Yet Another Value Podcast - David Kim from Scuttleblurb on Carvana $CVNA

Episode Date: August 11, 2020

David Kim from Scuttleblurb (https://www.scuttleblurb.com/) comes on to talk about his research process and Carvana (CVNA).

Hello and welcome to yet another value podcast. I'm Andrew Walker, the host of the podcast. And with me today, I'm happy to have my friend and the founder of Scuddle Blurb, David Kim. So, David, how are you doing? I'm good. How are you doing? Doing good, man. So let me start off by just doing what I start every podcast with by plugging you. You know, when I think of Scuttle Blurb, I kind of think you're like the OG finance and your newsletter. Like, I think you were the first. And when it comes to like, you know, really end. depth research into companies and industries, I think probably the best. I think you can see that and kind of, I obviously asked readers and listeners for questions, and you can see most of the questions were actually not on the stock I wanted to talk about today. They were kind of on Scuttle Blurb and the process behind all your research. And I think that's just because of how impressive all the research is and everything. So it's a great service underpriced. I'd say anyone should subscribe if they're interested in fundamental research. So with that, David, can you kind

give us a background on yourself and how scuttleblower came to be yeah yeah sure um yeah thanks for all that and um you know i'm a fan of your research too and and a subscriber appreciate i think the real ogey of all is maybe uh ben thompson who was um kind of an inspiration for me when i got started on this over at trajectory and i mean i compare i think your your research on you know kind of fundamental valuation side is very comparable to his on the obviously tax and strategy side. So go ahead. Yeah. Yeah, well, cool. Thank you. Yeah. So, yeah, I guess just quick background, I spent, you know, my byside career basically consists of, you know, three years at Fidelity. And on the credit side, I was covering insurers at the time, spent three years at a

long, short, special situations fund in New York called Solace Capital, get a very brief stint at Citadel. And then I launched Scuttle Blurb sometime in, I want to say, like, fall 2016. But I think, you know, at the time, it was basically kind of like earning summaries and stuff like that, but the work got tedious and I didn't feel like I was providing much value. And so I kind of shifted it towards being more of a more of like a research journal. And yeah, I mean, I thought, you know, the funds I used to work at, we used to subscribe to a lot of like independent research that would cost, you know, thousands of dollars. And, you know, we'd have, you know, guys come in with these big kind of pitch books reports and it'd be mostly like kind of copy and

paste jobs with you know very little oh yeah a lot of like itself spittouts and stuff like that and so yeah i just thought like well you know if i could do some real analytical work and price it for like a fraction of the price um you know maybe people would want to read it and so um so yeah I mean, I guess I'm just kind of all over the board here with my research. I cover a whole bunch of different industries, just whatever I happen to find interesting. I think I had a frustrated subscriber once described my post as a Savoy tea time conversation that doesn't particularly lead anywhere. I think that's a pretty accurate description.

I think I would say, you know, there are maybe like three characteristics of the blog. The first is that, you know, I'm not really pitching stocks, right? And so I consider the blog more to be, you know, the goal isn't really to pitch you actionable ideas on a biweekly basis. It's more kind of like case studies on companies and industries that I find interesting. And so, so yeah, I think, you know, most investing newsletters, the main thrust of it is to get like actionable, actionable stock tips. And so, maybe that's a little bit confusing to people, but that's just kind of what it is. The second thing is, you know, because I cover so many

different industries and kinds of companies, I'm necessarily, you know, not really an expert on any of this stuff. And so, you know, I think of like, you know, like Matthew Ballie's kind of like thought leader in media, Ben Thompson who I talked to talk about earlier is kind of the same deal with, you know, fang, fang companies. Yeah, I don't know if you listen to that Matthew Ball interview with Patrick O'Shaughnessy. I have listened to all but the last 15 minutes of it. I've got the last 15 minutes queued up for my run this afternoon. So, yeah, it's pretty incredible.

