Chip Stocks Are On Fire — Will It Last?

Megan Rappino here.

“This week on a Touchmore, the beautiful game”

I am sitting down with fashion designer DJ

and the Washington spirit's first ever creative director

Domo Wells in her L.A. studio. We talk about Domo's unusual path towards sports fandom and how her work as a DJ makes her a better designer. And what designing for women's athletes and their fans should actually look like.

Check out the latest episode of a Touchmore wherever you get your podcasts and on YouTube. [MUSIC PLAYING] Welcome to Proftory Markets. I'm Ed Elson.

It is July 1st. Let's check in on yesterday's market vitals. The major indices climbed as stocks closed out their best quarter in six years.

The Dow finished its best first half since 2021.

And the Russell, 2000, Rappino. It's best first half since 1991.

“Meanwhile, Brent Crude was roughly flat on the day”

as investors awaited news from talks in Iran. The yield on tenure treasury's climbed as job openings data showed a stable labor market and finally Bitcoin dipped below $60,000 once again. OK, what else is happening?

Semiconductors just posted their best quarter ever. Philadelphia Stock Exchange, Semiconductor Index, rose 82% in the second quarter, and is up 94% so far this year. Western Digital is up 240% year to date. Micron is up 310%.

Sandisk is up more than 700%. But the rally hasn't been entirely smooth. Last week, Chip Stocks fell 8% in their worst week since April 2025. But that isn't saying that much. Still, the stunning run out and the turbulence along the way

leaves investors with one big question. And that is, how long can this semiconductor boom actually last? Joining us to help answer that question. We are speaking with Stacey Razzgon, senior analyst at Bernstein.

Stacey, thank you so much for joining us on the show. This has been just a crazy run up that not many people predicted. I mean, AI is winning, but not everyone in AI is winning. Big tech certainly isn't. But the semiconductor stocks all will get into how sustainable this

actually is, but first just your reflections on what's been a crazy

quarter. And I mean, Semicondu's been the primary beneficiary here. And as you said, the stocks indexes up almost 100% actually may even be 100% after today's close a year to date. And AI has just gotten so big.

It's dragging everything along in the space. You could have almost owned anything you would have been just fine. It's interesting though, some of the divergences we see this way. So that the traditional, sort of, like blind AI winners that you would think the Nvidia's and the broad comes of the world that are doing, you know,

the actual AI accelerators, they've actually had the worst performance. They're up, but they're up not anywhere near as much as the sector. And the reason is people have been playing the so-called bottlenecks. Again, as AI has grown, like, sort of one area of the space at a time followed by the next, you've been sort of hitting the limits of what they can

supply and then prices go up. And, you know, Semicondu's loved to play bottlenecks. And so the stocks have gone up. And, you know, we went from the accelerators to the memory to the semi-capped, to the optical, to the networking, to the power semis of the CPUs.

Now, now people are playing discreets and other things. It's really been kind of remarkable. Overall, though, you could have owned almost anything in the space. You would have been sitting pretty pretty. Yeah, Semiconduq's now make up a fifth of the S&P.

Is it that much? Wow. Which I just find stunning. 20% of the entire market. I mean, when you look at that divergence between sort of the obvious AI names and the

“less obvious AI names, and I think that is a pretty good distinction as to who's been”

winning in this market and who hasn't, at least in 2026. I mean, is that just investors speculating, having a lot of fun with these more obscure names? I mean, why are they so excited about a random name in a bottle necked-up sector, versus like, you know, Nvidia?

You know, it really has been about earnings. And for you know, even with the space up 100% year to date, you sort of like decompile it into the drivers, you know, probably 70% plus of that performance has actually the earnings growth. So, like, multiples in the space are up, and I won't say that like the space overall

is expensive, but it's not egregious. And by far, in a way, the earnings so far year to date have grown much more than the

And some of these bottlenecks, we've just seen massive revisions. I mean, take take memory, for example. I don't cover the memory space, by the way, to colleague of mine, but to talk about the industry rather than the stocks, I mean, we've just seen some phenomenal positive revisions in the earnings powers, you know, memory prices have just gone through the roof.

