so widespread because of

304
00:15:57,240 --> 00:16:01,320
Speaker 3: latency requirements, that all adds up right, So it's both

305
00:16:01,360 --> 00:16:03,640
Speaker 3: of them are quite important at the end of the day.

306
00:16:03,760 --> 00:16:06,120
Speaker 3: And on top of that, you have things like crypto

307
00:16:06,440 --> 00:16:09,520
Speaker 3: cloud based data centers that come into that mix to

308
00:16:09,680 --> 00:16:13,240
Speaker 3: actually build this copper demand over the next decade.

309
00:16:13,640 --> 00:16:15,960
Speaker 1: So, I mean, the question I have that I think

310
00:16:16,000 --> 00:16:18,400
Speaker 1: when I think about all of this is I remember

311
00:16:18,600 --> 00:16:22,040
Speaker 1: at school in science class when we were learning about

312
00:16:22,080 --> 00:16:25,520
Speaker 1: electricity and we were learning about conductors and resistance, and

313
00:16:25,560 --> 00:16:27,640
Speaker 1: the teacher was explaining, how, you know, all of our

314
00:16:27,680 --> 00:16:31,080
Speaker 1: wires are made from copper because it's got such low resistance,

315
00:16:31,360 --> 00:16:33,760
Speaker 1: And I remember saying, you know, one metal that is

316
00:16:33,800 --> 00:16:36,920
Speaker 1: even better at being copper than copper is silver, But

317
00:16:37,000 --> 00:16:40,000
Speaker 1: we would never do it because it's just so ridiculously expensive.

318
00:16:39,520 --> 00:16:40,960
Speaker 2: To have silver wires everywhere.

319
00:16:41,040 --> 00:16:42,760
Speaker 1: So my question is not like, are we going to

320
00:16:42,800 --> 00:16:44,440
Speaker 1: it's copper going to get so expensive that we're going

321
00:16:44,480 --> 00:16:46,880
Speaker 1: to start using silver in our data centers. It's more

322
00:16:47,240 --> 00:16:51,080
Speaker 1: what alternatives exist in all of these different applications, both

323
00:16:51,080 --> 00:16:53,640
Speaker 1: in data centers and outside of data centers.

324
00:16:53,880 --> 00:16:56,560
Speaker 2: I mean, I've listed one silver that doesn't seem.

325
00:16:56,400 --> 00:16:59,000
Speaker 1: Very realistic, but are there are there any sort of

326
00:16:59,040 --> 00:17:00,400
Speaker 1: realistical turn tips?

327
00:17:00,960 --> 00:17:05,159
Speaker 3: So they are definitely fiber optics and aluminum, as the

328
00:17:05,200 --> 00:17:09,800
Speaker 3: Americans say, are definite options for substituting copper and data centers.

329
00:17:10,000 --> 00:17:13,840
Speaker 3: The catch here is, though, that aluminum only has sixty

330
00:17:13,880 --> 00:17:17,480
Speaker 3: percent of the conductivity that copper has. And because these

331
00:17:17,560 --> 00:17:21,160
Speaker 3: data center operators have the cash flow, they have the kpex,

332
00:17:21,560 --> 00:17:24,760
Speaker 3: why would I substitute when I can afford copper. So

333
00:17:24,920 --> 00:17:28,119
Speaker 3: usually the rule of Thumb and industry is if the

334
00:17:28,119 --> 00:17:31,200
Speaker 3: copper price is three point eight times to four times

335
00:17:31,200 --> 00:17:34,879
Speaker 3: the price of aluminium, then we can consider substitution. But

336
00:17:35,400 --> 00:17:39,200
Speaker 3: because of things like the uptime requirements for these data centers,

337
00:17:39,400 --> 00:17:42,280
Speaker 3: you can't add any type of material that would risk that.

338
00:17:42,480 --> 00:17:45,119
Speaker 3: So when it comes to critical components in the data center,

339
00:17:45,200 --> 00:17:48,800
Speaker 3: like you're cooling your IT hardware, it's copper all the way.

