The Spotify Effect

[Music] Hello and welcome to the March 2026 edition of the Brookside Research Podcast. I'm here with my co-host, Jonathan Bloom. Hello, Alex. Good to be talking with you again. Good to be talking with you as well. Today we're going to discuss a topic near and dear to my heart. We're calling it the Spotify effect. We are. But what we're really talking about today is the future of AI. In 20 short minutes, we're just going to knock it right off. We're going to solve it for the audience. That's it. This is all you need to know in the next 19 and a half minutes. Get it all done. Well, I think so many, many years ago back in the post-napture era, I guess. I was doing a lot of reporting for the New York Times and other outlets on the effect that Napster, he'll appear to appear network, was having on the music industry and the recording industry in specific. Essentially what happened pretty quickly was that people were sharing music online initially for free and without the permission of recording labels or artists. And it was a huge hit phenomenon. And people realized we don't need to buy records and CDs and cassettes or whatever other format. The recording industry was pushing on us. We can just listen to music online and we can share it with each other. And the RAA, the Recording Industry Association of America, tried to sue the technology out of business essentially. And out of the ashes of Napster and other peer-to-peer networks came Spotify, which was touted in its initial years as the solution to the problem away for people to listen to music online through a streaming service. But pay artists and pay recording companies for the privilege to do so. Well, here we are years later. Spotify is a popular, exciting technology for music lovers. But the value of the overall recording industry and music industry has declined precipitously. And Spotify itself. And Spotify itself. And Spotify itself, if you take a look at the stock price of Spotify these days, it's will off of its peak. And our contention in this interesting little podcast is that what's happened to Spotify will eventually happen to the AI. Boom! Beautifully said. That's exactly what's going to happen. Yep. And these are metaphors that Alec and I, ideas that Alec and I have been discussing. It's been decades now. I remember hearing this from you when we were at Fortune back in the, it's the 90s. No, early knots is when we first met, correct? I think so, yeah. And I, you know how it is that past, you never want to say for certain, but I have a very clear memory of you when we were on on 6th Avenue going. Just because this stuff was worth, you made the distinction between the music industry and the recording industry. And just because something was worth something in the past doesn't mean it's going to be worth something in the future. And I remember thinking, oh wow. And that's when I started going out and doing that hard research. Okay, where is that number? And that led me to the RIA database. And this really marvelous set of sales of music based formats by format and by year. And I'm just going to put you on the spot because I'm looking at my now on my side copy. Describe to everybody what this looks like. Because to me, this is the benchmark reference of how new ideas deploy and not just practically, but also financially. And what would be the word with value? What we all seek is where is the thing of going to be something of value tomorrow? Where is that going to come from? And I think this, the status that does a great job of that. So break that down for everybody. Yeah, I mean, if you look at a chart of music revenues over the past, I have three or four decades. 1977, forward is when they started. That's the A-track era. Right. Right. And it looks, it looks like a small mountain range with the peak being right around the year. 97, right, which is when they, which is when the RIAs, when they basically got together with Metallica. And they tried to pseudenaster, which is exactly what you talked about. Right. Then there's the other side of the peak. The right hand side of the chart. And incorporated in this chart is streaming revenues, which are a little tiny, little, I don't know what you would call it. Yellowy musteredy thing way on the right side that this thing starts in about 2017. A total revenue in the industry is about 8 billion. Now, what's great about this chart is you can see sales in nominal dollars. So when you look at it in today's dollars, overall streaming revenue is now up above $20 billion. But music is so old and it's so universal. And everyone has used it for so long. And that's what makes music data so great. Is that isn't just normalized for ideas and distribution. It normalizes for time. And if you take the music revenues and adjust them for inflation, what you see is, is music sales have never recovered. And it's absolutely true when you factor in live event sales. They've been the same number that they've always been now they're changing and the way the industry structures fundamentally different. There's a monopoly in our live nation, which basically is the promoter for the world. But the idea that when you digitize information, when you go from VATRAX, vinyl records and cassettes, compact disks, and you digitize it and put it on a network, it is fundamentally deflationary. It drives prices and value down. Now, that's not necessarily a bad thing. Not necessarily into the world. In fact, it is not the music industry is still going. I don't know. Strong is an interesting qualification. But it's very much there. Music is still very good. Gold is into last year's Grammy Awards. This is fabulous content. The Taylor Swift's to the world are the world's first music billionaire. There's still plenty