
Loading summary
A
For the past three years, IBKR individual clients averaged 24.3% annually, beating the S&P 500's 23.1%. Lower costs and 170 plus global markets matter. Interactive Brokers Member SIPC Visit ibkr.com performance welcome to Chit Chat Stocks, the podcast that helps you find your next great investment. I'm one of your hosts, Ryan Henderson, and I'm joined as always by the one and only Brett Schaefer. We this is our weekly Power Hour episode. We do these every Thursday at 5pm Eastern time unless we have any schedule changes, but usually it's 5pm eastern time on Thursdays live on YouTube if you want to ask us any questions. We've got a full slate today on these episodes we talk all things financial markets. Any headlines, any news. We have lots of news. We've got Claude's latest revenue numbers. We've got consulting industry in the gutters. We have some insider buys, a potentially very flawed compensation policy at a big tech company that you called out, which I thought was hilarious. We've got Delta earnings, we've got. I mean of course we have to. More AI news. And also I don't know if you saw markets today, but a little more AI wreckage for software companies as well.
B
I did see that, Ryan. Yeah. Welcome in everyone to the show. The, yeah, software stocks. I saw my, my old friend, my back and forth flame wix. They're down below $70 again. I mean as we. You own some software stuff. You probably see it when looking at your portfolio. What, what's happening? Is it just more product releases from AI startup?
A
Every cloud announcement seems to destroy billions in market cap for the software industry. Adobe is. I saw this today. They're in their largest drawdown ever. I believe the data is a little spotty going back to 86, which is when they went public.
B
Yeah. But at least still probably a hundred bag.
A
Yeah. It's actually kind of insane. What do you. So the stock is down 67%. It's trading at its cheapest valuation ever. I think shares are like flat for the last eight years. What do you think their total annual return has been over the last 10 years?
B
Total annual return 8%. 10%.
A
10%. Isn't that kind of staggering after a 70% drawdown that they've been flat for eight years?
B
Yeah, I spent a two good two years before that. That SAS transition really helped him. But now apparently if you're a SaaS company, you're absolutely dead. You're still a holder shareholder.
A
Yes, I am we can maybe talk about that a little bit. There's some interesting stuff. Monday.com has just been destroyed. They have, I think they traded on EV to free cash flow of five times. I mean there is, there's some hairy situations but cheap situations in the software space. But we also have OpenAI's user count which I think is interesting because this is a potential stagnation, which I'm curious how much that hurts IPO demand as well.
B
Yeah, well we're going to talk about a lot of that. I also have some small caps that we can look at. I know a lot of people are suggesting small caps we talk about. I actually asked, hey, what type? You know this is maybe our fault for targeting smaller cap community sometimes, but I asked in the substack chat which people should join. It's completely free link is in the show notes. Hey, what stocks would you like us to cover either from interviews or ourselves? And we got tossed a lot of random micro caps that no one's going to listen to. So we're going to talk about them in the small cap of the week. Maybe some interesting stuff there. But yeah, Ryan, I think we should just kick things off. What's, what's going on with your number one topic here?
A
So I was pretty blown away by this number, but CLAUDE or Anthropic had an announcement this week. I'll just read through a tweet from Anthropic. It says we've signed an agreement with Google and Broadcom for multiple gigawatts of next generation TPU capacity coming online starting in 2027 to train and serve frontier CLAUDE models. Our run rate revenue has surpassed $30 billion, up from $9 billion at the end of 2025. As demand for CLAUDE continues to accelerate, this partnership gives us the compute to keep pace. So just to reiterate that they went from $9 billion in ARR at the end of 2025 to 1/4 later, $30 billion. This, this has to be the fastest growing company of all time at this size.
B
Sure. Yes, definitely. We'll give two caveats. One, ARR is not necessarily ARR because if someone's using a lot of compute up front to train an application for themselves, that may not be. You understand what I'm saying? You're putting all your development resources into hey, let's build an internal AI tool for our enterprise. And then when you use it, it's not going to be as compute intensive. But I mean still, regardless, insanely impressive. There was the, I'm going to call it famous Interview. It's just one of the ones that Emoti went on earlier this year. We said, well, our trajectory has us going for 100 billion in ARR by the end of 2026. And he's like, I don't know if we're going to get there, but we got to prepare and kind of if it happens. And hey, they're on track.
A
So far it is. I've been using Claude a bit myself.
B
The got a. Got a nice company. Company. Company account that you could use. Yeah, nice, nice.
A
It is staggering how many this might be the biggest. Like platform shift. I don't know if that's the right term for businesses since the Internet. Like you see, you are seeing people seeing companies like building on cloud, which I know I'm late to the party with all this, but I just don't know if I've ever seen one single company take over the business world the way Claude has in the last two years.
B
Yeah, that's fair. I don't really know much about it from a use case perspective because I'm not in the corporate world. We'll look like seeing what people are doing. But I mean look at those numbers. It's impressive. We also have here information from Amazon shareholder letter. Maybe we can save that for the next topic. Did you have anything else or is this our Claude or is this our AI startup update for the week?
A
Google Cloud. I mean more.
B
What did they. Oh yeah. What do they say here? Something Google Cloud's doing well, sure.
A
Yeah. They signed this agreement to extend capu is with Google Cloud.
B
Okay, well that, that relates to Amazon's cloud business end. They released their annual letter today. As usual, extremely optimistic on growth and reinvestment Runway which I'm not sure shareholders care too much. Although the stock is up 5% I think about today. Either way, Jassy of course is writing bullish stuff. He has things on.
A
You mean Jassy, former high school soccer coach. That was interesting.
B
He's a big sports fan in Is he. He's more of an American football guy now but you know, he's an athlete.
A
He.
B
I don't know if that bodes well for the executive world, but I like it. I think it's funny for context.
A
He started the letter describing his past roles and how he got to Amazon and he mentioned he was a high school soccer coach for a while there. So that's the reference. But what was. Give us some highlights from the letter.