340
00:17:48,960 --> 00:17:53,240
Speaker 3: When it comes to external things like security, parking within

341
00:17:53,320 --> 00:17:57,240
Speaker 3: and around the data center, non critical components, that's when

342
00:17:57,280 --> 00:18:00,960
Speaker 3: they could look at using aluminium. Optics in itself has

343
00:18:01,000 --> 00:18:04,920
Speaker 3: already been used for networking, so copper is only used

344
00:18:04,960 --> 00:18:07,680
Speaker 3: in like the first one hundred meter run, and then

345
00:18:07,720 --> 00:18:11,719
Speaker 3: the risk is fiber throughout the facility. So that's basically

346
00:18:11,800 --> 00:18:14,640
Speaker 3: the where you see substitution come in. But it's critical

347
00:18:14,680 --> 00:18:18,119
Speaker 3: to note that without copper in our critical components, you

348
00:18:18,200 --> 00:18:21,480
Speaker 3: can't maintain that ninety nine point nine nine nine time

349
00:18:21,520 --> 00:18:22,440
Speaker 3: at the end of the day.

350
00:18:22,600 --> 00:18:22,960
Speaker 2: Got it.

351
00:18:23,160 --> 00:18:25,000
Speaker 1: So what I'm hearing is there's a little bit of

352
00:18:25,040 --> 00:18:28,840
Speaker 1: substitution in the data center applications, but ultimately the companies

353
00:18:29,080 --> 00:18:31,960
Speaker 1: building data centers, they have the money to pay for

354
00:18:32,040 --> 00:18:34,119
Speaker 1: more expensive copper, so they're not going to be the

355
00:18:34,119 --> 00:18:38,520
Speaker 1: ones to either substitute it out completely or to not

356
00:18:38,640 --> 00:18:43,439
Speaker 1: build their thing, so they're going to be displacing someone

357
00:18:43,440 --> 00:18:47,520
Speaker 1: else's copper applications. So I suppose what is the maybe

358
00:18:47,560 --> 00:18:50,560
Speaker 1: the marginal application. Who loses out if data centers come

359
00:18:50,600 --> 00:18:53,199
Speaker 1: and hoover up a bunch of copper supply? What is

360
00:18:53,200 --> 00:18:56,119
Speaker 1: the thing that doesn't get built or becomes unviable or

361
00:18:56,160 --> 00:18:57,200
Speaker 1: has to substitute.

362
00:18:57,520 --> 00:19:00,960
Speaker 3: That's an interesting question because I guess when we looked

363
00:19:01,000 --> 00:19:05,040
Speaker 3: at those sectors before, like your traditional and energy transition sectors,

364
00:19:05,119 --> 00:19:08,399
Speaker 3: those are data centers competitors. Now, when it comes to

365
00:19:08,440 --> 00:19:11,679
Speaker 3: looking at the scale of kpex or just money that

366
00:19:11,800 --> 00:19:15,600
Speaker 3: data center builders and operators are throwing at this infrastructure,

367
00:19:15,720 --> 00:19:19,520
Speaker 3: the question really turns around to who has off take,

368
00:19:19,800 --> 00:19:24,320
Speaker 3: Who has long term supply chain security with these fabricators

369
00:19:24,320 --> 00:19:28,520
Speaker 3: that make these components, Who has those established relationships. So

370
00:19:28,840 --> 00:19:30,879
Speaker 3: I think the story should be not so much that

371
00:19:30,960 --> 00:19:35,080
Speaker 3: there's competition, but a bit of a refocus on who

372
00:19:35,320 --> 00:19:39,240
Speaker 3: is setting themselves up, given that we've given these indications

373
00:19:39,240 --> 00:19:41,720
Speaker 3: that this is coming. Who's setting themselves up for long

374
00:19:41,800 --> 00:19:45,240
Speaker 3: term success, Who's securing their supply chain to ensure that

375
00:19:45,320 --> 00:19:47,000
Speaker 3: they have a piece of the pie.