It's through the roof as supply has gotten really tight as as AI has grown. And it's just driving massive revisions in the earnings. We're seeing that among a number of the bottlenecks, I think for some of the other ones, there's the hope or the strong belief that numbers as strong as they are right now are just too low, given where an eye demand is growing.

And again, you can look at semi-cap or maybe some of the optical names or, you know, some of the other ones that are playing how right now. You can just look out at where people of Forecasts in demand to be. And you can see where the numbers are sitting right now. And it's clear that one of those things is wrong.

Like either the demand is not going to be, or numbers broadly, probably still need to come up and in many cases across the board.

“And that's why you've seen some of these other names respond.”

And so I don't know that it's necessarily unjustified. Again, I wonder why some of the other compute names haven't performed as well, because again, numbers, there I think are going up too, but I think maybe they're more well-known or more like heavily anticipated that you will see that kind of performance. And again, it was only so much money to go around, right?

Investors have to invest in something. And I think there's been a belief among the, the more traditional guys that, you know, they're there, they're safer. You can use them as a source of funds and some sense to play some of these more other like esoteric names.

Because you know, like numbers, at least for the big guys, you're going to go up anyways, but there's less risk from, from, from doing that. And so we've just, some of the fun shift in other areas, I think.

It seems like one of the most important questions for

Sundays and therefore for the entire market, because they make a preferred of the entire market, is will this last? Specifically, these incredible earnings, which are largely a result of huge demand leading to huge prices. And the question then becomes, is this a one-off?

Is this cyclical, which is a big question in the semis business, or has something fundamentally changed? And our earnings just going to skyrocket or at least be at this level forever. What do you think? Probably there's, there's some elements of truth from all of those things.

“Look, semis, I think, are cyclical and they always have been.”

And they probably always will, but there's a variety of different types of cycles and different durations. And like, you could argue that this cycle has duration. And let's take the memory space, for example. And that's probably one of the areas where we've seen the biggest price in price increases

as well, which is what's driving all this. But the reason is, as you said, demand is very strong and supply is very tight.

And, you know, this time is different, it's always sort of a dangerous statement, but there

are some things that are different. I'm taking, take the DRAM space, for example, like DRAM is, like in your PC, it's like the system memory that's running things as you're, as you're working. And AI uses a specific type of DRAM memory. They call it HBM, or it's called Salesforce, high bandwidth memory.

And it sits on the AI chips that are good sold. And to make say a gigabyte of high bandwidth memory takes four times as much capacity versus making a gigabyte of like the standard would be called DDR5 DRAM that would go into your PC or smartphone.

“So, you're to scenario where you could be adding, you know, wait for capacity to make DRAM”

for AI. And actually, even though you're adding a lot of wafer, you're not necessarily adding as many bits as you are nearly, would be because of this, this differential in how many wafer is an example. Played ratio right now, it's like four to one between AI, memory, and traditional memory

for DRAM. And so because of those dynamics, I mean, the other thing I should say is, it takes time to add supplies. And one of the issues right now is we're short on what's called clean rooms. So to add and supply, like there's a whole industry that's called semiconductor capital

equipment. They make the tools that make semiconductors. But before I can sell those tools and add capacity, I need to have somewhere to put them. You need what are called clean rooms.

These are the buildings, right, that the factories that make it the factories. And we're short clean rooms right now. So they have to build the clean rooms first. And then you can add the tools and add the capacity. And it just takes time.

I mean, this year for total semiconductor manufacturing equipment will probably do $145 billion,

which is up 20% year of year, so it's a strong year. But as strong as it is, it's a constrained year, the clean rooms don't come online until we're willing to next year and beyond, and the mean to have demand is still growing. And so it's entirely plausible that this cycle, this obstacle could last quite a while. Are there any names?

It sounds like you think that ultimately the growth is pretty justified given the fact that the earnings have been pretty staggering, so far. So far, again, I would say as long as AI demand continues, and I would say if AI demand does not continue, we're probably screwed. We'll put that aside.

