Yet Another Value Podcast - Byrne Hobart from The Diff on The Death of FANG+

Episode Date: August 20, 2020

Byrne Hobart from The Diff (https://diff.substack.com/) joins us to talk about his recent "The Death of Fang+" series. We dive deep into the question "Which of Amazon / Apple / Microsoft / Google / Netflix / Facebook" is most likely to see serious declines over the next ten years?

Hello and welcome to the yet another value podcast. I'm your host, Andrew Walker. And with me today, I'm excited to have the writer of the diff, the newsletter, and my friend, Bern Hobart. So, Byrne, how are you doing? I'm doing great. Great. Well, hey, let me start this podcast the same way I do every podcast, and that's by pitching you. And, you know, I've recently been reading the book Hamilton that obviously the place based on. And one of the things that jumps out at you when you read it is, you know, Hamilton, he's, he's reading military history and finance all this and he's just expert at one after other. And you're kind of like, oh my God, how can this guy be so good at everything? And every now and then when I read your newsletter, I kind of get the same feeling where I'm like, are you kidding me? Like this guy's an expert on privacy and economic history and all these companies. And it's just, I mean, it's so widely buried. And, you know, if you were only doing it once a week, I'd be like, oh my God, this newsletter covers a lot. But to do it every day is just so impressive. It's one of my favorite newsletters. So, That's my plug. Thank you for coming on. I'm really happy to have you. And I was hoping you could maybe dive in and tell me, did I sell you too hard? And can you give us some background of yourself and how you came to the due to the death? Yeah, sure. I mean, the Hamilton comparison is really flattering. I think there's a trick to sounding like you know a lot about everything, which is that if you know a lot about a handful of topics, you eventually find some commonalities between them. Like if you dive really deep into business history in one sector and, into military history and into, I don't know, the arts or something,

you'll find some of the same archetypes across them, like people who are really smart but don't get along with anyone, and that's why they're not successful, or people who are pretty good in their field, but actually really good at making friends. And, like, pretty good can vary a whole lot. I read this biography of Robert Oppenheimer a while back, and there was this quote from a physicist saying,

you know, Oppenheimer is not that great a physicist. He probably would, should not have ever, like no one would have thought he would win a Nobel Prize or something like that. Like it was like an incredibly high bar of like not one of the 100 greatest of the century, not that good. But, you know, for a guy who is not that good, still had a big impact because in Oppenheimer's case, he made a lot of friends and was able to connect people and organize people and things like that.

So like those architects show up in a lot of different domains. And you can do this trick where it looks like you found this really crazy unprecedented to comparison between, you know, how France reformed its military in this period versus how airlines restructured after oil prices went up. But it turns out you just knew those two topics really well. And so, of course, you found the way to connect them. It's kind of like baseball doesn't explain everything, but people who use baseball analogies can use them in every domain. So it's, if you didn't really care about baseball, you might assume that these people knew about everything. But if you knew that baseball analogies are a trip, then you'd be less impressed. So

that's my trick. That's my secret. There are giant, giant gaps in my knowledge. And then there are a few things I've just gotten really into over time. That's really interesting. I love the baseball because as you said it, the first thing is whenever you're doing anything, you can say, oh, we're an inning three or inning seven or he hit a grand slam. I'm like, it makes tons of sense. Obviously, those are simplified and comparing French military history to oil prices in the 1970s is a little tougher. makes sense. So that's how you became an expert at everything. But maybe tell us, you know, just a little bit about yourself, your background, how you came to write the diff. Yeah, yeah, sure. So I have a pretty eclectic background. I got into stocks really early in middle school, basically,

and spent way too much time reading about investing in high school and not as much time as I really should have doing homework. So I ended up graduating from high school and not really having my pick of colleges, which was disappointing, but I went to college for a year on a full scholarship at Arizona State University, and I kind of decided, like, I am learning, but I could be doing other things with my time, and I should try something else and see if it works better. My thinking was if I drop out of school after freshman year and spend a year doing something else and then come back, that I graduated in five years instead of four, and that doesn't have a big effect of my life. But if it turns out that there are better ways to spend my time,

then it's a really good move. So that's what I did. dropped out after freshman year, moved to New York, bounced around a couple jobs, and ended up working in online marketing. And while I was working on the marketing, I started writing blog posts about internet companies. So this was in 2010 and 2011 when there was this renaissance of online internet companies where they were, there were finally viable businesses that had been created online that could go public and not just coast on this late 90s hype cycle. So I started writing about those, and especially writing about how you could use online marketing, market intelligence tools to track them. So if there's a company that spends a lot on AdWords or that spends a lot on display ads or that gets a lot of organic traffic from Google, I found some ways to track those and see how they're performing over time and just get a deeper understanding of the business.

And through doing that and writing about it, I managed to get in touch with someone at SAC Capital who liked my writing. talked me for a while, decided this wave of internet IPOs is going to happen soon and it's going to keep going and that there weren't as many full-time internet analysts as there had been 10 years ago. So it was actually hard to find people who could get a handle on Facebook's business model or Twitter's business model or something like that. So I joined SAC in 2012 and was there for two and a half years, learned a whole lot, did a lot with tracking internet companies, tracking traditional media companies, doing stuff with data. This was pretty early in the alternative data era in finance.