I mean, that's just like, you know, like Moses coming down from the mountain with these stone tablets. Just, you know, great stuff. And, you know, he's really got his shit together. I'm more of kind of like, you know, like an informed amateur. I would say, like just kind of jumping around different industries and just trying to figure it out. And I guess the third thing would be, you know, I think like most investment newsletters, they focus mostly on like the quantitative stuff on the numbers. And so you might have like a write-up that, you know, maybe spends a paragraph on the overview and then the competitive situation. But

a lot of times the bulk of the analysis will be around, you know, the assumptions that go into an Excel model. And here's kind of like the target stock price. Whereas I think what I try to do is more of the inverse and spend most of the time on the qualitative stuff and then have the valuation be, you know, almost more of like a sanity check than anything kind of actionable or precise. And so, yeah, that's in a nutshell.

I love it. You know, I do agree wholeheartedly with pretty much everything you said. You know, one of the, it can be frustrating or it can be great.

One of the things for me is, I'll read this and, you know, your thing, it'll be five pages and very detailed industry overview, company overview,

all that. And then there'll be a page at the end. I'll just say, hey, here's some assumptions that lead, like,

yeah, stock seems okay, based on pretty reasonable. assumptions. At then, I'm like, oh, I read all this, and I'm not sure if I can do anything with it. But, you know, it's great for anyone interested in every time you send an email, learning about a new industry, learning about a new company. And again, you can see how in-depth it is by how many questions we got on research and process. So if it's okay with you, I'm just going to dive and take a couple of the reader questions on a process that they had for

you. Yeah, yeah, sure. Yeah. So the most popular one is, you know, you called yourself an amateur, I would say you're kind of more than an amateur at a lot of different industries. And the most popular question I got was, how do you dive deeply into so many different industries? Yeah, I guess I would say like, so I do jump around in different industries, but there can be like common threads between them. Yep, yep. And so it's not necessarily that I'm starting from a totally, you know, blank slate. So like I spent the last few weeks sort of looking at like S&P Global and FACSET and market access and trade web.

And those are kind of part of this whole large ecosystem of, you know, financial infrastructure companies. And they're kind of loosely related to like the exchanges, which I wrote about way back. And so there's some like commonalities between these companies. And there are oftentimes similar, similar kind of competitive advantages that they have. Yeah. Like between like a Moody's credit business and, you know, call it like an S&P index business where they're kind of both kind of standards oriented type of modes. And so there's oftentimes like more similarities than you might think across different companies. But yeah, I mean, I would say there's nothing particularly differentiated about about my process or my research, really.

I mean, I read through the SEC filings. I read through conference call transcripts. I try to talk to the company where I can. And so it's all just kind of like bread and butter stuff that any kind of any analysts might do. I would say that I do spend a lot of time. like trying to put together like a timeline for a company and trying to create like a narrative so I'll go back you know however long the financials go back and just try to see where the company was like 10 or 15 years ago if

that's possible and then try to build up like a narrative to how it got to where it is today and and then yeah I just try to do that for its companies and for anyone else relevant in the ecosystem as well And I think, you know, the other, maybe the other, like, differentiating thing is, you know, I feel like when a lot of people write their research, what they'll do is they'll first do, like, all the reading. And then at the end of it, they'll have, like, a stack of notes and whatever. And then they'll say, okay, now I'm going to, here's a blank sheet of paper. And then I'm going to start, like, writing and typing. Whereas, like, typically what I do is, like, I'll start writing.

as I'm learning about stuff. Okay, I'll try to like put things in my own words and I'll come up with questions, come up with theories. A lot of those theories are wrong, but the more I'll learn, I'll go back and I'll kind of revise what I wrote earlier. And I think just the idea behind that is just to kind of encode information

in your own kind of way. I think it's useful to like make sure your brain is plugged in, and as you're going through the work and that you're not just kind of like sitting back and passively consuming content because it's hard to make things stick if you're just kind of like sitting there reading and whatever, at least for me.

