Are we in this dot-com era's bubble again? Is everything looking too crazy, too rich, or are things a bit more reasonable this time? Why we're worried about it now, though. What are the triggers? We've got these mega IPOs coming, each of which are just sort of eye-watering valuations.

The concentration risk. Already, we have a huge amount of concentration in the US market.

They didn't have a chance to figure out their business model.

But what about the failures? And I think what we should think about is what we can learn from those failures.

Amazon, eBay, real dot-com, dot-com era stories, just eye-wateringly high valuations, completely collapsed in the crash, and then figured out their business model, and then have obviously grown to be just these incredibly transformational companies.

Thank you for joining us for this week's episode of "What's the Big Deal?" where we take a look under the hood of major deals in the public and private markets, and we explore finance industry developments.

My name is Deborah Taylor, and I'm going to use my career in investment banking to bring insights to our discussion from a public markets perspective.

And I'm Graham Smith. I'll use my career in investment banking and private credit to bring the private market perspective here.

And Graham, my question for you, as always, is what's the big deal this week? Big deal is everyone's partying like it's 1999 again.

We've got these three mega IPOs that we've been talking about coming to market.

We've got Anthropic, we've got OpenAI, and we've got SpaceX. And we want to talk about are we in this dot-com era's bubble again? Is everything looking too crazy, too rich, or are things a bit more reasonable this time? Absolutely. So is 1999 calling? We're going to talk about why we're worried about it now, though.

What are the triggers? We're also going to talk about what actually happened in the dot-com era. We'll try and draw on our memories of what happened in the dot-com boom and bust. And then the big question, the bull and bear case for whether we are in bubble territory at the moment. And I think, Graham, maybe we're going to disagree slightly on this, but we'll see. But we'll draw on the data points that we think are really important in answering that question.

Yeah.

Fantastic. Well, let's start off with why we're discussing this now.

What are the triggers? What's happening in the market that makes us worried that we might be back in 1999? I'm really actually keen to get your take on this as the kind of public market expert here, really. But what everyone is kind of talking about in the news, basically saying we've got so much concentration, not only in tech stocks, we've got these mega IPOs coming, each of which are just sort of eye-watering valuations. It's going to make the public markets even more concentrated. One theme we started talking about the other week, which again I want to dig into a little bit, is when you have these big companies lists, you've got public managers, you've got pension managers, you've got mutual funds that effectively have to track the market.

So if you're in an index fund, the index has to, it's got to buy Anthropic, it's got to buy OpenAI, got to buy SpaceX.

And what does that mean from a regular Joe's perspective even, who's got a bunch of holdings in say index funds, if one or more of these IPOs doesn't perform? That's the thing I'm interested in- Mm ... digging into a little bit here.

Anything else on your list from that perspective? Yeah, no, I think just to kind of expand on some of your concerns.

Just to put it in context, the concentration risk.

Already, we have a huge amount of concentration in the US market.

The Magnificent Seven are the big seven tech stocks which have huge AI focus. They're about 35% of the S&P 500. And we're going to add to that.

Which seems insane.

Yeah, it seems insane. We're going to add to that with the big- Yeah ... three IPOs this year, SpaceX, OpenAI, Anthropic, which expected later this year. That potentially takes the concentration of the US market to these big tech stocks to nearly 50%.

And that's unheard of, even in the dot-com boom.

As we said, the AI IPOs which are coming our way, they are at quite frothy valuations.

Some of the numbers that we've heard make them seem very expensive, particularly compared to the rest of the market.

So that does call into question whether we're in a bubble territory when you hear these very high valuations. And then the third thing that you've mentioned is what is new this time is the fact that we have huge presence of index funds, funds which have to track the market, which potentially have to buy in when a large company IPOs and then joins the index.

So there's lots that's new and lots that potentially triggers concerns around a bubble in the market around AI. And we've already talked- Yeah ... in previous episodes around the enthusiasm for the market to jump on the AI theme, wherever it is in the value chain, be it hyperscalers, cloud service providers, chip companies, and potentially now the AI companies as well. So, so much stuff that we could talk about. We'll keep it concise, I think, in terms of where this might go. But yeah, this is definitely one to keep watching.