But I got early exposure to a lot of that stuff. And then left them in 2014, bounced around a bit after that. So did some work in crypto at a company called 21, which was really fun. Then spend a while as more of a cell site analyst, so working at data firms, helping them write reports, contextualize data, and working with hedge funds to plug that data into their investing models and figure out what to trade based on that. So I was doing that for a while,

but there's this dynamic with this trend with a lot of professional money management where the area where they have an edge, it's A, the edge is diminishing. And B, the way that you actually get a decent risk-adjustive return is you have a super diversified portfolio, a lot of different portfolio managers who are really, really focused on one narrow thing. And they're generally supposed to not have a lot of market exposure to be long short and have minimal exposure to the overall market. And in some cases, this gets really

extreme. Like I talked to people who would trade cruise lines, as I analyzed cruise lines. There are three. And the mandate was tight enough that you were either long one and short the other two or long two and short the other one. There were just not a lot of ways to do anything really exciting if your mandate is have no net exposure to the industry and pick good stocks within it. So it seemed like the industry was getting more and more focused on pretty short-term developments and also on just not having, on being able to understand the cadence of data. So back in the day, you would, if you were a short-term investor, one of the things you'd do

is trade ahead of the border. So you think they're going to be, you buy, you think they're going to miss and sell, and let's say you made your money. But now the timeline has gotten shorter because there are so many intra-quartered data sets. So it's more like there are five different companies releasing different reports based on credit card data. We know roughly when each report comes out. We have the underlying credit card data ourselves. So we are not trying to predict how well this company, we're not trying to predict the value this company is a business. We're not trying to predict how well the company does this quarter. We are now trying to predict what change in investor

opinion will happen over the next week based on what data is being released. So that turns into this very adversarial game because you're making money by trading against people who have slightly worse data than you. And that tends to lead to this weird, weird situation where the stocks track the performance of the business really well throughout the quarter. And then when the quarterly report comes out, whatever it is, the data sets either don't track or completely got wrong, that's what drives the stock. So stocks have gotten a little bit less volatile intra-quarter, way more volatile at the quarter, and there's just a lot of economic pressure not to think about the next five to ten years. But that's the stuff I find really interesting.

So I decided to take a break and see if I could make a living writing. Did that in a couple different venues, did some consulting as well. And then earlier this year, I started a subscription newsletter on substack, the diff. And that's what going really well. And I think part of what happened is that there are a lot of people who will read a given article and find it interesting or not, but it's really hard to create your audience every time you publish something. There are people who can do it, and that's part of why people use things like Twitter and Facebook.

That's also why people work at mainstream outlets is those outlets create your audience for you or they help you create your audience. But with email newsletters, you can find people who are not just interested in one thing you said, but interested in the next thing that you look into. and who are interested in not just are you going to follow these stories everyone is talking about, but are you going to identify weird stories that are not getting as much attention or that matter for the narrative around large-cap tech stocks, but that are not usually thought of as relevant to large-cap tech stocks. So that's a lot of what I do is I try to look for the inflection points,

ways the future will be different 10 years from now that we can identify today. And then I try to look at how those are playing out. The other reason I call it the diff is that it's, So, Diff is a Unix command that just says, take two files, give me a list of all the ways they're different as compactly as possible. So what I'm trying to do is take the overall flow of news about finance, about the macro situation, about tech companies, et cetera, and just try to extract what's actually valuable or what I think everyone is getting wrong and just publish that. So it's not, I'm not trying to do all of the news.

There are a lot of people who do that, and so that's a very competitive area. trying to do what the main news sites will typically miss or the connections they won't make. That was great. That was a lot. I actually, every time you hit on a different subject, very similar to your newsletter, every time you hit on a different subject, I had a different question for you. But let me dive into what just for you, right? You publish the diff every day.

I think the insights are great and everything, but you published every day. Obviously, that's a conscious choice. Why go every day versus once a week, once a month? You know, if you're doing once a week, you could focus on specifically one story, dive deeper, make sure it's something you really want to be writing about where one of the issues with, you know, if you do something every day, some days there might just be nothing you want to write about. So why go daily versus weekly, monthly, yearly? Sure.

So I do have some longer term projects. Like I'm working on a book with a friend, and that is a multi-year project. We thought it would be a shorter project, but it's gotten more involved. Finance, Fantasy. So this one, semi-fantasy, I guess it's about the, so there's this book, The Rise and Fall of American Growth. It's all about how productivity growth was really high in the middle of the 20th century. And it looked like it was just getting higher and higher and higher. Every half century, something would change and the rate at which growth went, would accelerate.

And then things started slowing down around 1971, and no one really knows why. And part of this book is trying to look at some of the specific, moonshots that really drove a lot of growth. And one of them is a literal moonshot, the Apollo program. That's the chapter we're working on right now. So we're just trying to understand everything about the Apollo program that made it work, whether that's the political setup, like how did this get approved? Why did we decide to spend?

I think in present dollars, it's over $100, maybe $200 billion. Why do we do this for this sort of vanity project? Was it that little? It was, I think, $20 billion in dollars at the time. I could have this off. It was, anyway, it was a significant chunk of GDP. Maybe it speaks to how warped I am by like the current budget deficits and spending stuff, but you say $200 billion, I'm like, that's like less, that's one bailout package

right now. Well, yeah, there's also the fact that real GDP is a lot higher now. So it's always, it's always tricky to make these long-term historical comparisons because the farther back you go, the less you're actually comparing meaningfully similar quantities. Like, if you look at GDP in a subsistence economy, discretionary income is basically zero because everyone works really hard to generate enough calories not to start to death. And most of the time, that's what happens. And they don't start to death, and then occasionally they do.

So you can say the market value of wheat was X and this is how much wheat was produced in this country and therefore their GDP was why, but it's not GDP in the sense of you could actually shift it around. Like, you know, a defense budget means people are dying. So that makes the comparison hard. But then if you look today, one of the reasons our budgets are so high is that we have a lot more, a lot more services, especially health care and education. And we have a lot more people who are out of the workforce, mostly for demographic reasons. So a lot of our budget just goes to things like that.

and that's part of GDP, it's part of the federal budget, but it wasn't nearly as big a factor when life expectancy were shorter, and there was just less, there were fewer fancy ways to save people's lives at a great expense. Regardless, it was a huge expense. It's not the kind of expense we do anymore. It was a hugely risky project. We literally didn't know if it was physically possible. The Russians, Russia demonstrated in 1957 that you could launch a living thing into space And then in 1959, they demonstrated that they could actually get a living thing back from space. A little bit of a gap there. Apparently, there are still living descendants of the first dogs that Russia successfully sent in this space and then retrieved afterwards.