And so I just try to like, you know, write and proactively do things as I'm researching things. And then I would say like the other thing is, I try to think about companies, less as a collection of attributes and more as like systems, right? Where don't start saying, well, this is kind of an asset light business with recurring revenue and so forth. I try to think more about, you know,

if you change this one thing here, how does that affect this other thing over here, which affects that thing there? Because like all companies are systems that are part of these larger systems. And I think that's useful because it forces you to sort of like think about context you know and like one thing I was thinking is just um if you take like the less than truckload industry the LTL uh industry there these are basically

like logistics companies where they're you know taking no ODFL all those type of company exactly yeah so they're taking freight from you know multiple different shippers and then they're kind of recombining that freight based on common destinations and And in that group, like Old Dominion is probably the most capital intensive company. But they generate the highest returns on capital. But you know, IRC is also very capital intensive and they almost went bankrupt twice.

So it's like, I think it's hard to sort of draw sort of these conclusions like capital light or capital heavier sort of way to go. I think what's more useful is just to think about, well, it just seems like what really great companies do they focus on, you know, what best serves the customer? How do I kind of set up my organization to fulfill that need? And then they kind of manage the hell out of their costs

and make sure that they're doing this profitably. So like a company like Old Dominion, what they'll do is, you know, they invest in their own trucks, they invest in service centers. Yep. And by doing so, they're able to offer a higher level of on-time service to their customers and they're able to make sure that they're always there when their customers need them. And so, yeah, so during, you know, hard times, they're able to steal share from other competitors who aren't set up the same way as they are. And by doing so, they're able to leverage their fixed costs

and invest more in infrastructure and provide even higher levels of service and so forth. And so it's kind of like this, there are these like, you know, but you start to realize that there are these sort of like feedback loops that you see all the time in these kinds of companies. And yeah, I would say like in my younger years investing, I mean, when I was in my 20s, and I was more of kind of like a paint-by-numbers value investor where I was mostly just looking for kind of like statistically cheap stocks. Yep. Maybe I was short like the statistically expensive ones.

And, you know, you're just hoping for. like mean reversion in those kind of cases. But a lot of times, like, you know, in certain industries like that enjoy scale economies, you don't really see multiples mean revert because the returns on capital don't mean revert because there are these feedback groups involved in scale economies. And so what, so, you know, these competitive disparities that you see between, between two companies just get wider and wider over time. So, yeah, I kind of rambled there. No, no, no, that was perfect.

And I've got a couple things that I've been on. You know, I think the one thing, a couple things stuck out to me, but one that stuck out was you talked about how you like to look at a company as a system, not as kind of numbers or a bunch of different other things. And in prep for this, I was rereading your post on Carvana and CarMax in back in 2000, It was 2018, I think, and, you know, this stock was like $40 versus $200 today. But one of the things I was impressed with was it stuck out of my mind. You said, hey, this company, you can't just look at one individual piece and say, hey, you know, that's not much of an advantage here, right?

Like, they all work and tie into each other. And I think it almost kind of predicted exactly where Carbana was heading. So I guess my question would be, you know, just thinking about these systems, do you think it helps you pick up on flywheels? Like, I don't think a lot of people really thought about the Carbana flywheel. at that point. Do you think it helps you pick them up early or am I just like kind of taking one example and making too much of it? Yeah. I think, no, I think that's exactly right. I mean, I think, well, I mean, it seems to me, and maybe this is, maybe this is just a reflection of the environment, but I mean, it's just, um, obviously there are a lot of

very expensively priced stocks that are priced very, very expensively right now. But it seems like, okay, you've got like kind of these steady established compounder bro type companies, whatever, you know, MasterCard and and that whole bunch. That seemed like priced for, you know, the excellent companies that they are and will most likely continue to be. And so it almost seems like maybe the opportunity are, it is like in these emerging. remote companies in a sense, where like the gap financials just look like absolute dog shit, but the unit economics look compelling. If you can imagine a state of the world where there's just enough volume sort of running through the system. And so maybe like, I don't know, I'm not really saying anything original there. I think that, you know, you look at companies like Shopify and

