WEBVTT - Where are Semiconductors Headed Under Trump? 

0:00:02.440 --> 0:00:06.800
<v Speaker 1>Bloomberg Audio Studios, podcasts, radio news.

0:00:07.320 --> 0:00:10.600
<v Speaker 2>You're listening to Bloomberg Business Week with krol Messer and

0:00:10.680 --> 0:00:14.440
<v Speaker 2>Tim Stenebek on Bloomberg Radio. We've got with us The

0:00:14.520 --> 0:00:17.239
<v Speaker 2>guy who wrote the book on chips, Chris Miller is

0:00:17.280 --> 0:00:20.840
<v Speaker 2>Associate Professor of International history at Tufts University. He's the

0:00:20.880 --> 0:00:23.319
<v Speaker 2>author of the New York Times bestseller Chip War, The

0:00:23.360 --> 0:00:25.880
<v Speaker 2>Fight for the World's Most Critical Technology. That book, a

0:00:25.960 --> 0:00:29.680
<v Speaker 2>geopolitical history of the computer chip, also a winner of

0:00:29.720 --> 0:00:33.240
<v Speaker 2>the FT Business Book of the Year Award. Chris also

0:00:33.479 --> 0:00:36.720
<v Speaker 2>advises semiconductor companies and tech startups and investors. He joins

0:00:36.800 --> 0:00:38.000
<v Speaker 2>us from Naples, Florida.

0:00:38.080 --> 0:00:39.880
<v Speaker 3>Oh my god, Chris, Like, where to start? You wrote

0:00:39.920 --> 0:00:41.960
<v Speaker 3>this in twenty twenty two. A lot has happened in

0:00:42.000 --> 0:00:44.800
<v Speaker 3>the last two years. How are you thinking about what

0:00:44.840 --> 0:00:46.720
<v Speaker 3>you wrote and kind of where we are today in

0:00:46.760 --> 0:00:47.440
<v Speaker 3>the chip space?

0:00:49.120 --> 0:00:52.360
<v Speaker 1>Like, there are two big shifts happening simultaneously right now.

0:00:52.640 --> 0:00:56.800
<v Speaker 1>One is the surge of spending on AI infrastructure, which

0:00:56.840 --> 0:01:00.840
<v Speaker 1>means lots of very expensive AI accelerator in high powered

0:01:00.880 --> 0:01:04.080
<v Speaker 1>data centers. And two is the ratcheting up of tension

0:01:04.120 --> 0:01:07.160
<v Speaker 1>around the US and China with the new administration in

0:01:07.200 --> 0:01:10.120
<v Speaker 1>the US coming into office in just a month's time,

0:01:10.280 --> 0:01:16.040
<v Speaker 1>planning to increase tariffs and controls on China's semiconductor industry.

0:01:16.120 --> 0:01:18.720
<v Speaker 1>So those two trends are intersecting in ways that impact

0:01:18.800 --> 0:01:20.639
<v Speaker 1>pretty much every company in the sector.

0:01:20.760 --> 0:01:23.000
<v Speaker 2>Chris, I'm curious, you know what we're hearing about the

0:01:23.040 --> 0:01:26.320
<v Speaker 2>Trump transition team and their plans when the Trump administration

0:01:26.400 --> 0:01:29.560
<v Speaker 2>takes office on January twentieth, they're ramping up into full gear.

0:01:29.600 --> 0:01:34.080
<v Speaker 2>No question, have you heard from anyone on the transition team.

0:01:34.480 --> 0:01:36.840
<v Speaker 1>Well, I think the transition is in the process of

0:01:37.200 --> 0:01:39.280
<v Speaker 1>sketching out who will be the officials who will be

0:01:39.319 --> 0:01:43.480
<v Speaker 1>responsible for chip policy. We know the cabinet secretaries Marco

0:01:43.560 --> 0:01:46.560
<v Speaker 1>Rubio at State, Howard Lutnik at Commerce, but a lot

0:01:46.600 --> 0:01:48.600
<v Speaker 1>of the key decisions we will be shaped by the

0:01:48.640 --> 0:01:51.919
<v Speaker 1>mid level officials will be responsible for implementing the Chips

0:01:51.960 --> 0:01:55.800
<v Speaker 1>Act or implementing the export controls on chip technology, and

0:01:55.840 --> 0:01:58.000
<v Speaker 1>those are names that we just don't know yet, But

0:01:58.040 --> 0:01:59.960
<v Speaker 1>I think it'll be really worth following who gets a

0:02:00.080 --> 0:02:02.600
<v Speaker 1>pointed to those roles because they'll have a pretty big

0:02:02.600 --> 0:02:04.080
<v Speaker 1>influence on the shape of policy.