They're still around. That's a long, those are old dogs. Descendants. Yeah. Yeah. But even descendants, right? Because most dogs live about 10 to 15 years. These are like great, great, great grandchildren.

Okay. Okay. I thought you meant the actual child. No, not the actual kids. These dogs are. Let me ask you. So it's funny you mention Apollo because every time you mentioned Apollo, I love that you mention it because your insights have been so interesting there. It's actually one of the things you've mentioned it so many times. I've been really curious. I've had a blind spot to it. I'm thinking about figuring a book to read on it. If we announced tomorrow, hey, we're going to send someone to Mars in 10 years. Would that be of equivalent, greater or lesser difficulty than kind of Kennedy announcing he was going to send a man to the moon in 10 years in the early 60s? So my read on this is that it would be technically a lot easier, but politically and organizationally a lot harder. And we're just very averse to doing anything that is novel, has an uncertain, possibly positive, possibly negative outcome, and where people can die if you mess up, that's just not something. I think America at least is very much okay with people dying sort of accidentally as a side effect of a decision, but not okay.

with them dying as the direct effect of the decision. Do you think that's because politicians are better today at gaming out? Like with Corona, right? If you spent $5 billion at the beginning of this year on virus prevention, you may have lost your job, right? People would have been coming at you and saying you were a... Now, if you had spent that with the benefit of hindsight, people would have been like, you deserve all the awards.

Take all the awards, take everything, right? So do you think politicians are saying, hey, if I spend $200 billion going to Mars in 10 years, I might not even be office. I'm going to get blasted for wasting money. Or do you think it's just because the country is so politically kind of at each other's throats that we just can't get the consensus? Yeah. I think even going back to the 1960s, the political situation was pretty challenging for Apollo. In fact, it's one of those multiplicative things where there are a lot of things where if one thing had gone wrong, we wouldn't have gone to the moon or if one person had to participate in it would not have gone to the moon. And one of the people who was basically essential for us to get to the moon, was probably Lee Harvey Oswald, because once JFK had been assassinated, Apollo became this thing we were doing to honor someone in his memory. And especially popular in retrospect.

He was certainly popular as the president, but not as much more popular as a, as the late president. So that is part of why it happened was a lot of the American public wanted it to be done for that reason. LBJ wanted to be done for that reason LBJ also really wanted it done because by then a lot of the work was happening in Houston so it was to his benefit

for Houston to be this center for space exploration So yeah a lot of But I think the political challenges are just Much more dire right now It's a lot easier to create a viral narrative around something about how it's a waste of money

You can already imagine the tweets about how many people we could send to call how many people we could feed, how many, how much insulin we could buy? Well, not much, but how much insulin we could theoretically buy if prices were lower, et cetera. There's just a lot of a lot of ways to spend the money. I think people are a lot more aware of that. There's that ritual paradox where to tax people really heavily, you generally need to let them vote. But once they vote, they actually care about how the money is getting spent. So countries with a large tax base and a lot of electoral participation are just run very differently from countries that don't have that.

And when you look at a lot of the countries that have gone through a, have gone from poor to middle income and especially poor to rich, a lot of them have not especially democratic systems. Like even countries that are ostensibly democratic like Japan, if you look at how decisions were actually made in Japan, the legislature was pretty much rubber stamp. Like bureaucrats could introduce bills and legislators could introduce bills. This is one of the 50, 60s, 70s. The bills that legislators introduced were almost universally shot down. They would just get voted down by their legislators because it was usually, someone would say, well, in my hometown, we should build a big bridge. And then everyone would say, well, that's tax dollars from my hometown that don't go to my constituents, so no. But when the industrial planners would say, we've chosen the site for a new steel mill, and I think we should build it, that everyone votes yes.

So it was democratic in the sense that people voted. There were elections. There was somebody in charge. But it was non-democratic in the sense that the actual decisions were made in this very hierarchical form where there was a bureaucracy that was self-sustaining, super elite, had really rigorous testing standards, made people work extremely hard, had kind of an up or out approach to promoting people and this very, very linear promotion path. But once you got to the top, you were extremely well qualified, you knew everyone there was to know, and you had effectively unlimited power to rearrange the economy. So it seems like that that's just something you sort of need. And the fact that Apollo was partly a military project with a lot of military participation meant that it did have that more top-down approach,

where there was just not a lot of feedback from the electorate or a lot of the feedback was tacitly disregarded. You know, I want to dive in more here, but I also want to get into the death of fang. And I want to go back to one more question, more quick, the diff, one day versus one week versus once a month. Yes. All day every day. So I've found when I think about when I've been really productive,

it's usually the 24 hours before deadline. Like I've done a lot of projects at work and for fun where I give myself two months to do it and most of it still gets done in the last day. Sometimes I had this screenshot from a time when I initiated coverage on an industry and it was going to be done, I think, on a Tuesday morning. so I had to get everything done by Monday night.

And I took a screenshot of my inbox at the end of Monday because my first email was sent at like 407 in the morning and the last one was sent at like 11.30 at night. So I was really proud of that. But that's sort of how I end up working is that I'm very sensitive to deadlines, but only when they're very close. So I decided to just give myself more deadlines

spread out very closely. I've done some things, where I work on it for a while and then end up publishing later on. Like, I have things that are posts that are in the works where I actually have to do more background reading. And I'm also, a lot of the time I'm working on something in the background before I decide to write about it. Like I'm reading books on different topics or collecting new stories on a topic that I'm going to write about at some point. But since most of my write-ups are not that time sensitive, I can publish it pretty much whenever.