B
Okay, so first let's talk about commerce. I thought two interesting quotes before we get To AI for a larger discussion here, quote. In India, where we have more than 316 micro fulfillment centers, Amazon now orders are increasing 25% month over month with prime members tripling their shopping frequency once they start using it. So it sounds like they've hit some nice product market fit with rapid delivery in places like India. Here's another quote and this is I think applicable to North America. Since introducing perishables into same day Delivery in early 2025 perishables, sales have grown by over 40 times. Again, this is two years, right? No less than. Less than two years. Barely over a year 40x growth. That's kind of astonishing. And it now makes up nine of the top 10 most ordered items for same day delivery where they're available. I gotta say, someone not being in the United States, the biggest thing you miss is just anything you can get on Amazon if you want and like, oh, I might need this. All right, I'll order whatever it is. And I would recommend people to try Amazon Grocery because it's quite cheap and you can now get it within an hour or two delivered to your house.
A
And that's grocery specifically is one of the things I think Jassy's been the most optimistic about. I remember listening to, I don't know if it was an interview or a conference call, but at one point he's like, I think the world really underestimates the impact we can have in grocery. And it seems like he's investing there.
B
Sure, yeah. All right, now let's talk AI. There are AI revenue of $15 billion. How they define that? You know, you guys investigate yourselves. But the, they talked a lot about their computer chip business, their internal one that they're building. So for listeners, instead of a lot of the times them just buying third party chips from Nvidia, AMD or Intel, Amazon now, along with the other cloud providers like Ryan mentioned with the tpu. That's why I think this relates to that announcement from the first topic. They have their, they have a lot of names, Graviton, Trainium, a lot of names. But they have a big business now and, and they said if they estimated that they're going to sell their own inventory because they use it for their own cloud, so they're not necessarily selling it to third parties. I don't think they do that at all really. But they estimate that they were going to sell this to like a third party cloud provider, they would be doing 50 billion in revenue for the segment. I do think over the next five to ten years, Nvidia that's their number one worry is these internal builders. I mean, I think for sure that's revenue that's completely lost to another player. And you have. Not necessarily a commodity from tmc. Tsmc. Of course there's engineering, design going into these things, but they give these companies an opportunity to put, hey, we invest a lot of money, steal some talent. We can build something that's applicable for our cloud business and we don't have to sacrifice that huge cost of sales to Nvidia, Intel, AMD and others.
A
Yeah. The one pushback I'd have there is that every time big tech talks about the success of their chips business, they always preface it by saying, we still have it. We still have a great partnership with Nvidia, don't get me wrong. But we have a growing chips business. And it's like, you can tell they still need those GPUs.
B
Well, they want to make. Maybe they just want to make Nvidia happy. I guess. I wouldn't think they would. Yeah, yeah, I always read those. It's funny reading that because. Why do you need to mention it?
A
They don't want to piss him off.
B
That has to be it. Yeah.
A
Well, I mean, he mentioned it. Like Jassy said, a lot of the customers still want the GPU capacity from Nvidia specifically. There, there. Here was a quote I did want to share relating to what you just said. Jassy says, as an aside, two large AWS customers have already asked if they could buy all of our Graviton instance capacity in 2026. Graviton is our w adopted custom CPU chip. We can't agree to these requests given other customers needs, but it gives you an idea of the demand.
B
He's a pump artist now. Oh, is that what we're. Is that what we're doing? Of course there's a lot of demand. Yeah, there's hundreds of billions of capital flowing into the industry. Of course. Of course.
A
Make up my mind about Jassy. Like, I read his letters and I like him. I think he's directionally doing a lot of the right things, like strategically. But he does feel a little more promotional than Bezos was.
B
Oh, Bezos is promotional. Do you remember the 60 minutes he did the drone thing in 2013 or something like that. He's saying this is our future. Yeah, come on. He's a hype man.
A
Yeah.
B
Jesse is just not as articulate.
A
I have a hard time with like, the acquisitions and the investments like that, the OpenAI stuff. It just. It feels like a red Flag. And we're about to talk about the user stagnation, but I'll let you kind of finish up any, any other big standouts from the letter.
B
Well, I won't mention on Jassy, I agree. Like, he's not as great at describing the narrative of what the story the company wants to build. Bezos was great at that. Other executives are great at that. They kind of craft a narrative that is very easy to understand in a few sentences. But I will say the one time I met someone, Bellevue Municipal Golf Course. We're waiting on, I think it's the eighth hole. There's always a backup. There's like three people, three groups always backed up there. There was a guy there he worked for, for Amazon. He goes, andy Jassy is the smartest man I've ever met. So I was like, oh, that's the, that's, that's positive there. Now he's running the company.
A
Have you talked to him since? 5 years of flat returns.
B
Oh, this is. No, this is a random guy from. Oh, yeah, yeah. He also said advertising is the guy that said that advertising will be more profitable than us, which turned out potentially to be right. But let's, let's go on the next quote here. We're not investing approximately $200 billion in capex in 2026 on a hunch. The recent OpenAI commitment of over $100 billion is an example of this. But there are several other customer agreements completed and unannounced or deep in process of the AWS capex we expect to spend in 2026. Much of that will be monetized in 2027 and 2028. And we already have customer commitments for a substantial portion of it. Sure, that's good. Yeah, that's good. But I will say OpenAI better come with the money. That's kind of the push and pull here. Yeah, like AWS has good, nice few years ahead of it. But the funding better be there for these startups that are losing a lot of money. It all comes back at the end of the day to the AI startup ecosystem.
A
Yeah, I'm starting to wonder if the Amazon investment in OpenAI was sort of a strings attached deal where it's like, look, will invest if you turn around and plow it all back into a capex for Amazon or AWS. The or orders for AWS. I would love to see a tally on OpenAI's full cloud capacity commitments because it's gotta be in the ballpark. It's gotta be north of 500 billion
B
yeah, I think they're supposed to be spending. And this is the one that's really suspect. $50 billion of the Oracle every year starting next year.
A
But they already. Oh, okay. I was going to say you look at the Oracle remaining performance obligations, but they've made massive commitments now to Azure. If you go look at the backlog from Microsoft on the cloud side, they've now made a $100 billion commitment to AWS. I'm sure they've made some commitment to Google. I believe they use Google Cloud in some capacity, if I'm not mistaken. And then Oracle. So call it 300 billion in projected cloud commitments on. What are they doing? 20 billion in revenue, maybe 30. Outrageous.