Agreed, and it feels like in some ways there are a lot of parallels between what happened in the late '90s, just in terms of, I mean, we're both getting up there now.

I was in high school in the dot-com bubble, but I suspect a lot of people listening here just hadn't even been born yet. And it's a little history lesson.

But I remember in those days, just through talking to friends at school, everyone's parent in the 1990s was a day trader because it seemed impossible to- Mm ... lose money. It's like, "Oh, are you investing in this and this?" My dad even had his whole little dashboard set up downstairs, and it felt like he made it his job for a good couple years. Because it felt like it was impossible to lose money.

There's just so mucheuphoria around anything technology, and if you had a tech platform, it was just destined to succeed.

And I feel like we're in a similar position with AI right now, just in terms of if you want to invest and make money, it's just so easy to get excited about the hyperscalers, about the actual AI companies themselves, about anyone in this entire ecosystem, and it feels like those for me are the investing parallels in terms of the themes that we're seeing right now.

And we'll talk a bit about the differences anyway, but that feels similar.

And Graham, so when we look back to the dot-com era, the boom and the bust, we know that there were survivors.

You've got Amazon, Google, still thriving.

But what about the failures? And I think what we should think about is what we can learn from those failures. So what do you know about the companies that didn't do well, the ones that did actually go bust or just struggled as a result of the dot-com bust? Yeah. So let me see. I have a little list because some of these companies I remember, and some of them I don't know of or forgot about because I was kind of a kid.

Let's see. A couple that are top of the list that I do remember. One was this company called Webvan.

Webvan was just grocery delivery, and obviously we have those services now. Companies have figured out the model, figured out how to make it work. I think what we're finding, just as a general theme with the dot-com bust, you had these companies just raise insane amounts of money, but with unproven business models. So pre-revenue, really had no kind of proof that their economic model actually made sense. So you had this company called Webvan.

They were doing, in essence, what-- God, the name is escaping me today.

Instacart does. Where you've got a-- I guess a little bit different in the sense that they actually owned a fleet of vehicles, and were arranging grocery delivery.

I'm pretty sure, I don't know Instacart well.

I think Instacart is a bit like the Uber model, where you're using kind of the gig economy. That really wasn't a thing in the '90s.

And they just over-expanded in terms of capacity, spent all their money on infrastructure, and the economics just didn't make sense.

Another big one was this company called pets.com.

Same kind of thing, right? You raise a bunch of money because you've got all this excitement about an online presence for pet food and pet products. And it turns out, actually, you're going to buy, like, a 50-pound bag of dog food. I'm talking pounds weight, not pounds- Mm-hmm ... the currency here. And of course, that's going to cost a ton to ship, so all your margin gets eroded through shipping and logistics costs.

And again, they never managed to make the model work.

And of course, we have companies that are doing this now.

We've got Chewy, we've got Petco, and all the other pet businesses that sell stuff online.

So I think a lot of what happened in the '90s were, you had companies that were a little bit too early, but they had raised so much money based on expectations that just never materialized because they didn't have a chance to figure out their business model.

And even the ones that survived, so Amazon, eBay, real dot-com era stories, just eye-wateringly high valuations, completely collapsed in the crash, and then figured out their business model, and then have obviously- Mm ... grown to be just these incredibly transformational companies.

So, we still had some real success stories, but what you had in general was just companies raising tons of money off the back of just nothing real, right? Just expectations- Yeah ... that there might be something just because you had an online presence.

And I remember from high school, people talking about companies being valued on a per-click basis. Like, "Oh, the home page had- That's right. Yeah ... like a million clicks. Each click is worth X." There's like a multiple of clicks.

So I think when we're inventing new multiples to value stuff off of that aren't related to anything fundamental, that's also when you can get into trouble.

It's such a good point, though, what you said before about the fact, though, that we have got some companies who are still around doing the same thing that these other companies that failed, that they tried to do.

So it wasn't that the actual underlying business idea was flawed.

It was that the execution wasn't quite there, that they hadn't quite worked out how to generate the returns you need in that internet era. So- Yeah ... I guess my very crude takeaway is that being first isn't necessarily going to guarantee that you're best.

And that- Yeah ... maybe that is a useful takeaway in this AI era, where we know that some are rushing to be the first. That doesn't guarantee success.