Yeah, yep, yep, yep. No, that's, and I love the point on deadlines, because like one of the struggles I've had with with COVID and working for home is like I like my structure right so I was wondering if having something every day like I come in for you if it was for you I come in every day and I start writing I know I need to get next tomorrow's newsletter out like if that structure was helpful because for me COVID it's been so hard like my structure was hey I wake up at six I leave the house at 645 I'm at the office by seven work you know work from seven to five seven to six take a break for lunch stuff come home get a little bit more work done not having that structure of getting out of the house has kind of made, I feel like I'm working as much, but I'm more scattershot

because I don't have that rigor. So, yeah, I was just wondering that. Anyway, anything else here, or are you ready to move on to death of fang? Yeah, let's do death of thing. Great. So this week, you've been, and I've been honored, it was inspired by a post of mine where I said, hey, you know, if you look over history, the market leaders in terms of market cat at the start of the decade, very rarely do many, any of them survive to be the market leaders at the end of the decade. And right now, kind of the fang stocks, Facebook, Amazon, Apple, you can throw Microsoft, Google, and Netflix in there. They just seem inevitable, right?

They're the biggest companies, the most profitable, most powerful fast is growing. They just seem inevitable. But history tells us it is likely that in 10 years, many of them will not be at the top of the charts. One of them will have spectacularly failing. It seems so difficult now. Anyway, I'll let you, I'll turn it over to you. I'll let you dive a little further into the series and how you wrote it.

Yeah, yeah. So the way that I thought about it was, as an investor, you're always being paid to take risk. And you can think of the equity risk premium as that payment. And there has to be some risk. Like, if there's no risk, it's a treasury. So part of the exercise was just asking what is the risk that you're actually getting paid to take when you buy Alphabet or Facebook or Netflix?

And you can give the obvious answer of some of these stocks look really expensive. And I didn't really touch that at all because, When I look at their business models, most of them have some form of recurring revenue and they have some way to invest OPEX and get revenue, get very high margin incremental revenue over a very long period. So when rates are really low, that kind of model is extremely valuable. And depending on what assumptions you make, you can easily mess up your discount of cash flow analysis and say these companies have an infinite value. like if you make the right assumption about what long-term inflation looks like, what long-term rates look like, and how much Netflix can increase prices over time, you can easily get to an argument that actually the net price of value of their cash flow is infinite.

So that's clearly not true. Clearly something can go wrong in the meantime. But I wanted to look at some of the explicit reasons that a particular company could do poorly. So for a lot of these write-ups, regulation is a factor, but it's too easy to do something. say, well, Facebook could be forced to divest everything they've ever bought and that they work less. It's more fun to go through the event path because that could have happened, but it didn't happen. So what will make it happen is, I think, the interesting question to ask. And look, let's stay on regulation for a second. So I think you and I emailed about these a little bit.

I was happy to, I was really thrilled to talk about it, but you obviously drove most everything. But regulation was the thing we kept coming back to for a lot of these guys, right? like what happens of regulation and you know I guess history I was telling you earlier to me history suggests that regulation is not what kills an industry and I think you had a different take on it like to me the last time something really got killed by a market leader getting killed by regulation it was very difficult for me to come up with the one but you had a couple of different ones I was hoping you could go through those maybe yeah happy too so I I suspect like I've seen this pattern where when there's a category of company

that becomes the largest by market cap or one of the most significant by market cap, there's usually some kind of reversal and often it is driven by policy. The more powerful an industry gets, the more it becomes a threat to the government's ability to do things. And governments are much more powerful than companies,

so they respond to those threats accordingly. And my thinking was, if you look back at 2007, the banks were becoming huge and financial revenue as a financial profits as a share of S&P profits were at historically high levels. And it's, you can't really point to any regulatory change that, that forced those companies to, to get smaller. There were always these incremental tweaks. You could say that some of the, the relaxing of mortgage standards created this race to the bottom and that since

mortgage availability affected housing prices and housing prices affected mortgage and fault rates, that it indirectly, indirectly regulation killed them, but that certainly wasn't the purpose of regulations that favored lower standards for mortgages. What I think you can say is that you can only have a huge financial sector in, you can only have a huge financial sector if the government is willing to backstop major banks, if the government is willing to do the very traditional central bank approach of when there's liquidity crisis, which we know there will be at some point, we will lend to everyone who needs it and we'll make sure that the system doesn't run out of cash, even though it might take an equity kit. And that didn't really happen in the

2008 crisis, especially part of what happened was just a lot of the people who had short-term dollar borrowings were not in the U.S. So it was just more technically and politically difficult to get dollars to where they needed to go. But also, the system had gotten really, really complicated to the point that it was it was kind of unclear what the magnitude was. And then once it was clear what the magnitude of the dollar shortage was, it was just politically really hard for the Fed to say, things are going really, really terribly. And everyone is suffering. And that's why we are giving hundreds of billions of dollars to big banks. And they still get politically punished for the fact that they did that. But it was, it was the right

decision as step one of dealing with the crisis where step two is make sure. sure that they don't get that levered ever again and make sure that they don't have as much much reliance on short-term wholesale funding and that to a large extent has happened like if you look at the the way banks have performed in this crisis they've done a great job and part of the reasons they have a whole lot more capital let me let me ask you a question on this crisis so I and I haven't thought deeply enough to this yet but you know in 2008 I you can remember like the the days where the house first voted down the bailout package and markets were down and obviously markets were markets drove a lot of what happened here, right? Where, hey, we're not going to shut the country down

or something. Stocks down 10%. Okay, we're going to shut the country down. But one of the things the Fed did here was they stepped in instantly, right? I mean, for a second, it looked like everything was going to freeze and the Fed said, you know, they just sprayed everything with liquidity. Do you think that was them learning lessons from 2008? Or was there just some different mix of kind of political power here that resulted in that? Why do you think the Fed was so much faster, so much more aggressive here? Yeah, I think that's exactly right. I think they learned a lot from 2008, and one of the things that they learned was, or maybe internalized, was when people start running out of money, there are a lot more companies going through liquidity crisis than you see in the headlines. And part of having a sufficiently