B
Yeah. Yeah. Well, and they're also. We'll talk about this later. We don't want to do a full hour on OpenAI, but maybe I'll save this for the end. There was a great video that our friend Aria, who's been on the show before, talked about this new circular revenue deal and the investigation that this random YouTube channel did. Wall Street Millennial. I think it's quite promising. They have a good theory on what exactly is going on there. But let's talk something else for a little bit besides AI consulting, which might be getting killed by AI. Ryan, what made you want to do this topic? You research your investments, you analyze markets, you manage risk. But did you research your broker? For the past three years, IBKR individual clients averaged an annual return of 24.3% compared to 23.1% on the S&P 500. IBKR's lower trading cost, competitive rates, efficient execution, and access to more than 170 global markets helped investors keep more of what they earn and put more capital to work over time. The broker you choose matters. Interactive Brokers member sipc. If you care about performance, find out why the best informed investors choose interactive brokers@ibkr.com performance.
A
Yeah, the topic is, is consulting dead? So I'm working on a piece right now about the consulting industry. Part of it's trying to figure it out for myself kind of piece. But also the industry just seems to be in the gutter, especially as of late, like 2026. Especially like year to date. A lot of the stocks have performed really poorly. It's worth noting most of the large consulting companies are not public, so we don't have perfect visibility into their results. I'm sure there's anecdotes out there, but if you look at some of the big consulting or pseudo consulting businesses that are public, you got gartner who's in a 73% drawdown, Booz Allen Hamilton, 57% drawdown, Accenture, 54% drawdown. And then you have more niche or I believe sort of domain focused consulting businesses like Huron Consulting, Marsha McLennan, FTI Consulting, all down more than 20%. The obvious culprit here is AI. If you look at the revenue growth, Gartner's seen a big slowdown, but most of the other ones haven't seen huge slowdowns. So I guess, and I was trying to read up on what is the narrative, what's happening at the AI companies specifically what kind of changes are they making. From what I've understood so far, they are reducing the number of junior level employees at consulting firms or at least the number of new hires for junior level roles. So like recent MBA grads, that kind of stuff, the hiring's kind of dried up a bit because a lot of that work. I SAW estimates that 20% to 50% of junior level tasks are data retrieval, market research and slide creation. So that is probably the most automated part of AI or of consulting since the rise of AI. I mean whether it's slides going out and asking for data, getting like a quick, simple market research report, that is stuff that literally these junior level employees were just doing Google search after Google search to pull a lot of this stuff together. I think that will maybe not be obsolete but will be highly automated in the future. What they are hiring more of is apparently they're switching from these junior level employees that they're bringing on to more of these mid level technical or like ex tech employees or data analysts that are kind of senior or mid level and they have like real technical expertise. And maybe there's more to the consulting industry than this, but the way I see it is there are two reasons for consultants. There's one, you don't have the capacity to do something yourself, so you bring them on which, or you outsource it, which I think is ripe for disruption from AI. And then two, you're renting expertise basically. So for example, if you're switching systems, maybe you like you're switching to a big different CRM. Maybe you get like a HubSpot expert or whatever from a consultant or a salesforce expert. They help with migration, that kind of stuff, like very technical.
B
I have a third category, scapegoats. They can be the scapegoats for your product strategy. Compensation told us to do it, so you can't blame us.
A
I think compensation consultants are perfect examples of that. But I Think the like the domain expertise, renting, that will all. I think that'll still be relevant in an AI world. I am interested in a couple of these businesses. Accenture is probably at the top of the list. Gartner, I don't really know. But if you look at the backlog or the remaining performance obligations, people are still adding more orders from Accenture, more service. Would you. I mean these are generally asset light other than, you know, human beings. And I would guess that if a lot of the contracts stay in place, margins will go up as the junior level employees start using AI.
B
As they use AI. I think this is a chance you're picking up pennies in front of the steamroller. It's not something I go, well, I just want to marry this business for the next 10 to 15 years. Yeah. Of course you have to reevaluate your investment every quarter or year. Every week. I guess if you're an investor in OpenAI and they do something crazy, they seem cheap. If you have a good capital allocator at the helm, a management team you can trust, maybe. But I worry that these are just a factory of consultant mercenary robots. Sorry for any consultants listening, but they have a reputation for doing nothing for companies for a reason. For example, one of them goes to like a Pepsi. Oh, what are you going to do? Well, we'll split off a business. Yeah. Why? Well, we're going to get leaner. Okay. Like it's just stuff like that. It's not going to where. Where where the game changing return is going to be. But on the other hand, I pulled up on our friends fiscal AI right now. Gartner PE of 15, EBD EBIT of 12. Yeah, not bad.
A
The Magic Quadrant. Brett, people need to have the Gartner Magic Quadrant.
B
Where's chit chat stocks in the Magic Quadrant?
A
I. I assume if you pay Gardner enough, you could be wherever you want to be on the map.
B
If we, if we take our full budget, we can go to the Magic Quadrant. I don't even know what is it? What is the Magic Quadrant? I have no idea what it is.
A
Just like the. Well, I could totally be wrong, but it, I believe it's like they'll map competitive landscapes. So like, okay, they'll go industry specific. Here's where each competitor kind of fits on the magic Quadrant and based on certain like expertise or features that they provide. But I think. And then every company that ranks well on the Magic quadrant just turns around and like puts a press release about it. So I wonder if they just pay Gartner they're like, yo, we need a magic quadrant that makes us look good. How much, how much do you need for this?
B
Yeah. And the hype cycle, you know that. That is something, I think. What are we going to spend a bunch of money? Oh, they put us at this point of the hype cycle. All right, cool it.
A
Yeah, I think you could be right. I would like to get to the bottom of how much of paying consultants is slide decks, is data retrieval, is market research, as opposed to actually something a little more advanced. But anyway, I did want to share this little tidbit from this week. I thought you might like this. In case you didn't see it, Oscar Health CEO bought $12 million worth of shares in the open market.
B
Yeah, I did see this. Since Oscar Health is some sort of Twitter, not meme stock, but it's popular. Let's just say that. Yeah, I thought that was nice to see. The insurance companies, they trade. I mean, the sector trades together and if there's any sort of news, like for example, this week, Oscar Health stock was up. Yeah. Because of the Oscar or the insider buy. But there was a ruling that Medicare insurance payments for a certain category, I believe, Medicare Advantage, where instead of the payout bump for the insurers, I think instead of it going up 0.5% this year, it was going up like 2.8% and it's better than expected. So all the healthcare stocks or all the health insurance stocks ran up, but Oscar Health doesn't even play in the Medicare Advantage space. So the thing trades wildly and the CEO has said that, like, if we execute, our stock should be at 50. So I like that he's putting his money where his mouth is.
A
We got a comment here in the chat. Thoughts about WIX today?