376
00:19:47,400 --> 00:19:51,280
Speaker 1: Kind of feels like the organizations that have kind of

377
00:19:51,320 --> 00:19:53,520
Speaker 1: done their homework and set it kind of reminds me

378
00:19:53,560 --> 00:19:56,639
Speaker 1: of you know, when you're at university and there's always

379
00:19:56,640 --> 00:19:58,879
Speaker 1: that person who books the one book out of the

380
00:19:58,920 --> 00:20:01,800
Speaker 1: library that everyone is going to need for this week's assignment.

381
00:20:01,880 --> 00:20:04,760
Speaker 1: This week's assignment is having copper, and someone has booked

382
00:20:04,800 --> 00:20:07,440
Speaker 1: that book out the library. We can't not talk about

383
00:20:07,480 --> 00:20:09,560
Speaker 1: the US in all of this because it's kind of

384
00:20:09,560 --> 00:20:11,800
Speaker 1: fifty percent of the world's data centers. Are there a

385
00:20:11,880 --> 00:20:14,800
Speaker 1: huge amount of the growth in data centers expected there.

386
00:20:14,800 --> 00:20:18,479
Speaker 1: And we mentioned earlier that the domestic supply of copper

387
00:20:18,480 --> 00:20:21,840
Speaker 1: in the US is not sufficient to meet US demands.

388
00:20:21,960 --> 00:20:23,199
Speaker 2: So what are they going to do?

389
00:20:23,480 --> 00:20:26,600
Speaker 3: In twenty twenty four, they consumed about one point six

390
00:20:26,640 --> 00:20:30,400
Speaker 3: million times right domestically, they produce about half of that

391
00:20:30,600 --> 00:20:34,200
Speaker 3: for their applications for their demand, and then they import

392
00:20:34,359 --> 00:20:37,600
Speaker 3: from China and the rest. So the narrative goes back

393
00:20:37,640 --> 00:20:41,600
Speaker 3: to how they strategize one, build more minds. But in

394
00:20:41,640 --> 00:20:44,200
Speaker 3: the US the minds are also a lot of great

395
00:20:44,280 --> 00:20:48,840
Speaker 3: quality two decision making around tariffs with specific countries, so

396
00:20:48,880 --> 00:20:52,080
Speaker 3: they have to be very strategic about what they tariff

397
00:20:52,080 --> 00:20:54,919
Speaker 3: and what they don't. For example, in China, there's a

398
00:20:54,960 --> 00:20:58,720
Speaker 3: stalemate in terms of tariffs that have been paused because

399
00:20:58,760 --> 00:21:02,000
Speaker 3: the US needs some of the refined material that only

400
00:21:02,080 --> 00:21:04,919
Speaker 3: China produces and hence they've kind of stopped that and

401
00:21:04,960 --> 00:21:07,160
Speaker 3: we're waiting for the end of this month to kind

402
00:21:07,200 --> 00:21:10,040
Speaker 3: of see what's going to happen, what conversation is going

403
00:21:10,119 --> 00:21:13,480
Speaker 3: to open from their other countries that they could form

404
00:21:13,600 --> 00:21:17,919
Speaker 3: better relationships with Peru, Chile and then the DRC. But

405
00:21:18,119 --> 00:21:20,959
Speaker 3: that in itself there's a lot it can do, but

406
00:21:21,000 --> 00:21:23,679
Speaker 3: it's also the way it does it in terms of

407
00:21:23,760 --> 00:21:27,600
Speaker 3: trade relations to secure their own domestic supply and hence

408
00:21:27,600 --> 00:21:29,280
Speaker 3: the buildout of these data centers.