high risk tolerance such that you lever up way more than you should is you also have the risk tolerance necessary to just extend and pretend day by day and not admit to your creditors that you're totally insolvent. So when you see some liquidity crises, you know that it's getting worse. The other thing that they learned, one of the constraints in 2008 was that oil prices were so high that the Fed and ECB were worried about inflation up until the summer of 2008. They thought that was a bigger problem. So they felt like they had a lot less firepower. And then, of course, oil prices crashed right afterwards. And the oil supply situation is totally different now than it was then. So in some ways, in some ways, the Fed did a lot better. In some ways, the Fed just had fewer

constraints because of frackers. And those two things, the confluence of those two things just meant that the Fed was way more effective this round. It's just funny how everything like is somewhat loosely connected, right? Like who would have, who would have thought frackers had some incremental thing in helping the Fed bail out the COVID crisis? But something you mentioned when you were talking about the government, when a sector gets big enough, the government like starts shutting it down because they see a threat in it, right? Financial profits got too big. Financial people have too much power. The government maybe starts to crack down. I mean, you say that and I say, what has the most, what has the potential for the most power? And tech obviously has the most profits, but it's not just that. Like we've seen from

2016 and all across the world, tech can also literally influence the voters and literally change the government. So you say that and I say, oh my God, like just alarm bells are going off my head. Regulation is going to come for this harder, probably harder than anyone thinks at some point. Yeah, I think that's fair. Well, I, the tech companies are in a really tough situation because in some ways, they are just reflecting the way that human beings behave. And people spread rumors, people vote for dumb reasons, people try to influence other people to vote also for dumb reasons. The scope of Russia's political manipulation in 2016 is way more minimal, at least the stuff that happened specifically on Facebook, way more minimal than people typically

assume. Like, I think the ad budget was something like $100,000. This is really minor compared to the ad budgets of the Trump or Clinton campaigns. And they were also, they were not just trying to help Trump. They were also just trying to cause chaos in all sorts of ways. So they, they would do things like try to hold two rallies for very opposed groups. Black lives matter and black lives don't matter. Right. Yeah. Exactly. So, so they, They were certainly trying to cause trouble, but the extent to which they, like, my, my working theory is that they wanted Trump to lose by a small margin, because they knew that if Trump thought that he almost won and somehow got cheated out of it, that Hillary would have no legitimacy or far less legitimacy and far less political flexibility, I think that's what they were going for. I think they overshot. And I think a lot of, a lot of people just underestimated Trump and they underestimated the extent to which the polling data, missed people who were more likely to vote with Trump. And we still don't understand entirely how that happened.

I hear that, but I also think there's something with, and I don't want to get to into politics weeks here, but I think there's something with James Comey's letter. We're reopening the case, you know, what, six days before the election. Yeah, that was just wacky. I still don't understand what he was thinking. Yeah. It's just absolutely insane.

But anyway, let's see. So let's go back to, so you did it on Facebook, Amazon, Apple, Google. I think you had one on Netflix and Microsoft today, right? So Amazon's tomorrow. The last one, I remember. Yep. So that's six companies. What, when you were writing these, like, what was the company you were most surprised when you were, hey, like, I went into this thinking they were bulletproof, but oh, man, I see some risks here. Or conversely, hey, I went into this thinking, I see some risk here, but as I dive a little bit more into it, I think, I think this is a lot stronger than I thought. Yeah, Google was one where it was, you could

come up with ways that the business wouldn't do as well as it had done recently. But it was actually really hard to come up with things that would actively destroy Google other than just a really crazy aggressive antitrust thing. Yeah. European courts trying to tax, like trying to find Google for absurd amounts of money. Like that stuff could happen. But the actual core of the business is really good. And they've been able to adapt to a lot of the changes that were threatening them. Like, for a while, when people were bearish on mobile as an advertising medium, there was that concern that search is moving to mobile faster than Google is getting good at monetizing it. And Google is able to address that by just integrating the bidding and basically doing a lot of the optimization for the advertisers.

Google was also able to circumvent the problem of apps where one of the worries for a while was instead of doing a mobile search on Google and then going to a site that people would rather open an app and make a purchase, like open a wafer app and buy something. And now Google lets you link to individual pages in an app. So the ads can point to exactly where they should be. And since apps convert really well relative to mobile pages, that actually meant that it was often a bit more for those clicks. And of course, the value of someone who signs up,

who does a search, purchase a product through an app, and then is logged into that app and bother them with push notifications and that you have more data on them. That's pretty high as well. So a lot of the things that look like risks to Google also look like things that Google had dealt with. And the core search business, like typing text into a search bar and getting a list of links,

that's a business with really, really strong network effects because you want to, you want to continuously improve search results and the way to do that is looking at a lot of quickstream data and seeing what search results work at which ones don't. And then the other thing that they've been doing a little bit of over time, they've experimented with over time, is trying to build more vertical search where instead of sending you to a page,

they might send you to a tool that's Google controlled. And the way that I think about this is for something like hotel search where it used to be you Google hotels in NYC and you get to hotels.com and booking.com and Expedia and all that. And now you actually get a search interface and you can shop directly. And then that shopping is fulfilled by one of those online travel agencies. I think the way Google was thinking about that was we don't want to give any advertiser a conversion rate advantage.

We want to minimize a conversion rate advantage for advertisers because when the best advertiser has a really large advantage, they capture all the incremental value of bidding for the top click and paying whatever they have to pay to be number one and then getting all the incremental traffic from that. everything they do to improve their conversion rate also improves their ROI on those ads. And that means that the more they improve their search relative to Google search, the more they benefit from the Google search they're getting. So Google's very afraid of that.