B
You know what, we got an interview coming on this two weeks. Should be fun.
A
I think I'm going to take a flyer on it. Honestly, they bought a third of the company at 20% higher than today's prices. I mean, even if it does decline from here, say cash flow decreased every year from here for the next five years. If they just turn around and put all that into buybacks, I think you can make out a decent return.
B
Yeah, it's extremely pessimistic. I've covered them on emerging modes before. I will say tomorrow or today. As you're listening to the recording of your podcast players, I'll have a quarterly update out on Gamma.com group and I'm outlining. You know, there's been a lot of volatility lately. I'm going to be outlining some various trades I am making today. Well as we're recording it tomorrow, but as you're probably listening today. Yeah, it's getting cheaper. And wix, yeah. Everyone hates these software stocks but I don't think they're going out of business in three years. That's kind of what they're getting priced at with these 80% drawdowns.
A
Okay, let's do a little dumpster diving real quick. I've pulled this up. 10 worst performing stocks in the S&P 500 year to date. I'm gonna rattle them off. You tell me what you like. The worst performer, the trade desk. Second Worst performer, work day.
B
Third. Nope. No and no.
A
Third Co star group. Fourth Intuit, which there's a lot of interesting stuff going on around Intuit right now.
B
Perplexity is coming for them. Correct?
A
That that is. Yeah. Seems to be the narrative a bit. I do as a customer of TurboTax and really as a tax paying citizen of the United States, I would love there to be a cheaper solution to
B
paying taxes or have to pay to pay for taxes. Yeah, it's frustrating
A
and it does feel like that would be like the target of a lot of AI improvement. Like. Like I could see competition heating up eventually And Perplexity had a big announcement this week that they are trying or. I don't. I didn't actually use the tool but basically they launched a tax tool. Okay. 5th Robin Hood.
B
They're down this year. Really?
A
They're down 38%. The would you ever own Robin Hood? Or is there are there some stocks that are just.
B
Oh yeah, it's much worse than tobacco and fast food and defense contractors. No, I don't. Robin Hood. It's just they're a little casino like right dad on a casino. Yeah, it's a good business. Them and IBKR I think are going to take over the world investing.
A
They are pretty, pretty darn good businesses. And layering on the prediction markets. Even though I have some concerns about that. Like it's gonna be lucrative for him. Okay. A couple more. Aries management.
B
Don't know anything.
A
Don't know anything. Gartner. No App Lovin.
B
Which no, not a series company. I was. Michael Burry had a piece this week that was insanely detailed. He needs an editor on stock based compensation. There was a lot of good info. What I did like is that Mercado Libre kind of came out and ASML came out clean on the other side. And their SPC is actually understated. Maybe it's because they're foreign companies. But Applovin was in the category of I can, I can maybe pull it up and it's going to take me a minute. Same as Tesla, maybe Shopify. But it was where like your sbc, your true sbc. So not the SBC that stayed on gaap, but your true cost of equity issuance was like wiped out your entire net income of the last decade. So Applovin is. It's in a risky business. There's pure fraud allegations which seem potentially credible because of how they're describing, how they've grown so quickly. And if you comedy's named Apple Oven.
A
Yeah, I don't know if I can ever own a company with that name. But the. I saw someone say like every time they hear the name they think of McLovin.
B
I do. Oh, I 100 do.
A
If you've ever been served one of those mobile ads where the exit button is like, yeah, they just get. Not formatted properly.
B
It's fake CPC.
A
Yeah, 100%.
B
And they're training at. What is it? Let's pull up a real EV to sales. 22 times EV to sales. Get out of here.
A
And there are companies who I think genuinely treat in. They think investors are the enemy of employees. Like I, I think the ones that give like really issue just a ton of stock, treat it like being public is a nuisance. Like, like we got. We gotta prioritize employees over our owners because they're just, you know, they're just
B
merchants and just ignore them because they don't care about us, the minority shareholders. Why would we invest in anyone that doesn't care about us?
A
Yeah. Okay, last two. EPAM Systems.
B
What was it?
A
EPAM Systems. It's actually, I believe like a tech consulting business. And then Boston Scientific.
B
What is Boston Scientific?
A
I don't know. But you know who's pretty close to this list? Fico.
B
Nice. What are they at 80 times.
A
I would. Well, let's pull it up. I actually could see myself being an owner of FICO at some point.
B
I'm just going to spam my banking app and keep, keep getting my FICO score.
A
$1 multiple makes the most sense for FICO.
B
All right. Boston Scientific for anyone that is medical devices, it looks like. Yeah, yeah.
A
What's fiber EV to EBIT 28.
B
Oh, wow. Yeah, yeah. I was saying that as a joke, but it's cheap. What is the risk? Are they getting disrupted by AI or something else?
A
I believe the concern is that people are outraged about the price increases.
B
Oh, meaning like real Estate people.
A
Yeah. So I. Apparently FICO increased the cost of its scores by literally a thousand percent in like the span of a few years. So it frustrated some people. But it. It's still pretty low cost when you think about the it in context to like the transactions that it. That they enable.
B
So $10 in the mortgage.
A
Yeah. $10 for a FICO score to buy a home and to penta potentially lower costs on. On your interest rate. I think it's a pretty big deal. But yeah, people. People are upset about it. And then on top of it, I think there's probably some ties to just like housing in general as well would be my guess.
B
In terms of volume still frozen. Yeah.
A
But yeah, all right, that's the dumpster dive. Any of those that would be the highest on your interest list?
B
Honestly, Robinhood would be number one. The trade desk is, is ah, it was just such a. It was like the bell of the ball. I would just look at them to a 25% revenue growth trading at 40 times sales. Now the whole business model is in. I'd say question like why were we paying 40 times sales for that? And it's still not cheap.
A
I think some businesses are better off with people not knowing exactly what they do. And like Palantir and I think there was a lot of people that just knew digital ads for the trade desk but maybe didn't understand the competitive position. I'm one of those.
B
No, they're getting outside the walled gardens. Just visualize that in your head. That's their business.
A
It's a good story. We're the anti big tech, but big tech tends to win.
B
So meta builds a model that knows how your brain waves function when an ad gets targeted. Do you think you're going to compete with that? No, it's pretty dystopian and you should get off Instagram, but you're not going to compete with it.