Yeah. Exactly.

Okay.

All right. So how do we feel about what's going on today? Because obviously some real parallels, I think, some major differences as well, which we can talk about.

But if we roll forward to 2026 and we think about the big IPOs coming down the pipe, how do you feel about this, Debs? Yeah. So, I think I actually feel quite sanguine about this.

I think the IPOs themselves present challenges, particularly to fund managers, which maybe we can come onto. But in terms of where the market is now, we are not in the same place that we were in 1999, and I think my go-to metric on this is the valuations. The valuations of the market at the moment compared to where we were in 1999, just as a useful reference point, P multiples are great for this.

Now, P multiples, the value of the shares relative to the forward earnings, that's usually the go-to number that we look at.

They do wax and wane through the cycle.

But if we look at the market today, the S&P 500 has a P multiple of about 23 times.The long-term average is about 18 times, so it is expensive, but not dramatically so.

Yeah.

In 1999, the Nasdaq, the forward P multiple for that index was 60 times. It was huge.

That's crazy. Yeah.

Yeah, exactly. And we are so far from that in terms of valuations. And also, the big companies, which are really focused on AI, they are generating a lot of profit and a lot of cash flow. So we're in a situation where the market is made up of these AI businesses, which a lot of them are actually generating profits, and that was not the case in the dot-com era.

We had a lot of companies, obviously, a lot of them were IPO-ing at the time, but they, as you said, they were pre-revenue or definitely pre-profit. So we're in a world where we have much more certainty about the businesses. They are more proven, at least in some of their other activities, and the valuations of the incumbents aren't as crazy as we saw in 1999. Now, that doesn't mean that the IPO company valuations are sensible.

We've heard some very sort of wacky numbers being bandied around.

But, in terms of the incumbents, I don't worry so much about the valuation side.

Right.

So I'm not too worried about, in general, a repeat of 1999. I think we have other risks that need to be addressed.

But what about you? What do you see as being a concern around the market today versus 1999? Yeah, look, I agree with you in terms of the fact the landscape is different. Because if we had 60 times forward earnings across the Nasdaq in '99, we're nowhere near that level today.

So let's assume we're in a world for a second where AI is overblown, and the hyperscalers and a lot of the people in the value chain that are supporting them are going to have some kind of correction.

At least we're not valuing the entire market on the same kind of crazy level we were in 1999, 2000. So that does feel a little bit different. I guess where I look at this is if you take a step back and look at what's happening just generally in-- We're really talking about AI, right? I know we keep talking about it, but it's important.

It's driving a huge amount of value.

It's just consuming so many resources right now.

If you look at what's happening just in general, we talk about these companies are different. Yes, they have earnings, they have revenue.

Say they have cash flow, do they really have cash flow? They're burning through cash like there's no tomorrow, right? Hmm.

The CapEx bill for these data centers is just insane. I don't have a view yet on what the return on CapEx on this spend is yet. It's going to be interesting, I think, when we actually get the real information through these IPO prospectuses to see some of maybe what that looks like.

I think the other thing that gives me a little bit of concern is just the kind of insane pace of revenue growth.

And we talked a little bit the other week on run rate revenue, the fact that it's not like run rate revenue is a GAAP accounting concept.

There's so much scope for managing that number.

And we were talking even before we got on the call, I don't remember what the numbers were, but Anthropic's run rate revenue has gone from some number that was five times less than what it is now, just a couple weeks ago, a couple months ago.

So you're seeing this crazy growth, and it makes me wonder, how much can we buy into these revenue numbers? And when you look at the full kind of picture here, and it really comes down to cash.

Talking about discounted cash flows last week.

Can't really do a DCF on a company where cash flow is negative.

At what point are we going to start seeing a return on this CapEx spend and these companies actually generating some real cash flow? Because you need to get there in the medium term in order to make all this work. Otherwise, we're just spending on this premise that AI's going to change the world, it's going to come, it's going to be amazing.

Right now, that still feels kind of unproven.

Yes, although I guess part of me feels that there are different parts in the value chain that you can be exposed to. And if you are investing in the infrastructure or even in the cloud providers or the chip providers, as long as AI does sort of have some benefits and it's not fully priced in, there is still opportunity there. I think my concern is the IPO businesses, because for them, they are the kind of the new kids on the block.