And Google wants it to be a level playing field when booking.com and Expedia, you have phenomenal conversion optimization teams and phenomenal advertising teams. Like when they're competing with lesser, smaller OTAs, Google wants it to be a more fair fight because that's also. fight where those companies have to pay a larger portion of their margin to Google. Let me click on that real quick. So when you broke it down like that, obviously, I think most people know, hey, the reason Google is opening their own, hey, book through Google hotels instead of Expedia or something

is because they can capture a lot of money by doing that, right? Like if they offer it themselves. But originally when I heard of it, I kind of thought, this is a mistake. It all comes back to regulation, right? Like Yelp sued Google because Google, instead of showing Yelp at the top, Google started doing their own, and that's big bad Google beating up on Yelp, opening up hotels, opening up airlines, that's big bad Google. I don't think anybody would say little old booking.com or something, but you know, if you piss off booking.com, you piss off Yelp, you piss off enough for these people,

there's political enemies to be made, and regulation will come down in you. Do you think they thought that through properly, or do you think this increases their kind of regulation risk in the long term? I'm sure it incrementally increases their risk, and I'm sure that's part of what they discuss. And in fact, I think that's really the way to think about how these risks affect big tech companies is that it's not so much that they will be forced to undo something and they'll pay a big fine. It's that now they have more meetings where instead of talking about, is this product going to be profitable? They're talking about, is this product going to be legal? And is this product going to, if we launch one more vertical search, does this mean we don't just

lose that, but we lose every other vertical search product we launch? So it makes them more cautious. And that was part of what I talked about at the Microsoft write-up was just Microsoft spent a long time trying very hard not to be sued and as a secondary goal, trying to grow their business. Since a lot of these companies are growth companies, the multiple can re-rate pretty strongly if they go from a growth company that you can count on to keep launching new products and keep monetizing the usage it has to a company that you know what you're getting and it's not going to change that much. And so there's no optimistic outcome. It's just stable. It's like, for Google, it's actually kind of lumpy because they're a variable cost for all of their advertisers, and they're the easiest cost to turn off. So everyone can cut spending to zero. When you compare Google and Facebook with Google, since they're selling ads that are targeted to individual searches, when spending goes to zero, it's not like there are other bidders for that inventory, whereas with Facebook, and this was Ben Thompson's excellent point a couple weeks ago after the boycotts.

when a branded advertisers stop spending on Facebook, prices ticked down until the direct response advertisers can afford to bid on those page views. And since the ads aren't targeted to any one topic, they're just whatever you see when you're browsing Facebook, the ad inventory has a lot of potential bidders. And everyone in online gaming has built their entire business around what is the ROI at various price points for Facebook ads. And this is, Google saw that this in March and April, right? Like, if you were searching, if you were, they make tons of money on hotel betting, right?

March and April, travel goes completely away. And I think Expedia said, hey, we cut out a billion dollars of Google advertising, right? So absolutely. One of the things that struck me when you and I were emailing about this, like you say, hey, what can kill Google? And your first thought is, oh, search goes away, right? For some reason, search goes away. And then you say, oh, well, Google also owns YouTube and Google also owns Gmail and Google Docs and Waymo.

And we can debate the value of all of these. but there's no doubt that there's somewhere between some call option value to just tons of value in YouTube, right? So all these guys since have it. Amazon has Amazon and AWS. Apple's pretty much only Mac, but the one that strikes me is Netflix, right? Netflix is the only one that is really a pure, pure play bet on one thing. Netflix is a pure play bet on streaming. So when you wrote this up, where did you fall on the risk of Netflix going away?

Yeah, so Netflix was the one that I thought had the most risk. because streaming works as a product. It works really well. Netflix has a lot of great shows, but streaming also works as part of a bundle of other things. Amazon's doing that a lot. And AT&T is doing that as well, and Spectrum is doing stuff as well.

So you have a lot of different companies that are saying video content can help us reduce churn for something that has a much higher dollar value. And so even though our margin on video is not great, the actual dollar ROI on adding video to our bundle is really great. And it's still, Netflix still has differentiated stuff and they've really worked hard to build out a library that no one else can match in terms of just how many shows they have that are total cultural events. But as more companies try to turn, try to create a subscription product that includes

video but is not based on video, that means bidding up on the production cost for creating new video content. And there was a piece in the journal a pretty, like I think a couple weeks ago talking about some of the smarter real estate PE shops starting to buy film lots because they

in places like in California it's just hard to build new stuff. And so if there's higher demand for the existing lots, then that just means higher rent. So you could expect some inflation there. That market on the TV production side, my

understanding is that it's pretty heavily unionized. Like, you can't really hire, you can't hire a crew unless all of them are using, basically. So unions are pretty good at taking any kind of increase in margin at their employer and turning that into higher wages and fixed higher wages. So you would tend to have those expenses ratchet it up when times are good. They don't ratchet it down when times are bad. unions, they basically own a sort of convertible instrument on whatever company they work for. So when things go really well, they ask for profit sharing.

There were cases with the airlines where one of the unions was offered profit sharing. They said, this is stupid because airlines don't make money, just pay us higher wages. And then when I think this was American, when the airline started actually earning really high profits, the unions were like, you were the only major airline that doesn't offer profit sharing. Yeah, this is outrageous too. So it's a convertible, and it's really tough to run a profitable business when part of your cost structure owns either equity or debt depending on which one's better to own. So that just means that that means that Netflix would face some pressure on the revenue side where if everyone else, every other video subscription is free because you're already buying prime or you're already getting internet access and then Netflix is the one you pay extra for. That's a tougher sell.

and then they face the pressure on the cost side. You can also, we can certainly look at Disney doing its play in that area. Disney is very, very different from what some of the other streaming providers are doing just because they have unique ownership of some IP. They also, because they are a media conglomerate, they have a lot of different ways to monetize a given piece of intellectual property. So they can do streaming and not make an especially large amount of money on it

because it gives them a better negotiating standpoint with theaters. They could say, we could put this in theaters, but if you don't give us a better deal, we'll just give it to Disney Plus subscribers instead. And then for the content that in the back catalog that gets a lot of interaction, but that Disney didn't know was popular for whatever reason, then they can make more money on that with their parks.