All the signs right now seem to be pointing to continue to strong demand, at least for now.

It seems like the base question from investors, can you put that aside? Is it, I mean, do you build that into your model or do you ignore it on a sim that this continues?

“How do you even grapple with that very big question?”

I'll say the same thing I've said since this started. At some point, will you see an air pocket? I mean, presumably you will.

I mean, this is what always happens eventually.

On the, I can say it's not now. It's certainly not this year. Okay. It really does not look like it's next year. I'm 20, 28.

I don't know. You know, we'll have supply starting to come online in the bigger way in 28, and then it will be a question of demand. Why not this year or next year, because some investors are concerned. There's no way in incremental capacity.

Supplies tighten demand is very starting to that. That's not slowing down. You know, you probably don't have to worry. You get all until supply starts to come on like in a big way. And then you'll see, you know, the title go up, we'll see if everybody's naked or not, right?

And this actually happened, you know, it wasn't that long ago, a year or two years ago when the actual sell rate is the GPUs were a very tight supply.

And by the way, I would say semi-investors always worry because there's a phenomenon that

happens. In many cases, when supply is tightened, customers can't get what they want on the time front of the one is that they order more. It's called double ordering, right? So the question is always that when supplies to the right-tighten demand is strong and you're

“adding capacity, is that demand you're adding for real or is it phantom, right?”

This happened during COVID. You know, we had big shortages and lead times stretched out and, you know, they added a bunch of supply. And as it turns out, like it wasn't needed and it took three or four years to actually work off that oversupply.

On the other hand, you go back a year or two when the GPUs and the accelerators were in tight supply. And that supply came online in a bigger way. And actually, demand as the supply came on demand got stronger rather than the weaker. That demand was real, right?

So that will be the question. We probably won't know the answer for a couple of years. In the meantime, for the next year, certainly for this year, like I said, and almost certainly for next year, supplies going to remain tight and in that environment, you know, the people will act like the demand is going to be sustainable.

So far, though, the everything we've seen just points to everybody's short compute, that that seems to be the case. And again, it's not like people are resolving compute and just sitting on it. Like they're, they're reserving the compute and using it. The compute is getting used.

We can have a discussion on what's the ROI and return on that.

“I think that's a valid discussion that we can have.”

But the compute is getting used. And in fact, they want farm where they want to be able to utilize far more compute right now than they currently available. Well, the ROI discussion is very interesting. I assume maybe your view is that discussion won't happen properly.

It's already happening or at least for the among the people who matter, in this case, the hyperscaler. I mean, it sounds like your view is-- Oh, but it makes happen, I don't know, no, no, no, no, it's investors are worried, right? Because the hyperscalers are investing a lot of money and they've sort of reached the limit

of what their free cash flow can currently support. So there's, there, there's top in the debt market, there's starting to raise some equity and like that kind of stuff. They still have plenty of capacity, but there isn't worry that, oh, they're just investing. And people for some reason think that they're idiots.

So like, I actually don't know why. And these are some of these smartest and certainly most largest and most profitable companies in the history of mankind. They can see things that we cannot. And I presume that they are not idiots, like that they have line of site to return.

I think that they're already seeing returns on some of the things that they're doing. You can look at some of the other things. I mean, you can look at, for example, GPU rental prices. You have like companies that they're called Neoclouds. They're business models.

They have computers running it out and we're seeing those rental prices going up. We're seeing five-year agreements on that capacity that are coming to an end. And that existing capacity, which at this point is fully depreciated, is out getting rented out and even higher prices, right? So those returns look just fine.

You can look at other companies like Enthrop, for example, who actually like has a, an agentic, they do agentic coding, it's, they have a product called Clawd that people will live in a play for. They're revenues have gone vertical.

They're doing, I can't remember what the number is now, 62 billion dollars in annualized

revenue, something like that. And a month ago, it was like 44 billion, and a month before that, it was 30 billion, and in January, it was like 14 billion, and in December, it was 9 billion, and a year ago, it was a billion dollars. And the revenues have gone like this.