and Carvana. Yeah, but there does seem to be sort of this increasing recognition that that I don't know, like maybe the game just isn't, isn't about like looking at historical financials and saying, oh, here's a company that does high returns on capital. It's more about like looking forward and seeing what this company that doesn't look so great today could possibly become. And the only way to like really do that, like there's different ways to do that, right? So I was reading Matt Levine's newsletter, I think last week where he's writing about, he's great, by the way. He's just like a national treasure. But he wrote about Nicola, right?

And he was like, this is a $13 billion market company that did like $36,000 in revenue last quarter installing solar projects at their executive chairman's house, right? And, you know, if you want to make that kind of like stock work, you've essentially just got to weave a story out of thin air. Just create some kind of fiction, right, to justify that mark cap. And maybe, like, I don't know the stock that well, but maybe like Tesla service similar case where I don't know where it's, maybe it's like a 300, close to 300 billion now. So, I mean, what would it take for Tesla to fill out that mark cap? It's probably not as prosaic as Tesla just becoming, you know, the largest auto manufacturing the world. Like there probably has to be a major kind of narrative arc where I don't know, maybe they start

selling the software that controls the automotive, you know, the autonomous driving space and they own that critical pinch point or something like that, right? I think it's something that a lot of value investors have gotten burned on, right? Like the classic ones are Tesla, Amazon, and Netflix. And two of the three, one of the three, I could agree with this short piece. But, you know, they looked at these and they said, hey, these things are priced ridiculously, right? They're burning money like crazy. They've never turned to profit.

They're priced to be one of the biggest companies in Earth. And with, I mean, I would only say this applies to Amazon and Netflix, but you would say, hey, these guys are the largest player with huge in a really growing market that could potentially be huge. Don't look at the historical financials. about their moat and run them out two years forward and all of a sudden these things they can be minting money if they want to right and i think a lot of times value investors have gotten uh burnt by shorting things thinking about trailing financials and not thinking hey you know elli turner and i in our podcast we called it the level up opportunity you control 100 million users you've got there them on recurring revenues what if you can you know cut your cost

or what if you can in amazon's case take them from you know just ordering from you and have them do everything from you, you know, subscribe to membership services and stuff. So, yeah, and it's always like a delicate balance between, yeah, between like, so I think, so what's different about, I think, names like Shopify and Carbana, even though those seem like pretty expensive statistically, is that it's sort of like, you know, if you, the future for them is more than sort of continuously repeating what they're doing. Yeah, right? Right? So it's not like you have to just make up some kind of story out of thin air.

It's more, I think what you more have to do is say, what are the natural consequences of this sort of feedback mechanism continuing to play out over multiple years? And that's a little bit easier to kind of get a handle on than like a story like Nicola or Tesla, where it does seem like there has to be some dramatic new, like Tesla, there better be some like aWS like hidden asset underneath those covers or something right and nicola there better be an actual business there like better be able to make cars or something yeah yeah you've compared shopify and carvana uh two or three times so far which i i can understand it in one sense where both have been screaming home runs both of them uh i think there was a huge short thesis on that has

probably been proven wrong but on the other hand like you know carvana and we'll talk about carvana more in a second, their basic businesses selling used cars, and there's only so much that can scale, right? They have to actually go out buy used cars and sell them, whereas Shopify's basic business is, hey, you want an online storefront, we'll run the entire back end for you, right? And that can scale pretty much infinitely. I mean, I guess somebody could say there's only so many online storefronts, but obviously that can be global worldwide and the cost of throwing one more onto that is absolutely nothing, whereas the cost for Carvana to sell another car, they do have to go buy another car, kind of screws it up, do everything. So what do you know, what do you