They're the ones that have to prove their business model.

They are the ones that potentially someone else could come along and completely disrupt them by having an alternative large language model that completely displaces them. How do you feel about that, Graham? I agree. I think the only other thing I'd think about in terms of this whole kind of CapEx and spend cycle is, look at a company like Nvidia, right.

Nvidia's been around- Mm ... a long time.

They historically were in the PC gaming market, making GPUs for people who want to play computer games.

And all of a sudden, they're this transformational company, and they think they're running the world, and they're just incredible in the forefront of AI.

They're at the forefront of AI because everyone's spending a ton of money on it, and they happen to have the best product that was- Yeah ... the best positioned at that point in time, right? So I guess my thing is, if this CapEx spend starts to get cut back because ultimately people come to the view that we're just not going to see the return on it like we expect to, then what happens to literally all these other companies in the value chain? All as we talked about the other week, we've got the Nvidias, we've got the Broadcoms, we've got the Samsungs, we've got the Microns, anyone who is participating in this crazy build-out right now benefiting from all this CapEx. If that CapEx tap gets turned off, or at least turned downThen what does that look like? And I don't think this is like a 1990s everyone is going to take this just insanely huge bath.

But I feel like it's more broad than just relating to say, OpenAI, Anthropic.

And I think that's where we do agree, is actually the concern is if you have a miss on key data points, affects AI-focused companies because there is so much concentration in the big indexes to this kind of business. That is something that we haven't seen before.

That is completely new, and a massive- Yeah ... risk to the market. At the moment, we have the magnificent seven, I think I said at the beginning, at 35% of the S&P 500, and once we add those IPO businesses to the mix, we're at 50%. I mean, that- Yeah ... doesn't take much for the market to completely shift as a result of a small change in the information.

Yeah. And we talk about the market overall not being overvalued, at least compared to 1999, but look at the valuation metrics for these big three that are coming. I actually don't really have a read yet into either OpenAI or Anthropic because I don't know what their run rate revenue calculation is, so I think we need to park that for a little bit and kind of see- Yeah ... see what's gone into it. But SpaceX, we talked about, I don't know, couple weeks, couple of months ago, and from memory, just the valuation that's being talked about there, at least on a run rate revenue, whatever kind of multiple basis you want to talk about, is just nuts, for lack of a better word.

So that's for me, where I'm like, there does just seem like there's so much AI tech euphoria right now, everyone buying into Elon's BS, frankly, to some extent.

I'm not saying it's all BS, but is a big part of SpaceX just driven by Elon wanting to be a trillionaire? Like maybe.

And why are we supporting that? That's where when I put my fundamentals investor hat on, I just say, "Okay, this just feels kind of nuts." Yeah.

I'll go back to my mid-market private equity investing where we're buying businesses for 15 times EBITDA.

That feels better.

Yeah, I have a- actually tried to look up some of the IPO valuations rumors. And I think what I've heard is with, on average, across the three, we're talking about a 30 to 50% premium compared to what we see in the market, for the valuation of these companies. So they are quite toppy valuations. And yeah, I mean, you're buying into hope for growth with SpaceX, possibly Elon's vision, and his track record.

But yeah, at the moment, we don't have the profits, and definitely not the cash flows to back up those valuations.

There's a lot- Yeah ... of future narrative there that's baked in.

Okay, so the thing I'm wondering, Debs, is what's the next step in this whole IPO process for the three companies we've got? They're at kind of different stages.

We've got SpaceX that's actually filed an S1.

We've got OpenAI and Anthropic, where we've got some private valuations and some reported run rate revenues.

From my perspective, these all seem kind of silly. We've got SpaceX that's 2 trillion, sure, 1.8, 2 trillion off of $18.7 billion revenue, so 100 times revenue.

They've actually got EBITDA reported in their S1.

They're at 303 times EBITDA, the purported IPO valuation.

And then we've got OpenAI and Anthropic both in the kind of 20 to 35 times revenue valuation. Again, going back to my point, this is run rate revenue. We don't know how real it is yet.

So when do we start to get some real clarity on what's going on here? Because I feel like right now we're still in just everyone's excited, everyone's talking crazy numbers, but we don't have any real data yet.