I think the question for the, for the ROI, people are concerned about the ROI though, is that all the, all the businesses that are using paying all that money for Clawd, are they seeing ROI on their investment into Clawd. It seems as though it goes one step further. Presumably they are, like I said, it's, it's the, they're more and more companies are

doing it. And again, it's hard, that's, this is the problem, it's hard to know on an individual basis of the end customer level, like what kind of a return they're seeing and what kind of return they're not. All I can tell is that the supply of that compute, like the, the, the, the, the, the, the, the

Anthropics of the world, the demands on their capacity that they are seeing where you have gone vertical.

Stacey Raskan, Senior Analyst at Bernstein Stacey, really appreciate your time. Thank you. Oh you bet my pleasure.

“After the break, calm cost is breaking up and for even more markets insights, you can subscribe”

to my weekly newsletter simply put at simply put.proftymedia.com. The average American spends $1,700 a year on clothes and a lot of us are spending way more than that. But this week on net worth and chill, I'm breaking down exactly how to stop overpaying for your wardrobe without sacrificing your style.

Because between the tracking pixels following you around the internet, the foam

moderops and the influencer halls, someone is always trying to get your money.

I'm sharing my best kept secrets for scoring designer pieces for less, why second hand shopping is having its biggest moment ever and the one rule I use before buying anything new. Plus, I'm answering your questions including how to dress well when you have zero dollars to spend, how to upgrade your wardrobe for as little as possible, and what signs actually give away that someone is truly wealthy.

Listen wherever you get your podcasts or watch on youtube.com/yourrichbf. We're back with profty markets. One of the world's largest media conglomerates is officially splitting up. On Monday, Comcast announced plans to complete a spin-off of its media properties. The newly formed company will include Universal's film and television studios, its

theme parks division, its broadcast networks, and its streaming business. Let's left behind, the broadband, the Wi-Fi and the cable connectivity businesses will continue to operate under the Comcast umbrella. The stock is up more than 5% on this announcement. So, to break down this decision from Comcast and why they're doing this, we are speaking

with Rohanga Swamy, business reporter at Summer 4, Rohan, good to see you. This is a development from a previous spin-off that we also saw. So it all gets a little confusing because last year they spun off, let's see MSNBC, the Gulf channel, all of these cable networks, which were under the company vessent, and

“now they're doing it again, but with more of that media assets, why are they doing this?”

What is going on here? The uncharitable view, which Comcast would disagree with is that they tried to take a baby step, get rid of those cable assets, and the market just didn't care. Because if you think about what the fast growing businesses were 15 years ago, they were cable assets, when Comcast picked up NBCU from General Electric back in, I want to say,

2014, 2015, for around $30 billion, it was by some accounts that deal of a lifetime.

It was one of the greatest seven ideals ever struck, because Comcast just knew how to run these businesses better than GE did, and for a long time they were the engine behind Comcast's sort of meteoric, certainly stock price growth. And as we know, that has stalled at UNI like to go on TV, but I don't think either of us would ever make our living going on TV, because it's a declining business.

It just no one watches TV and blurs.

“They tried to spin off first, that obviously worked for a bit, right?”

But now they've taken this much more dramatic step. And it's seen in a lot of quarters, if you look at the stock price performance of charter, right? Comcast chief broadband rival, it's seen in a lot of quarters as a prelude to more M&A, NBCU buying or being bought, Comcast similarly is widely expected to go after charter

communications, which itself is trying to buy or has bought Cox communications. We are in what John Waldron told us was sort of an era of end game consolidation or everyone is just buying everyone and an effort to get as big as possible before they're stopped by the next administration. Just looking at how these stocks have performed since the baby step spin off i.e.

Versant media, Comcast itself is down nearly 15 percent since that happened, and Versant

is down nearly 25 percent. Now my understanding was the idea is that when you're when you're sort of conglomeratized, you pay this or we kind of call the conglomerate tax, which is all of the sexier properties get lumped in with the unsexy properties, and if you can separate those out, then maybe you get a more attractive multiple on one of the other stocks.