A
Speaking of Meta, there were some interesting tweets that you shared this week that I thought were kind of hilarious. So I'm not sure who this is. Maybe it's a reporter, but this person says ranking engineers by token spend. And for anyone not living in the software or the tech world, lately there has been emphasis on how much token usage, how much money you're spending on tokens and AI to help you as an engineer as like a measurement of productivity, which obviously you can see some of the flaws in that. But she says ranking engineers by token spend is like me ranking my marketing team by who spent the most money. We may not have hit our KPIs. But Joe spent $200,000 on a branded blimp that only flies over his own house. And then someone, quote, tweeted it and says, plenty of my Meta friends told me folks have been building bots that just run in a loop, burning tokens as fast as they can due to this policy. It's an absolutely stupid policy and is similar to how Meta uses. I actually don't know what this stands for. Loc to measure engineering output. But yeah, apparently Meta has been measuring productivity by token burn, which.
B
Not great.
A
No, this is a true anecdote that is very concerning.
B
Yeah. And I think I got a lot of pushback for saying that. I don't believe they are good at capital allocation, which I think is quite obvious. But maybe people don't like that because they own the stock. People are saying, well, if you hate their capital allocation, you should short them. No, it's a phenomenal business. But I think their Stock could be 2 to 3x higher if they actually focused on not buying back stock at the peak in 2021, not over hiring like double the amount of engineers or for a decade, and then not spending money without any ROIC discipline. Like, for example, they came out with a model today, of course, everyone's hyping it up, It's a good model, but they've spent so much dang money. They had the $14 billion man leading the company, the AI division. So much other talent there. And then you have these stories of just unconstrained spending and they better gain market share or it's just going to be another money pit. Because I don't understand exactly what their full trajectory is. People have this vision of, oh, they're coming out with smart devices, they're going to have the AI layer, then they're going to have the social layer and it's all going to connect together. You're going to have this fused thing. I doubt people are going to use that, but they have no market share in general of these AI tools. Like who's using Llama? I don't, I think I, I think
A
I told you this, but I was getting beers with or getting drinks with friends and they were, they don't follow this. They're not chronically online like me, but they were like, what, what AI service do you use? And a couple people were like ChatGPT or Gemini or oh no, I use, yeah, I said I, I, I use Llama. And if for the few people that actually knew what Llama was, they thought that was hilarious. It is a joke of a. It's a joke in the AI space. Do you think there's any first mover advantages here?
B
What do you mean? Oh, you mean for OpenAI? Yeah, for sure. Alphabet. Yeah. The thing is, Meta has the footprint. They have 3 billion users. I made a facetious joke like, okay, well, they have this model who's going to use it and someone said 3 billion people are going to use it. First off, if a bunch of people in Asia start using it, which is half their user base, like they're going bankrupt because they have zero roic from extremely important people in Asia. Now if the models were good and they had a good go to market, people would start using it because they use Meta products every day. But so far you're seeing what not none of that traction. So I, I don't, I, I really think Alphabet is crushed it in AI and I feel like Meta is just aimless and is actually not developing stuff worthwhile. Like, look at Alphabet, Waymo crushing it. All the sort of stuff that DeepMind is doing, especially with healthcare, biotech research, plus many other things. And then you have Gemini actually not necessarily crushing it, but wiggling its way in there, growing impressively bundling across the entire Alphabet ecosystem where people are actually using it. And Meta is spending just as much money, if not more, and getting nowhere monetarily.
A
Did you listen to the interview with Sundar Pichai?
B
I did not. I can't listen to the stripe guys. It's too, oh, we're just Irish guys taking over Silicon Valley. We're so happy. I just, it's, it's too much.
A
They do get a little on my nerves sometimes. And they, they would. You could tell some people just are too in the Valley, like to just in the tech world, maybe like not seeing how normal people experience some of these technological advancements, but it was, I came away with it impressed by Sundar Pichai. He is like built by McKinsey. It feels like he was so resistant to saying yes, resistance to saying anything controversial. He's a politician, but I mean, I guess oftentimes that's kind of who you end up hiring.
B
Yeah, I'd rather have him. For a lot of times, I'd rather have him than Sergey Brin kind of being an Elon Musk light goofball.
A
I was impressed by the amount of detail he knew about all the aspects of his business, like all the different elements of Google. Obviously that's the role of the CEO, but for someone who's Google has like endless amounts of subsidiaries, he seemed to have a pretty good grasp on everything.
B
Okay, we have a question. Isn't Meta's new LLM proof that AI LLMs are a commodity? That you can just throw money at a model and be competitive doesn't exactly scream deep moat. No, that's not the moat. The moat is getting people to use your platform and stuck. Stick with it.
A
We.
B
We aren't leaving Google. The Google workspace slash, Gemini slash, what have you ecosystem. Right? We chose them. It's cheaper than Microsoft. We're not leaving. It doesn't matter if Meta's got the same product.
A
I. I think I. If I counted the amount of Google products I use on a daily basis, I would probably feel compelled to buy
B
more shares, buy more stock. I don't know. Just because you spend $100 billion to replicate someone, does that, that make it a commodity?
A
No, I mean there's habitual nature to AI use. I mean, that's. I think, I think about Gemini. Like, I go to Google first tab, enter. I'm in Gemini. Like, it's just habitual now. And I Gmail, workspace, YouTube, Drive, Docs, Gemini, Google search. I'm probably miss. I probably used Google cloud in some way without knowing it.
B
Sure, sure. Yeah. I think people get prisoner of the moment with this AI stuff. Like this week, Meta stocks up 8% because they came out with some model. Next two weeks, someone else is coming out with a model. That's not a thesis for the stock. Pepsi Prebiotic Cola in original and cherry
A
vanilla that Pepsi taste you love with
B
just 30 calories and no artificial sweeteners. Pepsi Prebiotic Cola, unbelievably Pepsi. This episode is brought to you by Indeed. Stop waiting around for the perfect candidate. Instead, use Indeed sponsored Jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate. C. According to Indeed data, Sponsored jobs have four times more applicants than non sponsored jobs. So go build your dream team today with Indeed. Get a $75 sponsored job credit@ Indeed.com podcast. Terms and conditions apply.
A
I, I know this is like I could have had this take six months ago, and I probably did. And I was probably wrong. I wonder when I look at like, some of these software companies, like Workday, like ServiceNow, like Monday.com, yes, in theory, you can replicate all this. I think people are underestimating how little most companies want to switch. Like, do you want to spend your day switching systems? Like, I, I am reluctant to switch off anything.