your mind connects these two that you've made this comparison so many times oh geez yeah i don't know maybe it's just the fact that like people just like are are um get are sort of divided about these stocks maybe it's the fact that they're expensive but also you know they're both kind of integrated vertically integrated businesses um i mean like Shopify is a platform in a way that carvada is not but you know it vertically integrates you know like payments and logistics and it provides like all the back end and that kind of stuff. And it just tries to make the entire process easier for merchants to get set up on. And, you know, Carvana, that's kind of like, that's kind of the thrust of their mission

and why they've sort of vertically integrated everything as well as just to make the shopping experiences as seamless as possible. So I guess there's that common thread, but I don't know why I put them in the same sense like that all the time, but I kind of do. Scuttle blurb, obviously the name of your newsletter, is based on the term scuttle butt, if I'm correct. And when we were talking about your research process, I didn't hear a ton of scuttle butt.

So is there a connection there or is just- Yeah, not really. I mean, I forgot where that word came from. I just thought like, you know, it's a, I could get the domain name because nobody knows. There's no, it's a made up word. And I don't know, I just thought it sounded like playful. And yeah, so that's why I kind of like went with it.

Last question here. So you said, you mentioned that you start writing proactively early, right? Like, how early do you start writing? If you're looking at a new industry, do you just pull up the first thing on this industry and you just start typing every, start writing an article then? Or when do you kind of hit the, let's start putting this onto paper moment? Yeah. Yeah, very early in the process, I'd say.

Yeah, very early. Yeah, almost as soon as I start, like, reading, almost as soon as I start, like, opening up the 10Ks and reading through stuff, I'll start putting things. Most of it's like, most of, like, my writing is, like, incoherent when I start. And I have to just, like, go back and revise a whole bunch. And there's a whole bunch that I delete. And so it's just kind of this back and forth process of tightening things up.

You spend a lot of time looking and writing about, I'd say either very high-quality companies or companies that, as we discuss, have the potential to kind of become platform, very high-quality companies. If we just ignored valuation completely, what would you say is the best business you've ever written up? Or you could go to the company with the potential to become the best business you've ever written up. Oh, geez. Well, I don't know.

I mean, like Moody's is a pretty phenomenal business. So, like, if you get embroiled in the whole, like, rating scandal of, I don't know. 2008 and 2009 and like essentially come out on scale rate like trillions of dollars of securities and still come out of that intact and just thriving today I would say that's a pretty strong indicator of a competitive advantage I think like S&P Global is a very very good business for for similar reasons I mean they've got a ratings business but they have also got an index business

which yep it's from like similar characteristics where you know you just have like buying from the ecosystem. But yeah, I mean, I kind of like these sorts of infrastructure utility type businesses where they play some kind of critical role that everyone else sort of relies on. And yeah, because those are just really difficult to replicate. So I mean, yeah. And if I could just put words in your mouth, both those businesses you mentioned capture a very small percentage of the value that they actually create. So they've probably got huge pricing power along term. Yeah, yeah, no, exactly.

And I think it's just, yeah, it's interesting because it's not, I was sort of thinking about this a few weeks ago, but like you think about like the index business where, you know, you have an entire industry that's sort of benchmark to like your IP. And sort of like, what is this like kind of based on, right? It's just kind of like a standard that everyone just kind of coalesced around. but it's a little bit of abstract in a way, but it's like, but yeah, I mean, those, but at the same time, it's not necessarily sure if that makes the business more stable or more vulnerable. I think I've come in more towards the side of it makes it more, more durable in the end.

They've just been around, I mean, I don't know how long the S&PIP has been around, but for a really long time and to kind of think that. that, you know, a new benchmark's going to, you know, spring up and just kind of replace that. And you would have, you know, trillions of assets kind of re-indexed to that. It's, it's sort of hard to imagine. So I don't know. I think those have, those are very durable. It is crazy, though, right?