0:02:04.200 --> 0:02:08.040
<v Speaker 2>What would you advise in terms of names in terms

0:02:08.080 --> 0:02:11.639
<v Speaker 2>of people, in terms of expertise and policy, if you

0:02:11.680 --> 0:02:13.200
<v Speaker 2>were speaking with the transition team.

0:02:15.120 --> 0:02:18.440
<v Speaker 1>What we've seen thus far is that Trump has opted

0:02:18.520 --> 0:02:22.120
<v Speaker 1>for a range of officials with different views. Some folks

0:02:22.240 --> 0:02:25.280
<v Speaker 1>very close to financial markets who want a very stable

0:02:26.000 --> 0:02:30.360
<v Speaker 1>US China economic and financial relationship. Others like Marco Rubio,

0:02:30.720 --> 0:02:34.079
<v Speaker 1>Secretary of State nominee and noted China hawk, who's advocated

0:02:34.320 --> 0:02:37.920
<v Speaker 1>more decoupling not just of the supply chain and technology relationship,

0:02:37.960 --> 0:02:40.440
<v Speaker 1>but also the financial relationship too. And I think this

0:02:40.520 --> 0:02:41.960
<v Speaker 1>is similar in a lot of ways to what we

0:02:42.000 --> 0:02:44.600
<v Speaker 1>saw in the first trumpdministration, where there really was a

0:02:44.639 --> 0:02:48.000
<v Speaker 1>broad spectrum of views among Trump's advisors and ultimately up

0:02:48.040 --> 0:02:50.000
<v Speaker 1>to Trump to make the final decision.

0:02:50.200 --> 0:02:53.079
<v Speaker 3>Hey, listen, I do think about the struggle between policy

0:02:53.160 --> 0:02:58.400
<v Speaker 3>and government a new administration versus the realities of big

0:02:58.440 --> 0:03:00.640
<v Speaker 3>tech companies who want to sell their stuff to everyone.

0:03:00.680 --> 0:03:02.400
<v Speaker 3>They're thinking about their top and bottom line. The New

0:03:02.480 --> 0:03:05.880
<v Speaker 3>York Times out with the story today and they get

0:03:05.880 --> 0:03:08.919
<v Speaker 3>into how the US firms, so with these big tech

0:03:08.960 --> 0:03:13.320
<v Speaker 3>firms battles with that government contract to keep tech sales

0:03:13.600 --> 0:03:17.120
<v Speaker 3>to China, to kind of keep that crack down going.

0:03:18.880 --> 0:03:21.200
<v Speaker 3>Where are we going to see that kind of line

0:03:21.240 --> 0:03:24.240
<v Speaker 3>in the sand when it comes to selling technology to China,

0:03:24.960 --> 0:03:27.639
<v Speaker 3>and then the big tech companies saying, but wait a minute,

0:03:27.720 --> 0:03:30.480
<v Speaker 3>this starts to impact our business because China is a

0:03:30.520 --> 0:03:34.880
<v Speaker 3>massive market. Does the battle continue? I don't know thoughts

0:03:34.920 --> 0:03:38.040
<v Speaker 3>about that constructive thoughts, because I'm assuming, as we said

0:03:38.040 --> 0:03:40.600
<v Speaker 3>in the introduction, that you talk to some of these

0:03:40.640 --> 0:03:44.360
<v Speaker 3>major players in the you know, our consult in the

0:03:44.400 --> 0:03:46.960
<v Speaker 3>semi industry, and I'm curious what you are hearing about

0:03:46.960 --> 0:03:49.400
<v Speaker 3>them about the pressure they are putting on government when

0:03:49.400 --> 0:03:50.360
<v Speaker 3>it comes to policy.