B
Yeah, we're not going to Vibe code a replacement to our recording software. It's not happening.
A
No.
B
We can probably do something if we worked hard enough.
A
But I don't want to say too many bad things about Riverside in case they're listening and they give us more technical difficulties.
B
But we're, we're. We're their custom. They have to please us.
A
We have had endless issues with them for a couple months now. I don't want to switch. It's just a pain. Like switching costs are generally quite high. Do we want to talk about Delta?
B
Yes. I have plenty of topics here. As we close things out here, Delta we'll go through quickly because I have some other things before we get the hour done. I like looking at them because they're always first for earnings. And it's kind of a great barometer on the consumer economy. Tons of revenue. Just pure almost consumer discretionary spending. And the revenue guide next quarter was pretty good. Low teens currently have low teens percent with 6 to 8% operating margin. And I also like to look at their spending and loyalty quotes. They're saying better growth from premium like usual quote loyalty and related expense increased 13% primarily driven by double digit growth in card spend and an expanding cardholder base. American Express remove who every quarter. I can't say it remuneration of over $2 billion grew 10% year over year. And apparently we found out. I found out that they have their own fuel refinery. These are some smart. They're definitely they the reputation as being the best and the smartest in the industry is showing up. I think I had no idea they had their own refinery which is fascinating but at the same time even with the refinery they're expecting $2 billion in higher fuel bill cost next quarter just because the price of oil went from 60 to 100. I think it's an example. You look at their stock chart. The company is good at good at their job. They're the best in the industry. They have a great ecosystem. But the business is just a bad one. Like the sector.
A
Airlines suck.
B
Yeah.
A
It's just such a tough business. Like even if you do everything right it's so competitive.
B
But I think more for me is macro. The consumer looks fine so far. We'll see what you know, Visa, MasterCard, American Express, others say but for one more quarter I'll say right now it looks like Q1 yet again. The recession, the consumer recession is not materializing. Even though we've seen bear porn around that for what, four years now.
A
Yeah, people still travel. We got a World cup coming up. I Don't know if that'll have any bearings.
B
Yeah, we'll see. There's a lot of hate around those ticket prices. But the. What was I going to say? Yeah, the what? The TSA thing. When people have three hour lines just. That goes to show how much people want to spend on travel. Three hours. You're going to go to that Sandals Resort. Lovely like that. People want to do it, they want to spend money on it.
A
If FIFA were a public company, would you own them?
B
No, no. I don't know how that's possible but their biggest segment line item, government bribes. This has been growing year over year.
A
They are, they're a monopoly.
B
Oh yeah. I didn't like it when they forced Nelson Mandela to like bribe them. Do you remember that story?
A
Yes.
B
I mean that's a sad one.
A
Government bribes across the board, Nelson Mandela,
B
come on, don't force him to do it.
A
Do we want to talk about Rocket Labs atm?
B
Yes, yes. This is a company I've been following. A lot of people are in love with this company. We've had interviews on them. Stocks up a ton. They did a huge raise of the. At the money offering which is just, I love that it goes shortens to ATM because you know, they're, they're going to the ATM to raise some money.
A
It's the perfect acronym.
B
They raised about a billion dollars for all intents and purposes. They use some financial engineering we don't need to talk about. But they had over a billion dollars in cash on the balance sheet last quarter. Quarter. I also saw that Peter Beck, the founder and current CEO, eliminate his RSU payment program, which was nice. Maybe not the whole thing but you know, it was a nice gesture to shareholders. Right now the stock trades at something like 40 times sales with slim margins. You know, this is rocket company. They're trying to become the second SpaceX. They're pretty. Doing a pretty good job about looks extremely overvalued right now. You know, if they execute on their product roadmap. But it looks like they're taking advantage of this high share price and I feel like if we get a huge fall fallout from this space stock bubble with the SpaceX IPO, they're setting themselves up to weather any storm with a hu. With a nice cushion on their balance sheet. Thought it was a good note. It's one I'm following. Maybe I would still be holding if I bought a long time ago, but this is one I'm trying to be very patient on because I think it has a promising business trajectory and they've executed brilliantly with the odds against them. So, yeah, pretty. Pretty interesting.
A
It's a difficult spot to be in where you're an ethical company. Let me double check and make sure Rocket Lab's valuation is still insane. It's not as insane as it was,
B
but it's still pretty insane.
A
It's insane if you're an ethical operator and you actually want to run the business for a long time. I feel like having an extreme overvaluation is an awkward place to be because the right thing to do is probably an at the money offering. But it does feel like you're diluting shareholders. I mean, in this case, I don't
B
know what you were.
A
You mentioned that he adjusted the compensation policy to forego some stuff. That's a nice kind of gift, I guess, to shareholders, but I don't know. So often you see companies with unethical operators, especially during the SPAC boom, that were taking as much advantage of the at the money offerings as they possibly could.
B
Yeah, that's fair. That's fair. All right, Small cap of the week. You ready for a gym operator? The Planet Fitness of Latin America.
A
I've actually read up about them before. Do you have some anecdotal evidence for us?
B
I. I've used them. They're busy and it's easy to hop in and out so you can sign up pretty quickly. It's better because, I mean, they still do the gym. B.S. where. Oh, you're canceling after a month? Well, you actually have to pay for another month, but I expect that with any gym the company is Smart Fit. This is a Brazilian company. You can actually now buy them through interactive brokers. I would say this is another pitch for our sponsor here, ibkr, because they just announced access to the Brazilian exchange As of late 2025. Smart Fit is maybe the Planet Fitness of Latin America. Fairly close. Maybe not exactly like them, but fairly close. They added 384 clubs in the last year. They're at 2,000 locations across 16 countries. And Ryan, maybe while I'm talking, you can pull up a chart from our friends at Fiscal AI. Use our code. Fiscal A chitchat. Get 15% off any paid plan, 50% gross margin. If we go down the unit economics here. From what I can tell, places I've gone, they're packed. Almost too packed, honestly, in my opinion. Let's jump up the prices there, guys. So we get the riff raff out of there. 5.2 million total members expanding into high end clubs called Bioritmo, which maybe is good because These gyms are pretty like basic ones. So there could be a nice little niche in that. Although it's a very, very early days for that. I think they only have a couple offerings. They are doing different bundled subscriptions, traveler subscriptions. They're targeting corporate clients with these blanket passes, which I think could hopefully make them differentiated versus other players. I think gyms are an interesting area. It's not the best sector because you have some operating leverage that can disappear if you lose just like 10% of your members. There's a ton of operating leverage. You can work both ways. But as economies get richer in emerging market areas, more people are going to get freed up with time and money to go to gyms, exercise. And then you also have the weight loss drug tailwinds. That could be helpful as well as the company itself is a 24% revenue CAGR over the last decade, currently priced to sales of 1.6 PE of 18. What do you think, Ryan? What charts are you looking at?