409
00:21:29,400 --> 00:21:32,840
Speaker 1: So there's a little bit of nuanced strategy required with

410
00:21:33,000 --> 00:21:37,080
Speaker 1: certain key players. And yeah, we'll have to see whether

411
00:21:37,320 --> 00:21:40,760
Speaker 1: the US can dance that dance effectively, because you know,

412
00:21:41,040 --> 00:21:42,760
Speaker 1: I mean, I suppose in all of this as well,

413
00:21:42,920 --> 00:21:45,080
Speaker 1: we're saying copper it sounds like it's in the top

414
00:21:45,160 --> 00:21:48,479
Speaker 1: tier of strategic materials, but it's not the only thing

415
00:21:48,560 --> 00:21:51,520
Speaker 1: in the top tier, and it's probably I guess what

416
00:21:51,520 --> 00:21:54,159
Speaker 1: you're describing is not as simple as it sounds, you know,

417
00:21:54,320 --> 00:21:56,119
Speaker 1: just being friends with who you need to be friends with.

418
00:21:57,240 --> 00:22:01,520
Speaker 1: So given the high demand for copper and in certain countries,

419
00:22:01,560 --> 00:22:05,040
Speaker 1: including the US specifically, there's going to be a bit

420
00:22:05,080 --> 00:22:08,840
Speaker 1: of a gap. What kind of actions and changes to regulations.

421
00:22:08,920 --> 00:22:10,840
Speaker 1: Could you see them taking to address that?

422
00:22:11,400 --> 00:22:15,800
Speaker 3: So I think just harboring on the US narrative in

423
00:22:15,840 --> 00:22:19,919
Speaker 3: itself when we go back to you ESG issues. So

424
00:22:20,000 --> 00:22:23,600
Speaker 3: let's take out tariffs, let's take out their lack of supply.

425
00:22:24,000 --> 00:22:26,640
Speaker 3: I think one of the key issues that they might

426
00:22:26,720 --> 00:22:29,720
Speaker 3: be able to start off with one. They've now added

427
00:22:29,960 --> 00:22:33,399
Speaker 3: copper as a critical mineral or it's in proposal stage,

428
00:22:33,400 --> 00:22:34,439
Speaker 3: which is great.

429
00:22:34,760 --> 00:22:38,080
Speaker 1: I just clarify because when you say they've been classified

430
00:22:38,080 --> 00:22:40,800
Speaker 1: as a critical mineral is I always thought that when

431
00:22:40,840 --> 00:22:42,879
Speaker 1: we say something's a critical mineral, that was just like

432
00:22:43,040 --> 00:22:46,080
Speaker 1: it was no more than a label reflecting our opinion

433
00:22:46,160 --> 00:22:48,879
Speaker 1: that it is critical. And I think I might be

434
00:22:48,960 --> 00:22:51,919
Speaker 1: learning something new here is critical mineral and official designation

435
00:22:52,080 --> 00:22:54,199
Speaker 1: in the US.

436
00:22:53,920 --> 00:22:57,000
Speaker 3: Not just the US. It's a term that has slowly

437
00:22:57,680 --> 00:23:01,879
Speaker 3: been developed for US as an industry to understand that

438
00:23:01,960 --> 00:23:05,840
Speaker 3: criticality means something that a region or a country may

439
00:23:06,000 --> 00:23:09,960
Speaker 3: have lack of and need to focus resources on to

440
00:23:10,320 --> 00:23:13,719
Speaker 3: close that gap, to not be behind at the end

441
00:23:13,720 --> 00:23:15,919
Speaker 3: of the day, to rather build up surplus of it.

442
00:23:16,080 --> 00:23:19,440
Speaker 3: So as soon as you see critical criticality means there's

443
00:23:19,440 --> 00:23:22,080
Speaker 3: a little bit of trouble we need these minerals to

444
00:23:22,359 --> 00:23:24,720
Speaker 3: grow our economy, we need these minerals to grow out

445
00:23:24,720 --> 00:23:27,520
Speaker 3: our industry, and if we might not have enough of

446
00:23:27,560 --> 00:23:28,800
Speaker 3: those on those lists.