0:03:52.000 --> 0:03:54.320
<v Speaker 1>I think you talked to any CEO or executive off

0:03:54.360 --> 0:03:57.000
<v Speaker 1>the record, and they'll say that my biggest problem is

0:03:57.000 --> 0:03:59.840
<v Speaker 1>that China is both my largest customer and my number

0:03:59.840 --> 0:04:02.560
<v Speaker 1>one competitor in most cases, and that I think structures

0:04:02.560 --> 0:04:05.600
<v Speaker 1>how the industry deals with this issue and tries to

0:04:05.640 --> 0:04:08.280
<v Speaker 1>lobby the US government. They all sort of understand the

0:04:08.280 --> 0:04:11.240
<v Speaker 1>logic of the US government, understand that chips are key

0:04:11.320 --> 0:04:14.600
<v Speaker 1>for AI, understand that AI is key to national security

0:04:14.600 --> 0:04:17.719
<v Speaker 1>as well as economic prosperity. But they all also realize

0:04:17.720 --> 0:04:21.160
<v Speaker 1>that they have really substantial markets in China, and so

0:04:21.480 --> 0:04:23.800
<v Speaker 1>wherever you look in the supply chain, of the makers

0:04:23.800 --> 0:04:27.159
<v Speaker 1>of the tools that make chips, the manufacturers of chips themselves,

0:04:27.320 --> 0:04:29.200
<v Speaker 1>they're all going to lobby, and they all have lobbied

0:04:29.279 --> 0:04:32.599
<v Speaker 1>very successfully to keep their ability to sell to China open.

0:04:33.279 --> 0:04:36.960
<v Speaker 1>And the National Security Apparatus, the Pentagon, the Commerce Department,

0:04:36.960 --> 0:04:39.760
<v Speaker 1>of the intelligence agencies are pushing for more restrictions than

0:04:40.040 --> 0:04:43.320
<v Speaker 1>companies themselves want, and that a balancing act that any

0:04:43.360 --> 0:04:45.640
<v Speaker 1>president of got a balance. I think we've seen the

0:04:45.720 --> 0:04:50.360
<v Speaker 1>Biden administration end up with a more lapse approach than

0:04:50.520 --> 0:04:53.800
<v Speaker 1>one might have expected given their early tough rhetoric on

0:04:53.800 --> 0:04:55.320
<v Speaker 1>this issue, and it'll be interesting to watch how the

0:04:55.360 --> 0:04:58.839
<v Speaker 1>Trump administration addresses this balance. There will certainly be voices

0:04:58.839 --> 0:05:02.200
<v Speaker 1>of administration saying, let's focus on continue to export to

0:05:02.360 --> 0:05:05.640
<v Speaker 1>China of technology. It's important economically. But there will also

0:05:05.640 --> 0:05:08.440
<v Speaker 1>be voices of the opposite side saying, if you export

0:05:08.640 --> 0:05:12.000
<v Speaker 1>advanced technology to China, you're only giving China the capability

0:05:12.120 --> 0:05:15.200
<v Speaker 1>to learn how to manufacture high end SIPs domestically and

0:05:15.200 --> 0:05:17.799
<v Speaker 1>eventually replace US firms in the Chinese market.

0:05:17.920 --> 0:05:20.160
<v Speaker 3>Chris, with your understanding and knowledge of this world, are

0:05:20.160 --> 0:05:24.719
<v Speaker 3>we right us, the United States right to be concerned

0:05:25.080 --> 0:05:28.560
<v Speaker 3>about exporting technology to China. We've certainly seen this play

0:05:28.600 --> 0:05:31.239
<v Speaker 3>out where things get exported the technology and China becomes

0:05:31.320 --> 0:05:32.920
<v Speaker 3>really good at doing things. We're seeing it kind of

0:05:32.960 --> 0:05:35.760
<v Speaker 3>play out in the EV market as we speak. So

0:05:35.839 --> 0:05:40.040
<v Speaker 3>I'm curious, how concerned should we that we will export

0:05:40.120 --> 0:05:47.800
<v Speaker 3>our expertise, our knowledge and lose control of another industry.