A
I'm pulling up a couple different numbers. So EV to EBIT. I don't know if you just said this 14 times. Nice. Doesn't seem too expensive. The. Let's go annual revenue, gross profit and operating income all look phenomenal. Has this all been organic like? Or have they acquired gyms?
B
I have not gotten that deep in my research, but they seem to have a pretty simple model. Expand into cities, you build the bundled pass. There's a lot of different. It's a large population area and you can do fairly cheap. I mean, the prices are pretty cheap. There's a lot. They're not even in every country yet. And interestingly, their first non Latin American market is Morocco, Casablanca, which I thought was funny.
A
Revenue is up almost 50 fold over the last decade.
B
50 fold? Really?
A
Almost 42% CAGR.
B
I was doing USD. I would do USD.
A
Switch it here. All right, so 22 fold, I guess. Yeah, big difference. But still 30% plus revenue growth over the last decade and you've had consistent margins. It looks like operating incomes pretty solid. This is interesting. I like this and it fits within my theme of trying to geographically diversify outside of the United States. Not for any political reasons. I know some people seem to tie the two lately. Uh, but primarily because I've had other than software, I've had a hard time finding super appealing valuations for.
B
Yeah, I mean, S and p is at 30 times pe and the US dollar might be overvalued. Hey, geographical diversification doesn't seem like the worst idea.
A
And it is a. It's a model. The gym model, I think, is one that's easy to run at scale. Like, once you've built the blueprint for launching these new gyms, I mean, you've seen it with Planet Fitness. Obviously they do some scummy things, I would say, in terms of customer retention. But I mean, there are things you can do to help the downsides of the gym business. And by the downsides, I mean the fact that people are taking your. Taking your capex and throwing it on the ground, literally. They are depreciating your assets as hard as they can.
B
It's all right. Your best customers, though, are the ones that don't show up. Just get them on a decently priced plan. Yeah. Oh, no, you forgot about it for a year.
A
That sometimes. I know this is insane, but that is sometimes my motivation to go to the gym is not letting them get that, not letting me be that customer for them that they're the most profitable on.
B
We had someone here say they're Globo Gym, but I'm an average Joe's guy. I'll say.
A
Would you say it which.
B
It's more Global Gym. It's. It's. Well, it's neither. It's neither. It looks. It looks more corporatey commoditize D like a Globo Gym, if we're talking dodgeball design, but it's not as nice as these are pretty basic operations.
A
What. What's the cost of living like down there in Brazil?
B
Do you want to just like a comparison to somewhere? I don't know. It's a big country. It's kind of like saying, what's the cost of living in the United States? Well, you can go.
A
What about where you're at specifically? Does it feel cheaper than Seattle?
B
Yes. Oh, yeah. Yeah, for sure. And it's a nicer area, so it's definitely. Cost of living's good, but they're, they're, I guess mainly in Brazil. That's the biggest country, but they're in a lot of. A lot of different countries. 16. I. I think though there should be, as the economies grow, pricing power with these type of businesses. Like, again, like, I'm more spending power than some people, but. Or down here. But I look at that and I went, hey, whatever the price is, I'll sign up. Like, it's not that bad.
A
I've been revisiting our conversation from our last episode on rebuilding the portfolio from scratch. I'm going to do it officially. I'll announce this now. Big announcement. I will be taking a position in Grab Holdings. I'm going to do it. Nice it. I'll wait until we have our own little waiting period so we don't pump it. But I don't think I have a sway to pump Grab. But I just. I like the business. I really do. And you know what? I know you hate it, but I like Uber.
B
Yeah, I like Grab more, but Uber would be okay. Should be all right down here. I've been using them but they don't have Whammo yet. Yeah, but people like that episode we did from Wednesday. A lot of good feedback. Appreciate it guys. And if you haven't, I go check it out. It's a fun one. We kind of go quick hitters through 10 different stocks we'd buy that we've never owned. K Pop Demon Hunters, Haja Boy's Breakfast Meal and Hunt Tricks Meal have just dropped at McDonald's. They're calling this a battle for the fans. What do you say to the that roomie?
A
It's not a battle.
B
So glad the Saja boys could take breakfast and give our meal the rest of the day.
A
It is an honor to share.
B
No, it's our honor.
A
It is our larger honor.
B
No, really, stop. You can really feel the respect in this battle. Pick a meal to pick a side and participate in McDonald's while supplies last. I get so many headaches every month. It could be chronic migraine. Fifteen or more headache days a month, each lasting four hours or more. Botox Onobotulinum Toxin a prevents headaches in adults with chronic migraine. It's not for Those who have 14 or fewer headache days a month. Prescription Botox is injected by your doctor. Effects of Botox may spread hours to weeks after injection causing serious symptoms. Alert your doctor right away as Difficulty swallowing, speaking, breathing, eye problems or muscle weakness can be signs of a life threatening condition. Patients with these conditions before injection are at highest risk. Side effects may include allergic reactions, neck and injection site pain, fatigue and headaches. Allergic reactions can include rash, welts, asthma symptoms and dizziness. Don't receive Botox if there's a skin infection. Tell your doctor your medical history, muscle or nerve conditions including als, Lou Gehrig's disease, Myasthenia gravis or Lambert Eaton syndrome, and medications including botulinum toxins as these may increase the risk of serious side effects. Why wait? Ask your doctor, visit botoxchronicmigraine.com or call 1-844botox to learn more.
A
Yeah, I won't spoil them, but there were a couple on there that I, I was taking a look at my own portfolio thinking, why have I not swapped these out yet?
B
Yeah, that's a good point. All right, well, we had this circular re revenue thing that Aria shared with us that we haven't gotten to, and it is firepowered through this because it relates to the AI talk. So again, thank you to Aria. And go check out the video from Wall Street Millennial. So remember the 17 and a half percent promised returns from these private equity firms doing a joint venture with OpenAI.