447
00:23:29,880 --> 00:23:33,480
Speaker 1: So the EU in questions around raw materials that EU

448
00:23:33,640 --> 00:23:37,160
Speaker 1: is typically not a supplier but a source of demand.

449
00:23:37,440 --> 00:23:40,480
Speaker 1: How do they play in this world where copper is

450
00:23:40,480 --> 00:23:43,560
Speaker 1: in short supply but is incredibly important for the future.

451
00:23:43,960 --> 00:23:46,800
Speaker 3: So the EU in itself has also come up with

452
00:23:46,840 --> 00:23:50,640
Speaker 3: its own critical mineral strategy, and one key point around

453
00:23:50,720 --> 00:23:55,120
Speaker 3: that is recycling or the circular economy. So they've put

454
00:23:55,119 --> 00:23:58,919
Speaker 3: a lot of effort in backing projects several projects that

455
00:23:59,119 --> 00:24:03,600
Speaker 3: now focus on recycling. They very consciously do follow like

456
00:24:03,760 --> 00:24:07,879
Speaker 3: fabricated scrap to ensure they understand where it's coming from,

457
00:24:08,000 --> 00:24:11,800
Speaker 3: where it's going within their region. They also focused on

458
00:24:12,200 --> 00:24:15,520
Speaker 3: limiting exports of scrap out so that they themselves can

459
00:24:15,720 --> 00:24:19,160
Speaker 3: use it, which helps with them maintaining their own inventories.

460
00:24:19,440 --> 00:24:23,280
Speaker 3: So while there are initiatives that the EU is taking

461
00:24:23,440 --> 00:24:26,640
Speaker 3: up because they are such a high demand center, they

462
00:24:26,680 --> 00:24:29,120
Speaker 3: still need to go to the countries that are actually

463
00:24:29,320 --> 00:24:31,960
Speaker 3: you know, supplying the actual or at the end of

464
00:24:31,960 --> 00:24:34,040
Speaker 3: the day, and they still need to be aware that

465
00:24:34,080 --> 00:24:36,200
Speaker 3: what they're doing is not a silver bullet for their

466
00:24:36,240 --> 00:24:40,280
Speaker 3: domestic demand. Exports or imports are quite critical for them

467
00:24:40,359 --> 00:24:44,040
Speaker 3: to maintain their copper suppliers. One interesting thing that I

468
00:24:44,080 --> 00:24:46,600
Speaker 3: found though, Tom is a bit of a project between

469
00:24:46,640 --> 00:24:51,320
Speaker 3: Glencore and Schneider Electric. Right, so glen Core minds copper,

470
00:24:51,680 --> 00:24:55,680
Speaker 3: Schneider Electric makes the components that goes into these data centers.

471
00:24:55,920 --> 00:24:59,760
Speaker 3: They have formed a partnership, so they've basically secured their

472
00:24:59,760 --> 00:25:02,320
Speaker 3: fe future copper supply. And we might see a lot

473
00:25:02,400 --> 00:25:05,560
Speaker 3: of the EU doing those partnerships to be able to

474
00:25:05,680 --> 00:25:07,520
Speaker 3: secure their own copper supply.

475
00:25:07,920 --> 00:25:10,680
Speaker 1: So basically the strategy for the EU is to sign

476
00:25:10,720 --> 00:25:12,960
Speaker 1: the book out the library before everyone else does.

477
00:25:13,240 --> 00:25:13,520
Speaker 2: Yes.

478
00:25:13,920 --> 00:25:16,200
Speaker 1: Finally, I mean when we ever were talking about something

479
00:25:16,240 --> 00:25:19,240
Speaker 1: being critical and in short supply, all of that nuance

480
00:25:19,280 --> 00:25:21,439
Speaker 1: can boil down into a number, which is the price.