The stocks have both gone down. So I guess my question is why do they think this will work, or are they just sort of crossing their fingers? It's a complicated question. So to go back to sort of the prototypical, the archetypal conglomerate was general electric.

And Jack Welch's argument, and Jeff Mell, his success was that the size and scope of these business creates a smoothing effect. If say your engine manufacturing business isn't doing well, increased ad sales from an NBC

what some called earnings manipulation, but generally was seen as a more consistent, reliable

producer of cash flow.

“But investors kind of realize in recent years, they can do that themselves.”

They can build their own conglomerates using index funds, using any number of ETFs, or they didn't need a corporate back office to be architecting it. That's generally the argument for why conglomerates have almost uniformly broken up. They're very few true historic conglomerates left. What's happened with concast is a little bit different.

So to rewind a clock, broadband was seen as sort of the fastest growing backbone of a lot of these companies. If you were called time Warner Cable was a massive deal for concast, concast itself, charter, cox, altis, all of these businesses took on a lot of debt to build out the fiber optics that actually power the internet that allows you and I to talk to each other from across

Manhattan.

And in doing so, they were making a bet that growth would continue to be sort of unflappable,

right? Everyone needs to be on the internet, more and more people want faster and faster internet. That was why you saw concast and charters share surge during the pandemic, is obviously more and more people flocked to upgrade their internet speeds to rely on the internet more. And then what changed was the mobile companies started to step in.

“So the AT&T is the Verizon's, but really the team mobiles, right?”

They really started to compete aggressively for home internet using these 5G networks. That, of course, are ubiquitous and we rely on for our phones. That led to slowing growth in what was the growth engine for concast after linear started to decline, the cable or the broadband business itself. And so earlier this year, charter and then concast both warned that they were seeing slowing

growth outright declines in customer acquisition and revenue in their core engine. And investors just freaked out, right? The stock re-rated, I want to say double-digit drop for charter and concast in the course of a week. And so you have two businesses here that are fundamentally challenged right now.

Obviously, of the studio business, which will be significantly smaller than what they paid for NBCU by any, by any metric, but you also have a really challenged broadband business. That is sort of fighting to keep and retain customers by any means possible. So yes, while there is sort of a conglomerate tax that you pay an unstructured conglomerate should theoretically lead to more, a better, multiple expansion.

What's happened here is you can't really put lipstick on a pig when the pig is drowning in mud. And that's both of these businesses here. It's a really interesting point. I mean, you've got broadband, which is, you're saying, is struggling itself.

You've got the cable channels, which are obviously struggling over a verse and you've got some of the more traditional media production studios like Universal Pictures, Dreamworks, etc. Which I don't think people are feeling very bullish on. I mean, I don't think anyone's very, unless you're David Allison, peacock.

Maybe, but we also know that it's still losing money itself. I mean, are there any assets in this business that Wall Street looks at and goes, I want a piece of it? Not the way that investors are looking at assets right now. We have gone from sort of a very conservative shepherd, your cash mentality coming out

of the tech boom into a, we are, once again, rewarding growth. And neither of these businesses are fast growing or predictably growing businesses. Now, Comcast for years has said that profitability for peacock is just around the corner. That, of course, has not materialized my cavernal, the, the, the, the, the now coceo of Comcast to elite NBC Universal, the spin off again reiterated that on the call with investors earlier

this week.

It's really hard to make a streamer profitable, basically nobody but Netflix has done it.

Not Disney, not HBO. It's hard. It's very hard to compete here. What is the saving grace for Comcast, however, is that they've got two folks who are stepping into respective leadership roles of these positions who know the business is really well.

On the one hand, you have, as I just said, my cavernal who's coceo of Comcast with Brian Roberts right now. This is a guy in inveterate bankers, spent a little bit of Carla, maybe a year, but a long-time banker at JPMorgan has been in and around Comcast for years, knows this business intimately well, not a Hollywood guy, but a great finance guy.