A
Yes, of course, very easy. 17.5%.
B
So what's happening here is OpenAI formed a joint venture, they own 70% of it, 30% is owned by these private equity firms like Brookfield Asset Management. But these private equity firms are going to their portfolio companies and giving them, selling them tokens of ChatGPT credits, which is flowing again back, or you could say circularly into the joint venture which they own a part of. And OpenAI owns a part of. You kind of getting one. Maybe you can share this graphic that the video makers had because I think it kind of illustrates what's happening here. There's this Frontier product, which is enterprise Chat GPT that's trying to sell resell through to these portfolio companies of these PE companies. And what's weird is that again, these PE firms are selling software to their own company. This is selling to themselves. And it comes back to the promised 17 and a half minute percent return. Because interestingly, all of this, because OpenAI owns 70% of it, will be consolidated in their financial statements, which is going to pump up circular revenue that just gets recycled over and over. Maybe not over and over, but just gets recycled right. Coming into their ipo. And it's, it's just a perfect way to say, oh, actually, we're not discounting anything or what have you. We're selling you to this. And then we're just gonna, instead of selling it to you at a discount, we're gonna just sell it to you a full price. But then we're gonna do this profit share with our own customer, but it's gonna make our revenue look two to three times as big. OpenAI is a walking right flag. And I'm thinking we can officially call them AI we work because I'm not sure this makes it to, makes it to market.
A
I, I think there are a lot of deals like this out there where it's net profit zero for both companies where they just give each other revenue growth.
B
And I should have brought up the always sunny thing. Just, we're circling, we're circling. Wait, where's our money? Oh, we. We have not. We have the same amount.
A
Yeah. That's nothing. Yeah. I think there's a lot of deals like that. They just want the logo win. They want to be able to announce a press release, make it feel like you're getting some sort of a. Some momentum with it. But, yeah, it's. I am worried that this is more common with OpenAI than we already know.
B
Fair. Fair. I want to see the audited financial statements by a Big Four auditor.
A
I get. I know we've already talked about this a bunch. OpenAI gives me the ick. Anything associated with OpenAI and it's astounding because Chat CBT is still the leader, but I've just be. I've, like, discarded them, almost as if it's not going to work out for them no matter what, so.
B
It's a good point. It's a good point. Yeah. I'm taking the name AI WeWork. That's what I'm calling from now on. We're going long. We had a listener question around a desalinization company, Consolidated Water. That's a fun name. Sorry, someone. They had a nice message in the substack chat. We'll do it next time. We'll do it next time. Besides that, Ryan, anything else before we get out of here? We're a couple minutes long.
A
No, I do recommend, if you haven't already, listening to the Rebuilding Our Portfolios from Scratch episode. I thought that was one of the more fun ones we've done in a long time.
B
All right. Yeah, go listen to that episode. We have some fun interviews coming up. Hims and hers, maybe a little wicks, studying some super investors. Gonna be a lot of fun episodes. And of course, the weekly power hour as we plug through. And next week, we're gonna be starting earnings season, so it'll be a fun one. All right, as a disclosure, we are not financial advisors. Anything we say on the show is not formal advice or recommendation. Ryan, I are. Any podcast guests may hold securities discussed in this podcast, may have held them in the past, and may buy, sell, or hold them in the future. Thank you, everyone, for tuning in, and we'll see you next week. I finally had a light bulb moment
A
about a stock we've all heard about growing at 18 a year and a 15 pe. I shared this insight in a special deep dive report to subscribers of my research service, Value Spotlight. The report is called a Generational Moment, Reigniting human connections through a tangible network of intangible assets. Chitchat listeners can get a discount to my research at stockwriteup.com. that's stock W R I T E u p Com.
Episode: Amazon’s Annual Letter Insights; New Insider Buys; Claude’s Record Setting ARR Growth
Hosts: Ryan Henderson & Brett Schafer
Date: April 10, 2026
In this week’s Power Hour, Ryan and Brett deliver a lively rundown of dominant financial market stories. The episode largely centers on the explosive growth of AI companies (namely Anthropic/Claude), key takeaways from Amazon’s latest shareholder letter (including cloud, grocery, and AI segments), the ongoing software and consulting industry shakeups, major insider buys, and a dumpster-dive into some of 2026's worst-performing large-cap stocks. The hosts also discuss the questionable circular revenue practices in the AI industry, notably at OpenAI, and close with a look at a promising Latin American gym operator as their small cap of the week.
On Claude’s ARR:
"This has to be the fastest growing company of all time at this size."
— Ryan (04:37)
Amazon/AI Chips:
“Every time big tech talks about the success of their chips business... they always preface it... 'we still have a great partnership with Nvidia.'”
— Ryan (11:21)
On Consulting Automation:
“I saw estimates that 20% to 50% of junior level tasks are data retrieval, market research and slide creation. So that is probably the most automated part of AI or of consulting since the rise of AI.”
— Ryan (18:45)
Meta Token Spend Policy:
"Folks have been building bots that just run in a loop, burning tokens as fast as they can... It’s an absolutely stupid policy."
— Brett (quoting tweet, 37:34)
OpenAI’s Circular Revenue:
“OpenAI is a walking red flag. We can officially call them AI WeWork.”
— Brett (64:36)
Insight on Switching Cost and SaaS Valuations:
“People are underestimating how little most companies want to switch. Like, do you want to spend your day switching systems?... I am reluctant to switch off anything.”
— Ryan (45:31)
| Segment | Timestamp | |------------------------------------------------|------------| | AI/Claude ARR Growth | 04:11 | | Amazon Shareholder Letter Highlights | 07:19 | | Consulting Industry & AI Automation | 17:51 | | Oscar Health Insider Buy | 25:32 | | Software Stock Dumpster Dive | 27:57 | | Meta/AI Token Policy Critique | 36:05 | | Delta Earnings (Consumer Barometer) | 46:41 | | Rocket Lab ATM Capital Raise | 49:53 | | Small Cap: Smart Fit (LatAm Gym Chain) | 52:41 | | OpenAI Circular Revenue Practices | 62:15 |
The show features direct, irreverent, and at times cheeky banter between two sharp, analytical hosts. They combine skepticism of promotional hype with respect for sound capital allocation and business unit economics. The delivery is conversational, accessible, and peppered with industry in-jokes, making dense financial topics highly engaging.
Disclosure: Not financial advice. Hosts and guests may have holdings in discussed securities.