Right now, all of that stuff is very tough to do and seems unexciting. But at some point, COVID, we will either have a vaccine or have sufficient behavioral changes that amount to a vaccine that keep R not below one, or we will just decide herd immunity is the best we can do. One of those things has to come true at some point in the next few years. Yeah, with Netflix, the one thing, so my last podcast with Marish Bele, there was a quote from the Uber CEO where he said,

hey, look, all of our competitors for a long time had basically a zero cost of capital. And when all of your competitors have a zero cost of capital, you know, it's tough to show off your network effects, right? Because everybody else can just, who cares if they can bid at a loss. And with Netflix, the thing I worry about is I have no doubt they have network effects, right? They have the best users, the best, they have the most users, the best data, all this type of stuff. I've no doubt about that. But my worry is Disney Plus can monetize in some other way. Apple can monetize in some other way.

So in many ways, their competitors have a zero cost of capital, right? Because Apple might say, hey, we can lose some money on this because it's going to help us make money on the phones and people spend more time on their phone. So the one thing I worry with Netflix is they've got network effects, no doubt about it. But if every competitive product has no cost of capital, in the long run, that can be pretty scary. Speaking of Apple, that's just the one, to me, they're the most exposed to regulation just because you see it with the Fortnite bad, the epic Fortnite battle right now with the App Store.

I could see a lot of different ways it goes, but you know, they've got everything locked in. If you want anything on iOS, anything on App Store, you have to go. through them. I think they're probably the most exposed to regulation. Maybe Amazon just because they're such a big beast, but I think probably Apple's the most exposed. Just maybe dive in for a second. Do you agree with me there? How you could see kind of Apple, the largest company of the world, getting killed? Yeah, sure. So Apple, I think, was the one, the one that is usually included in some variant on Fang that I did not end up writing up because I wanted to do five. It was pretty arbitrary. But, yeah, Apple is definitely exposed to regulation. They're exposed in the sense that

there's the whole antitrust risk around the app store. And that, we still have to develop some new theory of tech antitrust that can deal with this because they don't have a monopoly on smartphones, as they point out over and over and over again. Google does charge the same thing. So it's really not as if they're charging extortioner prices for something that they control. It's more like they're charging market prices for something that is really valuable. But maybe that's somewhat collusive or like indirect. the collusive that both parties chose the same cut.

Then they also have the regulatory risk of China risk. So a whole lot of their supply chain runs through Shenzhen, a whole lot of their revenue is China revenue. And that just gives them a lot of service area for getting hit by any kind of foreign policy of any sort. So whether it's China has started telling them they have to actually follow some of the rules that they've been skirting before. they have this special tariff zone, this bonded zone where they do a lot of their manufacturing,

so they get special tariff treatment in China. That could go away. Weechat gets banned and gets banned globally by the U.S. or if U.S. companies are not allowed to let users down with Rechat, then iPhones in China are worthless, or at least they're a way to signal I don't actually live here. So that takes away a lot of their revenue. So they have a lot of these random. and regulatory exposure.

And then on the App Store, maybe the better outcome for them is that they make some kind of concession to Epic and to publishers and to everybody else who is now saying that Apple takes a huge chunk of their revenue, doesn't provide nearly as much value, and then it's just not fair and not appropriate, and that they're bullies. Like, all of that is true. A lot of it is sort of something they backed into, like, early on. They just didn't think the App Store would be that big. They thought it would help them cell phones.

And as it grew, the pricing became more of a sticking point for participants. But there was no real catalyst until now for them to actually adjust that pricing. Now, I think that since so much value passes through the app store that they can probably find some way to capture similar amounts of value without just having a fixed cut. Like I think of cow bow and how listing is free, but they make money by ranking products. So they basically sell ads on what's otherwise a free peer-to-peer platform. And that model is perfect price discrimination. But it works really well for commodity products.

And as long as you want 5,000 of it, you could buy anything super cheap. It doesn't work super well for games or productivity apps or health apps or anything where you do actually have sensitivity to the brand. So they'll have to rearrange it somehow. But I think that some price discrimination would probably be. be useful for them and that they will eventually enough of these independent publishers or enough of these publishers who are actually part of large companies will get together and force them to come to some compromise.

So I think we've hit on a couple of them. So just so life down, so 10 years from now, you and I are doing a follow-up podcast. What company do you think we're most likely to be saying, yep, this company fell off the cliff? And it doesn't have to be, I want to be clear, like a lot of people say, say it's the fall. It doesn't have to be a complete zero, right? But down 50%, down 20% in the market. It's nowhere close to as important today in 10 years from now as it seems like it will be today. Yeah, I have to go with Netflix. The valuation, and I said I wouldn't talk about valuation, but I'm doing it anyway. Valuation just gives them fairly far to fall.

And there are just so many companies that are trying to do the same thing. That said, the more I dug into these companies, the more I realized they're all exceptionally, well-run, and all the managers are just deeply paranoid. I think the CEOs of every one of these companies sleep with a copy of only the paranoid survive under their pillow. They've just absorbed this idea that there are always inflection points, that technology changes really fast, and that you have to be five steps ahead of your competitors because they will be at least four steps ahead of you.

So they're all super aware of this stuff, and I think Netflix is the one where there's not a natural pivot. There's not a natural way for them to extend into a different market and counter. Like, they can't start selling a smartphone and then have Apple economics, like have Apple streaming economics. They can't open up hundreds of warehouses and have an offer Netflix prime that's Netflix plus groceries and electronics. They probably are not going to start an ISP, although they certainly have cheap capital,

and that's what you need. So maybe they'll do something like that or threaten to or joke about it down to office call. As a cable shareholder, I shudder to think that. No, just on the CEO, you know, it's interesting. you say because the one I think of, you know, I go through all these in my head and most of them are founder led, Netflix founder led, Amazon founder led, Apple is no longer founder led, but Tim Cook's great. The one that comes to mind is as such an interesting case study is Microsoft, right?