Yeah, to be honest, for retail investors, we get information very late. It's not until the IPO book is being built that you really get a sense of what's the pricing range for the IPO.

You start to have an allotment that you can try and bid for as a retail investor, likewise institutional investors.

So at the moment, we know what's being discussed, but we don't actually know what it's going to price at until much later in the day. And that's for SpaceX.

For Anthropic and OpenAI, the numbers that you've quoted, great, they are valid data points, but they reflect what's been raised in private fundraising.

Yeah.

Once they are in the IPO process, they will think about an IPO valuation, and then they're going down the road that SpaceX are currently going down in terms of sounding out investors to see what they can value their business at in the public markets. And that's always going to be at a premium to what you have in the private markets.

So yeah, it's still going to be a bit of time before we get some hard data on the valuations. And I think what we are all waiting to hear when it comes to OpenAI and Anthropic is getting our hands on their S1, which is when you first get a look at their financials.

When we were discussing just before we jumped online about some of the numbers, we get revenue run rate numbers from these companies in their press releases, but we don't know anything about their profitability really.

So, yeah, it'll be exciting times to get some more detail that we can scrutinize, and hopefully we can cover that in a future episode as well.

So if you had to pick right now, Debs, based on the information we've got right now, like I got some spare cash, and I got to park it in one of three of these IPOs. Where are you putting your money? Oh, no, really? Ugh.Okay. Which would it be? Okay, I'm going to go with Anthropic.

I think my view on SpaceX was quite clear in the episode that we did previously on that.

I think Elon Musk has an amazing track record, but there is so much future growth that's needed to justify the $1.82 trillion valuation of that business.

Yeah.

OpenAI, I think is a great business, but I think Anthropic is slightly stealing their thunder, and it kind of harks back to what we talked about earlier about being first isn't necessarily best, and I think Anthropic have definitely learnt a little bit from OpenAI. How about you, though, Graham? Yeah. I'm actually-- We're trying to take opposing views, I'm actually going to go the same. And the reason for that at this stage right now is actually they just seem like the least crazy and the least wild.

Just going back to Elon, Elon is this kind of crazy cowboy. I agree with you.

He's delivered some amazing stuff.

I don't think anyone in the IPO is getting particularly good value opportunity here. And then just look at what everything's going on between Elon and Sam Altman. Sam Altman, his CFO, getting in a public debate or argument around what run rate earnings are, and that just doesn't feel like a place where you want to park a bunch of money. I could be wrong, but- Mm ... just on the basis of sanity, Anthropic seems like the most interesting one right now.

But yeah.

Yeah.

Obviously, always open to revisit that as we get more info.

Yeah. I have to say, I definitely don't envy the fund managers at the moment.

I think it's a really difficult thing for them because they have to be part of that decision making as to whether they want to put their clients' money on the table to invest in these IPOs. Some of them, they're working for funds where the asset manager is selling at the same time. So you might have one VC fund in the same firm that's exiting at the same time they're thinking about whether to buy in, so maybe that influences their decision.

Yeah.

But yeah, I think it's going to be a really challenging decision as to whether you participate or how much you place on the table for these IPOs.

$1.

I don't know.

$1. You're so brave, Graham.

And this is capturing the credit analyst in his element, the risk averse.

No kidding.

I would put in $10.

Okay, fine. I'll do 10 each. How about that? Yeah.

I go crazy.

Woo.

All right. Well, still more to come on this, no doubt.

Yeah.

We've talked about these themes so far, fairly quickly evolving, and I'm actually really interested to see some of the financial metrics that get reported once we get- Mm ... some real detail on both OpenAI and Anthropic.

I think things will start to look hopefully a bit clearer at that point, either to the positive or the negative. We shall see.

So I think that's it for this week. Thanks for listening, and I hope you've enjoyed our deep dive into whether or not we are going to experiencing a flashback to 1999 and the thrills and spills that we all enjoyed during that time.

So that's it for me. Thanks for listening. Hand over to Graham.

Thanks, Debs. I'm going to go listen to this episode on my Walkman and give it a listen before it gets uploaded to YouTube.

But, in the meantime, we'll see everyone next week