481
00:25:21,760 --> 00:25:23,760
Speaker 1: And you mentioned that we've seen, you know, even in

482
00:25:23,760 --> 00:25:26,840
Speaker 1: the last month, the price of copper going up, indicating

483
00:25:27,119 --> 00:25:29,800
Speaker 1: a sort of a recognition that this is now becoming

484
00:25:29,920 --> 00:25:32,280
Speaker 1: a kind of a very very critical material.

485
00:25:32,440 --> 00:25:33,360
Speaker 2: What is the outlook?

486
00:25:33,520 --> 00:25:35,040
Speaker 1: Do you have a view on where the price of

487
00:25:35,080 --> 00:25:37,080
Speaker 1: copper is headed or does it do other people have

488
00:25:37,119 --> 00:25:37,800
Speaker 1: a view on that.

489
00:25:38,080 --> 00:25:42,320
Speaker 3: In being if we use the commodity price forecast consensus,

490
00:25:42,320 --> 00:25:45,880
Speaker 3: which is basically a tool on terminal where analyst comes

491
00:25:45,920 --> 00:25:50,280
Speaker 3: together and forecasts the mean, mid, and high copper prices.

492
00:25:50,520 --> 00:25:53,919
Speaker 3: And right now copper prices are quite elevated, close to

493
00:25:54,280 --> 00:25:58,480
Speaker 3: ten thousand dollars per ton, and we have analysts speculating

494
00:25:58,520 --> 00:26:01,280
Speaker 3: that by twenty twenty eight we could see highs about

495
00:26:01,280 --> 00:26:05,919
Speaker 3: thirteenery five hundred, which is quite substantial and it's quite

496
00:26:06,160 --> 00:26:09,600
Speaker 3: telling of what's happening within the market from a tight

497
00:26:09,640 --> 00:26:12,520
Speaker 3: supply point of view, and hopefully that will be a

498
00:26:12,560 --> 00:26:16,479
Speaker 3: wake up signal for these different sectors to see just

499
00:26:16,560 --> 00:26:21,240
Speaker 3: how material copper could be within their different infrastructural buildout. Now,

500
00:26:21,240 --> 00:26:23,960
Speaker 3: when you translate that to data centers, we did a

501
00:26:23,960 --> 00:26:27,360
Speaker 3: bit of study on data center builds that cost between

502
00:26:27,480 --> 00:26:31,200
Speaker 3: seventeen million and I think thirty six million per megawatt,

503
00:26:31,320 --> 00:26:34,080
Speaker 3: and what we found there is that copper can account

504
00:26:34,080 --> 00:26:37,879
Speaker 3: for about three to six percent of capital expenditure. While

505
00:26:37,920 --> 00:26:40,719
Speaker 3: that doesn't sound like a big number, and while we

506
00:26:40,800 --> 00:26:44,560
Speaker 3: know hyperscalers can afford this number, what we would like

507
00:26:44,640 --> 00:26:47,840
Speaker 3: to emphasize that this is a very important line item

508
00:26:47,960 --> 00:26:51,639
Speaker 3: on the materiality of copper and that that money still

509
00:26:51,720 --> 00:26:54,680
Speaker 3: can translate to something quite substantial that can be used

510
00:26:54,680 --> 00:26:57,720
Speaker 3: for something else. So it's just for us to encourage

511
00:26:57,840 --> 00:27:01,120
Speaker 3: these builders to just like watch the market and understand

512
00:27:01,119 --> 00:27:03,520
Speaker 3: how it impacts how you build out your infrastructure.

513
00:27:03,800 --> 00:27:06,040
Speaker 1: Yeah, and I think the other part of it as

514
00:27:06,040 --> 00:27:09,679
Speaker 1: well is that even if they can ultimately afford it

515
00:27:09,720 --> 00:27:12,399
Speaker 1: and decide it's worth it, someone somewhere is going to

516
00:27:12,480 --> 00:27:15,000
Speaker 1: feel the crunch on this. And yeah, it seems that

517
00:27:15,240 --> 00:27:18,479
Speaker 1: copper is critical to everything. I've learned so much today.