to be money to be made. But this essential notion of deflationary pressure when you digitize assets that in my estimate is absolutely what's happening with AI. And all you have to do is listen to the software guys themselves because right around the same time in the early zeros, when revenues began to tank is when Mark Andreessen, the founder of the world's first web browser, and probably the world, Andreessen Harrow, which really was the first major tech of VC firm. Correct. I mean, they were in that early handful in terms of those assets. And he very famously says, this is when software is going to be, this software is going to eat the world. And all that's happening is AI is software is eating the world, but it's eating itself. And what it's consuming here is its own assets and its own deflationary pressure. And armed with this analogy and then spotifies actual performance over the last 20 years. A very clear model emerges that I help the core of my analysis, but I use it a lot in terms of where are we really going to go? Now that automation is going to come primarily in the software industry. And that's this fascinating connection that works. But from the moment that we're in right now, AI is king. Well, and the news hook here was Jack Dorsey, founder of Twitter, now X, right, had this big announcement. Oh, in his own shareholder, you should read this. We'll put it up on the site of how we justifies this. And the quote was nearly half of over the 10,000 people that we had. We're going to lay down to just under 6000. They're going to lay off just under 6000 people, which means over 4000 are people being asked to leave or entering into consultation. Which by the way is exactly how the music and music, when I worked at Viacom when I came up through the ranks at the MTV Networks, I was at VH1 when it was a fast growing network. All of us were contractors, even though we worked there nominally full time. They all paid us on a day rate. So if block, if Dorsey is going to be laying off this number of people, you know, the idea is that everybody in tech is going to do the same thing. And so this idea that automation is going to restructure the industry, I mean, it definitely has. Like if you're using Claude code, I don't use it. I use a thing called cursor, but I don't pretend to code anymore, man. This stuff makes me very nervous on a lot of levels. I have enough time. We're going to have an example today here of me tangling with Gemini to show you just how deflationary these A.M. models are for somebody to do with the oil industry. And that's an interesting thing. enough for me to do what I need to do. But no, you do not need to code anymore, particularly, and if you're going to work in any type of basic script that's well understood. I mean, these things are really good at making software. They just are. Well, yeah. So, let's talk about, let's complete the circle here. So we both looked at a video on YouTube by Rick Biotto, who was a well-known music YouTuber, talks about the industry, talks about music, really a fascinating guy, but he put out a video, I guess, earlier this week, about how he agreed with us for slightly different reasons that AI would eventually go the way of the music industry. Yeah, man. And tell them why that it was a hardware-based issue. Well, I thought he had a really fascinating example in which he sort of describes that you can already download a pretty robust AI program onto your computer. You can use it offline, and it's very robust. Absolutely. I've done it. Absolutely. Right. And it has many of the same qualities that AI has, but he believes that over time, as word gets out about that, this whole notion, and again, it's not just AI, it's everything from Facebook to every social media network, the idea that your data, your personal data, is the product, and that so everybody who's posting on Facebook is basically contributing to the revenues of Meta. Yep. For example, we'll go away because people won't want to do that. And he points out the fact that you just have to pay $5,000 an hour for access to really expensive audio equipment, and now you can do it in a computer that you pay $5,000 once for the same 5G's that you'd buy one day of studio time, you go buy a Mac and do it all yourself. Now, full disclosure, those models were great, but not in our business. We have to be right. So when someone brings us on, I have to check it on a live, fully-enabled global model, just to make sure that I'm not crazy, and even then, as you will see very soon, I'm not really sure I'm getting top quality information there either. And I'm beginning to have to step up my access. I'm going to probably have to start paying even more for these tools, just so that I can remove that variability from trying to dog down what's actually happening. I mean, if you're going to ask for a restaurant advice from a neighborhood, you can use an office shelf model off a login face and drop it, and it works because it's using public source map data. It's going to be fine. But the fact that we're talking about live news, still Google, and they still touch trillions of data points, and you've got to sense how it's reacting. Now, with that said, if I have a particularly sensitive question, which I do, I use an offshore model that I'm not going to mention. I go, I am very aware of things that I don't want it to know, and I take the steps in order in order to do that. So all of this information is not to say that AI is going away. No, but it shows that there's no point being afraid of it, and all you need is a metaphor to understand it. And if you want to see what's going to happen, go look at the enterprise valuation, the stock price of the last 20 years of Spotify. What you're going to see is the peak