0:05:45.839 --> 0:05:48.200
<v Speaker 1>You know. I think it's worth noting that China only

0:05:48.279 --> 0:05:53.280
<v Speaker 1>imports a couple of types of manufactured goods semiconductors semiconductor

0:05:53.360 --> 0:05:56.400
<v Speaker 1>manufacturing equipment. Those are the two largest categories of China's

0:05:56.400 --> 0:05:59.400
<v Speaker 1>manufacture good imports, because China can produce everything else that

0:05:59.520 --> 0:06:02.400
<v Speaker 1>needs domestically when it comes to manufactured goods, and I

0:06:02.440 --> 0:06:05.640
<v Speaker 1>think that speaks to China's capabilities in this sphere, which

0:06:05.680 --> 0:06:08.720
<v Speaker 1>means that the more capabilities you give to China, the

0:06:08.760 --> 0:06:10.679
<v Speaker 1>more they can be able to use those to replace

0:06:10.720 --> 0:06:14.080
<v Speaker 1>Western firms. And they're desired to replace Western firms, and

0:06:14.120 --> 0:06:17.719
<v Speaker 1>I think eminently well documented. It's evidence in the fact

0:06:17.720 --> 0:06:21.160
<v Speaker 1>that they don't buy anything abroad unless they absolutely have to.

0:06:21.720 --> 0:06:23.920
<v Speaker 1>And I think some connector companies are all worried that

0:06:23.960 --> 0:06:27.000
<v Speaker 1>their market share is going to decline over time, and

0:06:27.040 --> 0:06:29.000
<v Speaker 1>I think they all realize that the more chip making

0:06:29.000 --> 0:06:31.040
<v Speaker 1>tools that are auld to China accelerates that process.

0:06:31.080 --> 0:06:33.719
<v Speaker 3>Hey, listen, we want to get to Broadcom because they

0:06:33.880 --> 0:06:35.599
<v Speaker 3>are out with their earnings and the stock is up

0:06:35.640 --> 0:06:38.440
<v Speaker 3>about two and a half percent here in the market.

0:06:38.520 --> 0:06:40.960
<v Speaker 3>Actually now it's down about one percent. It's bouncing around

0:06:41.000 --> 0:06:43.200
<v Speaker 3>here to be fair, but we did see a quick rally.

0:06:43.240 --> 0:06:44.840
<v Speaker 3>So let's get to some of the headlines, and let's

0:06:44.880 --> 0:06:48.160
<v Speaker 3>go to the outlook. Company sees first quarter revenue of

0:06:48.160 --> 0:06:51.159
<v Speaker 3>about fourteen point six billion. That's spot on in terms

0:06:51.200 --> 0:06:54.360
<v Speaker 3>of what analysts we're expecting fourth quarter just at EPs,

0:06:54.360 --> 0:06:57.240
<v Speaker 3>so looking backward at the last quarter, a dollar forty two.

0:06:57.279 --> 0:06:59.560
<v Speaker 3>That's three cents better than what the street was expecting.

0:06:59.680 --> 0:07:02.960
<v Speaker 3>Fourth titter, justin net revenue was fourteen point zero five billion,

0:07:03.160 --> 0:07:05.680
<v Speaker 3>pretty much in line a little bit light a hair

0:07:05.800 --> 0:07:07.640
<v Speaker 3>light the yes have been on the street was fourteen

0:07:07.680 --> 0:07:08.760
<v Speaker 3>point zero eight billion.

0:07:08.960 --> 0:07:10.720
<v Speaker 2>Chris Miller. We're going to be speaking in just a

0:07:10.720 --> 0:07:14.200
<v Speaker 2>few minutes with kun John Sabani, Bloomberg Intelligence senior semiconductor analysts,

0:07:14.200 --> 0:07:17.600
<v Speaker 2>who's going to give us the details on this report.

0:07:17.640 --> 0:07:20.160
<v Speaker 2>But just remind everybody sort of in the chip ecosystem

0:07:20.240 --> 0:07:23.560
<v Speaker 2>and the world of chip ecosystems where broadcomplays.