518
00:27:18,640 --> 00:27:20,320
Speaker 1: I did not know there was such a thing as

519
00:27:20,320 --> 00:27:22,760
Speaker 1: good copper and bad copper for a start, and that

520
00:27:22,920 --> 00:27:25,240
Speaker 1: was really interesting. And you know, maybe the future it

521
00:27:25,240 --> 00:27:27,080
Speaker 1: won't be good copper and bad copper will be good

522
00:27:27,119 --> 00:27:28,800
Speaker 1: copper and not quite as good copper.

523
00:27:28,880 --> 00:27:31,000
Speaker 2: If we figure out a.

524
00:27:31,000 --> 00:27:33,840
Speaker 1: Way to mitigate some of these environmental issues and of

525
00:27:33,840 --> 00:27:38,040
Speaker 1: the cost issues permitting, maybe that's the road to salvation

526
00:27:38,200 --> 00:27:40,120
Speaker 1: for us, is to figure out how to make bad

527
00:27:40,160 --> 00:27:43,720
Speaker 1: copper better. But it's been really fascinating, and I did

528
00:27:43,760 --> 00:27:46,960
Speaker 1: not have an appreciation going into this of quite how

529
00:27:47,119 --> 00:27:50,840
Speaker 1: critical copper is to just everything. So Rose, thank you

530
00:27:50,880 --> 00:27:52,479
Speaker 1: so much for joining today.

531
00:27:52,640 --> 00:27:55,240
Speaker 3: Thank you so much, Tom for having me today. I'm

532
00:27:55,280 --> 00:27:58,639
Speaker 3: happy to talk a lot more about this. It's just

533
00:27:58,720 --> 00:28:00,600
Speaker 3: one of those things we need to keep boff finger

534
00:28:00,680 --> 00:28:03,640
Speaker 3: on the doll on because the marketer is deffinitely becoming

535
00:28:03,760 --> 00:28:05,280
Speaker 3: quite exciting going forward.

536
00:28:05,480 --> 00:28:09,119
Speaker 1: Well, those of you listening who are BNF clients, you

537
00:28:09,200 --> 00:28:12,040
Speaker 1: literally can talk to Rose more about this. Get in

538
00:28:12,119 --> 00:28:14,560
Speaker 1: touch with your rep and they can set up a conversation.

539
00:28:15,040 --> 00:28:24,440
Speaker 3: Thanks Tom.

540
00:28:24,600 --> 00:28:27,760
Speaker 4: Today's episode of Switched On was produced by Cam Gray

541
00:28:27,960 --> 00:28:31,679
Speaker 4: with production assistants from Kamala Shelling. Bloomberg NIF is a

542
00:28:31,680 --> 00:28:34,840
Speaker 4: service provided by Bloomberg Finance LP and its affiliates. This

543
00:28:34,920 --> 00:28:37,600
Speaker 4: recording does not constitute, nor should it be construed as

544
00:28:37,640 --> 00:28:41,400
Speaker 4: investment in vice, investment recommendations, or a recommendation as to

545
00:28:41,440 --> 00:28:44,280
Speaker 4: an investment or other strategy. Bloomberg ANIF should not be

546
00:28:44,360 --> 00:28:48,080
Speaker 4: considered as information sufficient upon which to base an investment decision.

547
00:28:48,240 --> 00:28:51,200
Speaker 4: Neither Bloomberg Finance LP nor any of its affiliates makes

548
00:28:51,240 --> 00:28:54,959
Speaker 4: any representation or warranty as to the accuracy or completeness

549
00:28:54,960 --> 00:28:57,960
Speaker 4: of the information contained in this recording, and any liability

550
00:28:58,000 --> 00:29:00,680
Speaker 4: as a result of this recording is expressly it is quiet.

551
00:29:09,560 --> 00:29:09,760
Speaker 2: Hm