Like, why can't you and I make Andrew and David's 499 or Andrew David's 500 and charge nothing for it, right? Or like, Vanguard. Why do they need to index the S&P 500? Why can't they just go and say, hey, we've created, they could be. even call it the Vanguard 500, right? And they could have their advisors to tell their clients, like, it's pretty much the S&P 500. We just don't want to pay the fees. We're trying to save you money. But we're 50 years into like the real professionalization of kind of retirement investing. And nobody's been able to beat back. So it just seems like that can't, it seems like that's too

entrenched at this point. Yeah. And I mean, like some of these indices like don't even make sense or like track the things that you think that they're track. I mean, it's like the Dow, the Dow Jones. Is that even really relevant today? I mean, it's kind of like, like a price weighted index maybe made sense once upon a time, but, but yeah, I mean, it's just because they didn't have computers, right? So it was, it was the easiest way to do it was price weight. Or I think S&P value and S&P growth, I think there's like a huge overlap between the top components. I haven't looked in a while, but I seem to remember that correctly. Yeah. Yeah. And you know, like once you own kind of like the dominant IP, like, you know, the S&P 500, there are other kind of like offshoots of those

indices that you can just create it's just like so whatever the trend happens to be so maybe it's like esg then you can create like esg related indices yeah and it just becomes like this um you know uh you just like create these products out of thin air and have people uh um adopt them like once you own kind of the key ip it's that great like you've always got these you've got own a great business and it just presents one great opportunity after another right like everybody wants esg i'm just going to launch the best ESG index out there. Everybody wants momentum? Here's a momentum ETIF.

Yeah, and there will not be an end to this because asset managers are always trying to differentiate in one way or another. And so if it seems like, you know, ESG is the way to do it, then that will become the thing. But there will always be like another thing that asset managers do. And then there will always be a need for some kind of benchmark to, yeah, to evaluate what managers against each other. So.

Let's see. Very popular question. We talked about it a little bit. Your write-ups are generally in-death, but it's more, hey, here's a stock slash industry overview that it's never really, I think you should buy the stock or here's a price target or anything. You're obviously an investor. What type of stocks do you, what type of stocks do you own and what type of stocks do you actually invest in? How much overlap is there between Scuttle Blurb and your personal account? I would say a lot, but I don't own that many stocks. So I own only in between like 15 and 20 stocks. And so I obviously write about a lot more than that in any given year. And so yeah, I mean, it's probably, I try to keep abreast obviously of the names that I own. But there are a lot more that I write about on the blog. And so it's very possible that any given stock that I've written about years ago, if you were to ask me, like, what's going on with that? In the last few months, I'd like it.

I don't know. Last question here. What write-up do you look back on and say, oh, man, I really missed that one. And conversely, which write-up do you look back and say, I was really early on that? I kind of hit the nail on the head there. Oh, I don't know if I've ever been, like, earlier really made the awesome call or anything like that. I think, I don't know, I thought my payments, my four-part payments write-up was pretty decent. People seem to like that.

I guess the one that I have, that I regret the most might be my Equifax right up from three years ago. So this was like after the breach. And I don't know, maybe I let like my emotions carry me away because I just thought that was such a huge, like, terror election of duty. And I just thought like, well, you know. this must kind of seriously damaged their credibility it's a business based on trust or whatever i was just like basically wrong about that it was like a few quarters of of uh readjustment but they they basically just got back on our feet and all the grandstanding from the politicians just you know died down as that as it tends to and um and they're basically doing fine

I know, Equifax, and we mentioned S&P and Moody's earlier, right? Like, these are trust-based, basically ratings agencies, right? They have a lot of power with your credit and all this type of stuff. They had these scandals, and yeah, there was a lot of political noise. They paid fines. They suffered in the short term. But medium to longer terms, the businesses were completely unscathed. But then you look at Wells Fargo, which had the account scandal, right?

which there weren't really, I mean, obviously some people had their credit games and stuff, but it was pretty minor in the grand scheme of things. I mean, they were opening up a lot of very harmless pay accounts, right? But the company seemed to have really suffered from that scandal. And it's just interesting me, I'll turn it over to you, but like, what's the difference? Like, why is one bank suffering for what should be, would seem pretty minor versus these things that were a dereliction of their core business, and it seems not to even matter. Yeah, I mean, I think maybe the thing is that, well, okay, so I think about a company like Moody's, right?