0:07:24.320 --> 0:07:27.160
<v Speaker 1>Broad Comes one of the biggest designers of semiconductors. They

0:07:27.160 --> 0:07:30.200
<v Speaker 1>don't manufacture, but they do the design process, and they've

0:07:30.240 --> 0:07:34.840
<v Speaker 1>been focusing in particular on designing AI chips for other

0:07:35.000 --> 0:07:37.800
<v Speaker 1>companies some of the big hyperscalers that are investing in

0:07:37.840 --> 0:07:41.000
<v Speaker 1>their own ship designs. Those are often done in partnership

0:07:41.000 --> 0:07:43.600
<v Speaker 1>today with Broadcom.

0:07:43.280 --> 0:07:45.600
<v Speaker 2>So report such as this now shares up two percent

0:07:45.640 --> 0:07:48.960
<v Speaker 2>postmarket after results. The company sees first quarter revenue about

0:07:48.960 --> 0:07:52.160
<v Speaker 2>fourteen point six billion dollars. Looking across the ecosystem, who

0:07:52.200 --> 0:07:54.280
<v Speaker 2>does Broadcom have to be most worried about? Because we

0:07:54.320 --> 0:07:57.920
<v Speaker 2>got that report yesterday from the information about Apple working

0:07:57.960 --> 0:08:01.240
<v Speaker 2>with Broadcom, and we're reminded Ian King a little earlier

0:08:01.240 --> 0:08:03.560
<v Speaker 2>today that these two companies have sort of done this

0:08:03.680 --> 0:08:06.480
<v Speaker 2>dance of scorpions where they sting each other. But also

0:08:06.560 --> 0:08:09.480
<v Speaker 2>then well he was saying, it was like, what mating scorpions?

0:08:09.480 --> 0:08:09.840
<v Speaker 2>Is that right?

0:08:09.880 --> 0:08:14.360
<v Speaker 3>Yes, two scorpions mating. Yeah, we know it's not easy.

0:08:14.520 --> 0:08:15.840
<v Speaker 3>I want a little complicated.

0:08:15.880 --> 0:08:17.760
<v Speaker 2>I won't describe it, but let's just use the term

0:08:17.760 --> 0:08:19.200
<v Speaker 2>forrenemies instead perfect.

0:08:21.000 --> 0:08:24.400
<v Speaker 1>You know, I think that's that's a great analogy broad

0:08:24.400 --> 0:08:26.520
<v Speaker 1>companies is a really wide range of products, and so

0:08:26.560 --> 0:08:29.600
<v Speaker 1>they simultaneously cooperate and compete with a bunch of different

0:08:29.600 --> 0:08:33.360
<v Speaker 1>companies in the technology space. I think if you focus

0:08:33.400 --> 0:08:38.080
<v Speaker 1>on the AI portion of their portfolio, in some ways

0:08:38.120 --> 0:08:41.280
<v Speaker 1>they're competing with in Nvidio because a lot of the

0:08:41.320 --> 0:08:45.400
<v Speaker 1>in house designed processors that the hyperscalers are designing right

0:08:45.440 --> 0:08:49.560
<v Speaker 1>now are intended to be replacements for invidious products, at

0:08:49.640 --> 0:08:52.080
<v Speaker 1>least to a certain degree. And so I think investor

0:08:52.120 --> 0:08:54.840
<v Speaker 1>is going to look at this result and ask what

0:08:54.880 --> 0:08:57.920
<v Speaker 1>does this mean for the process of bringing more chip

0:08:57.920 --> 0:09:00.800
<v Speaker 1>design in house at some of the big hyph for scalers.

0:09:00.840 --> 0:09:02.280
<v Speaker 2>Hey, Chris, can I have to leave it there? I

0:09:02.280 --> 0:09:03.960
<v Speaker 2>always love it when you take the time to join us.

0:09:03.960 --> 0:09:07.560
<v Speaker 2>Do appreciate that very much. Chris Miller Associate professor of

0:09:07.640 --> 0:09:10.079
<v Speaker 2>International history at Tufts University. He's the author of the

0:09:10.080 --> 0:09:12.840
<v Speaker 2>New York Times best selling book Chip War, The Fight

0:09:13.120 --> 0:09:15.680
<v Speaker 2>for the World's most critical technology