And then you have Comcast former CFO, Mike Angelakis, who left in 2015 to run a Comcast backed investment firm. He is coming back to run the broadband business. And for M&A practitioners and viewers, they will know him from the actual acquisition of NBCU. He was the guy, as Comcast CFO, who kept it all together.

He's widely respected. He's seen as a steady operator. And with Brian Roberts having his hand in both of these businesses, there is actually a decent amount of long-term hope for both of these assets in as much as you can have hope in either of these troubled sectors.

“What is the future of these conglomerates do you think?”

I think it was an interesting point that, you know, it used to be considered a good thing. Jack Welch kind of led the charge, and then investors woke up one day and said, "Hold on. I already own lots of different stocks that don't need the operators to diversify. I can diversify myself.

Yes, and no.

I mean, if you look at GE, right, broke up several years ago now, and has created staggering

“and multiples of value for shareholders, the three split companies are worth, I think, four”

or five times what GE was when they split. It's been remarkable. That being said, so that model of conglomerates, the industrial engine being the sort of the engine that allowed conglomerates is that's dead. But as my colleague Liz Hoffman and I have written about a lot, AI companies are conglomerates

by any other name. Google is a prototypical conglomerate. Amazon is a conglomerate. I mean, they make anything from doorbell cameras to Wi-Fi routers, their logistics provider, they're obviously a huge tech company.

These are conglomerates that have a new engine in fueling them. That engine is to a certain extent web hosting web services, but also AI now. And AI arguably, if you look at Nvidia, which has taken stakes and dozens of companies large, large stakes, a Google of Microsoft, all these companies are turning into holding vehicles.

What's allowing them to do this other than Google's case, which is still largely advertising, is being a hyperscaler, is being an AI computer provider. That allows these companies to sort of become the new conglomerates and allow them to battle in self-driving cars, or yes, run a media company. I mean, Google just cut a deal with A24.

Amazon obviously is getting into sports and dipping its tone news. There are new conglomerates around the corner, and investors just don't care. The expected upside from AI, which UI and many other people have questions about. But largely the street thinks is unlimited, allows them to sort of paper over the lower multiple businesses that are really a few basis points compared to the massive amounts of

money they're taking in. It does seem that this is just a natural part of the corporate life cycle. And I know you've written about this. You can glomerateize as you grow, and then suddenly you wake up one day and you realize things aren't the way they were.

And then you can glomerateize and, of course, this is the old adage of bundling and unbundling. The only two things that really happen in business.

“And it is sort of happening right before our eyes, I think you made a very good point.”

Brian Goswami is business reporter, Art Semfort, Ron. Thank you so much.

And always a pleasure things for having me.

The president has a new strategy to get gas prices to come down. Tell them to. Yes, earlier this week on truth social, Trump demanded gas retailers to quote, get their prices down immediately as they are, quote, too high, no shit. This is probably the lowest of the low when it comes to economic policy.

You break something, in this case, global oil supply. And that, instead of cleaning up your own mess, you go out and blame other people for doing it. Gas companies, but let's be very clear. The reason gas prices are high is because of Trump and the mess he made in Iran.

And despite his attempts to convince us that the problem is now solved via this memorandum

of understanding the reality on the ground is quite clear, it is still a fucking mess. The US and Iran both traded strikes at each other over the weekend. And if you're wondering if that memorandum of understanding still holds, well, look no further than the US and Iranian government, both of which have accused each other of violating the memorandum.

Now, supposedly these strikes have been halted and supposedly the straight is back in business, but a boy can only cry wolf so many times.

“The only way to bring inflation back down including gas prices is to actually resolve”

the conflict in Iran. Until you do that, prices won't fall and they certainly won't fall if you simply tell them to. That's it for today. This episode was produced by Claire Miller and Allison Weiss and engineered by Benjamin

Spencer. Our video editor is Brad Williams, our research team is Dan Chalarn, Isabella Kensel, Christian Adonicue and Nia Savario, and our social producer is Jake McPherson. Thank you for listening to Propheech Markets from Propheech Media. If you liked what you heard, give us a follow, I'm Allison, I'll see you tomorrow.