Because they had Bill Gates obviously took over the world. Bomber ran it for, what, 15 years. And not that he almost destroyed it, but I mean, the missed opportunity cost. I mean, the stock was up 10% the day he retired, which speaks to it. And then you have Satya coming in. And kind of out of nowhere, I don't think he was that well known. You know, I'm not super deeply plugged into the circles. But he's just the one that's most surprises me because he's not a founder, but I think he's proven, as you said, he digested the paranoid survive. And just for him to step into that role and be so successful, you know, it's really interesting. And then Microsoft Underbomber is interesting as the case study of, hey, these companies that seem bulletproof, they miss 10 years worth of trends because they've got a bad manager.

and that's what this is what ends up happening with them. Fortunately for all of them, I don't think any of them really have a bad manager at the top. Just behind Netflix, what would your second, what would your runner-up be for most likely 10 years from now? Oh, that's tough. I think you could say that Facebook or Apple would be the ones where I think the regulatory problems could hit them exceptionally hard.

Yeah. Like in Facebook's case, they're just such a natural scapegoat. every election, the loser will believe that the winner won because Facebook was manipulated. And it's a narrative. And then since both sides, it's never visible to you if people you disagree with are getting punished by Facebook, but it's very visible that people you agree with are. So one of the most bipartisan views on Facebook is Facebook is helping my political opponents get their views across and censored people I like.

So there's just a lot of a lot of willingness. We just have to find something they actually did wrong. And we have circumstantial evidence from Zuckerberg's IMs and his text messages when he's trying to buy Instagram. But a lot of that, you could read it in a really damning way. You could say he's trying to take out a competitor. You can also say Zuckerberg knows a lot about social media. He found a social media product that's growing that would be complementary to Facebook.

And so he wants to own it. And he wants to help that company grow because they don't have Facebook's back. end, but they do have, they do have a similar kind of product. Got a kid coming in here? Yeah, yeah. How old are your kids? I have a one-year-old, a two-year-old, and a four-year-old. Oh, man, I can't believe you found out.

You know, one of the questions for you was, how do you write so much having young kids? Yeah, that I mostly. my wife does a really good job, making sure that happens. So I basically work a normal or normal plus length of time, a typical day. My schedule has shifted around a lot over time, but I generally get several three to four hour chunks to just work on stuff each day. So that's it. I just, I spent a lot of time with this, and that's how the output gets done. Last question, because I can see the kids are popping in so probably time to wrap this up just we covered you know fang fang plus whatever you want to call it today that's a lot of tech companies we didn't talk about you know

Spotify Shopify Twitter TikTok Snapchat Tesla Uber strike PayPal uh tons of legacy tech companies out of all the companies that we didn't cover all the tech companies you didn't cover who do you think would be the most likely 10 years from now for us to kind of lump them into the fans and say, this company is inevitable? Of the ones you mentioned, Stripe has the best shot. Everyone I've talked to with Stripe is really, really smart. They're actually really nice, but they also ship, and there must be some dark secret there.

Somehow, because usually they're like companies where everyone's really, really nice, and they all get along really well together, and the product is what the product was 10 years ago, and they haven't added anything new. And then there are some companies where they're constantly launching new things and you talk to people who work there and you check their knuckles to see if the knuckles are bleeding because they had a fist fight with someone at the last conference

or the last meeting they had. Like there are companies that are at both ends of that spectrum. And then Stripe does a really good job of shipping things, shipping things that look great and work great as soon as they come out. They think really far ahead. And one of the advantages they have is that they seem to do a very writing-centric company. Like a lot of their code gets documented thoroughly.

A lot of people write lengthy things internally. A lot of people at Stripe seem to read a lot. So it's a very, very word-centric company. And I look at Amazon, and that's one of the things that sets Amazon apart. But also, I guess, when we do look at a lot of these successful tech companies, when information leaks about them, typically the companies are run by people who write very long, but not excessive, thoughtful, impactful memos to one another. That's how they think.

So I tend to look up to companies like that, whereas companies where the decision is verbal or the decision is made based on a PowerPoint or something like that, for whatever reason, that just seems like the wrong caters to make long-term decisions. And with tech, you have to make a lot of short-term tweaks, but you have to have to have a fairly long-term outlook because you have to design your company around

how foreign factors change, how usage changes, is how markets change over really long periods. One of the things I go back to when I think about that is Facebook, when they announced their series A, it's the first press release you find if you scroll all the way back on Facebook's on our page or on their old PR page. The first press release, it says it's 2006. They say they've just raised their series A,

and it says we're going to invest in operating expenses a lot of data dot and mobile. So six months before the iPhone comes out, they think mobile is one of the most important things they can do. Great. Great. Well, hey, actually, last question, this has been so much fun. When I launched this podcast, one of the first names I wanted to have on was you. So really glad to have you.

I certainly would love to have you back at some point. Anyone else you think I should, you'd enjoy hearing a podcast from where you think I should talk to? I think, let's see, I'll have to get back to you. There are some good people who come to mind. I'm not sure which of them actually do podcasts. Like, some of them are anonymous. Like a lot of, one of the things I love about finance Twitter is that so many of the best accounts are anonymous and they're, so they're able to, they're able to speak a lot more freely and they're able to be actual, actual human beings, which is nice. Whereas if you have a professional account, you have to sort of pretend that you only have a professional life or that you have this incredibly made for Instagram personal life.

So like a lot of those people are super fun and can go really into the weeds on on their topics. choice. But I'd have to think about who would actually want to do that. So I'll have to get back to you. No, this is perfect. Hey, look, love, love collaborating with you in a small way on this series. Love the, love the diff. Really enjoyed having you on. So thanks for coming on. And we will talk soon. All right. Sounds great. Thanks.