of enterprise value was about a year ago, and just like in the rest of the story telling business, it all caught up with what everybody has known all along is trying to provide a unique experience to that individual person. In other words, what Spotify attempts to do is a terrible business to be in because everybody is so different. Their tastes are so dramatic. You're losing the ability to actually reach the entire audience. And oh, by the way, the business that works pretty well, it has its own issues, is live nation because it is an agnostic concept promoter. It doesn't care if it's rap, if it's soul music, if it's Christian rock, live nation produces it. None of this market segmentation and supposed rules of who can do what, where, and where this music comes, doesn't apply for live nation. They put the show on. And as a business, it has been a better bet. Now, there are a major problem of live nation because of margins and terrible. And there's a huge cap ex cost and there's other issues that put it on live shows like when there's a global world event, and forces everybody to stay home, you know, when there's a pandemic. But the idea that finding this safe harbor, that's what you want to go to. And we won't be able to get to it this week, but we do believe we have an area that's going to offer just that type of moat that is, that could be immune to AI, but it's going to offer you relative protection going forward and we'll end with a particular news hook as to what that is and we'll talk about that next week. You know, just to wrap up, I think what's always been true about technology is technology blazes a path. It doesn't care what people want. Driven by what people want, but it doesn't care. You can't legislate against it. It's agnostic. It doesn't have morals. It doesn't have feelings. It's just technology and it will find the most desirable path forward in terms of, you know, the attention that seeks it. Right. And the term that I use that has helped me go through this is technology. Yes, but capability. It's just something it does. And so the hook that got us here is Jack Dorsey announced when it was going to lay off, you know, 6,000 people. If you look very carefully in its press releases, what it also announced, even when it was making this major retrenchment, was it was going to block its square, you know, it's a payment store, which I use in my business. And I don't know if you use it, but I use it for several things that it's actually terrific payment tool for a lot of reason. And be very expensive. You have to be very careful with it, but it does do some terrific things. They went out and announced that they were opening an office. They were expanding in the face of what was supposed to be a secular sea change in the strategy of the business. And you know, the answer to this because we've talked about it, but they were opening up a shop in Dublin, Ireland, a beachhead in the EU into European payments. And I believe he's right. I believe the future of not the place that's going to protect you from AI, but a place that you can operate in an age of machine automation has to do with one of these fundamental capabilities of being human being, which is in currency, which is using a proxy for moving around value. And I believe that the place to operate is where this value was invented and where they have the greatest history for solving these problems and the place that they're being driven to coming up with their own solution by geopolitical events. And I very strongly believe that European payments and European currency flows is going to be a safe haven going forward in the age of ever, ever devaluing automation. And we're going to explore though where we go and how we do this and certain opportunities that are there. And if all anybody out there and reaction, we'd love to hear from you all about this. If you just look at the performance of Spotify, that gives you the operations perspective. Look at the music revenue database. That'll give you the revenue perspective and look at how live nation has evolved as essentially the promoters monopoly. And you'll see that these are the themes that are going to drive an automation restructured vertical and payments is not, is by no means the only place you can go. There's plenty of upside lurking out there and there are certain tools you can use to make that work. Absolutely. And I think that understanding AI through this kind of framework creates new investment opportunities. Correct. I think it's important to realize that technology while it can blaze that path, it can also fall out of fashion and out of purpose as something else new comes along unlike human intelligence. I totally agree with this. And don't you get the feeling that we've seen this before? Yeah. Right? There's nothing new about this. I mean, I guess that's the thing that comes with age. But I remember like it was yesterday and I said, oh, it's the end of the world, the music industry. It's all going to fall apart and you were like, no, just because it was valued before or it doesn't mean it's going to be value in the future. And that's when I thought, okay, what's this really going to be worth? And my big moment in the sun was that those digital skeptic things that I worked for with Cramer down at the street. And that was a good run for me. Being dubious about how this stuff would never have occurred to me unless we thought about music in this way. Well, again, we're going to have another episode on this, the Spotify effect and beyond. To read stories discussed on this podcast, please subscribe to our newsletter or visit us at ProxideResearch.com. Also, we'd love your feedback. Click on contact us for information on how to call, email, or even set up a 30-minute console.