Yeah. And so I remember like during the crisis, I was working at fidelity and credit. And, you know, when that, in the midst of that crisis, there were other rating agencies that were trying to be, that came on to our floor and they were trying to pitch their services and why, you know, they should be, they should be, they should. be seen as almost kind of like the trustworthy version of Moody's. They're not conflicted because they aren't paid by the issuer and so on and so on. The by side of the way. Yeah, yeah. And so, but it was just met with, they were just met with a lot of hostility, like quite frankly. And I think part of the reason is because it would just have been a massive pain in the ass to like reconfigure everything around these new ratings because, you know, like in all of our systems,

we had like Moody's S&P and Fitch and we have to update the ratings in there. And, and And there was a whole regulatory framework that was like kind of based on, on, you know, the three main rating agencies. And so to have like a new entry and all of a sudden, it's just like, oh, come on. Like, you know, at the other day, people just want to, like, get their work done and go home. And they kind of, we kind of like all sort of tacitly understand that, you know, a Moody's rating isn't, you know, the end all. It's not truly, you know, there's more to it than just a rating anyway. And so it's just like, all right, here's kind of a quick and easy standard that works most of the time, the vast majority of the time that we can just kind of like congregate around. And I don't know, maybe it's just kind of the same way with with with with with Aquefaxes too.

It's just, you know, it's kind of embedded in sort of the workflows that a lot of these banks, all this like credit data. And so it's just, I don't know. I think like maybe when you're just so ingrained or embedded in the system and it's just so hard to rip out. you know, people are willing to just kind of like overlook, overlook what seem like pretty horrendous offenses. It's just crazy to do with such a dereliction of, as you said, their core business and they got away with it. But anyway, let's turn to Carvana.

This is the talking, I sent you an email and said, this is the one I want to talk about. I'm kind of kicking myself because I miss the seven bagger from the bottom in March. But why don't you just, why don't we just sort of, just give me a quick overview of who is Carbana, what do they do, and maybe the quick elevator pitch for them? Yeah, okay. Well, yeah, so, well, I guess let me just preface up front here. So I don't own any shares of Kavana. And I'm not, I'm certainly not pitching Kavana as a long or short.

And so I'll say some things that are positive, some things that are negative. And they can send their hate mail to you, Andrew. So, yeah, I mean, I think, like, you know, this is. So the basic problem is that, you know, buying a used car is sort of a stressful sub-part experience where it has been historically, where you've got, you know, like, you know, Sharkey salesman on a plot of land trying to sell you the highest price cars because their commissions are kind of tied to the price of the car. There's limited selection on a small lot. There's usually lots of paperwork that you need to get through to get the car. And so, you know, I would say,

like version 1.0 of sort of rethinking the consumer experience when it comes to car buying was CarMax, I think, right? So CarMax was set up as a subsidiary of Circuit City back in 1993. And the idea was just basically to apply all the best practices of retail to car buying. And, you know, like arguably, it was as disruptive to car buying. then as, you know, Carvano might seem to car buying today, you know, they were put in place like pretty sophisticated inventory management systems. They have like RFID tags on all the cars so that they can track like how quickly certain models were turning and that would inform kind of how much that they would be willing to pay for certain types of cars. And yeah, I mean,

they were basically kind of shifting from like a variable cost model to more of like a fixed cost model that that relied more heavily on data and technology. And so then in like 2012, 2013, you had like a handful of these online first players that came onto the scene. And a bunch of these were kind of these asset light peer-to-peer type models. were a few like shift technologies and room that were asset light in some ways in that they didn't own logistics. They didn't own their reconditioning centers, but they kind of owned everything else. But Carvana sort of emerged as sort of the most vertically integrated of the

bunch in that they owned their logistics. They own like the inspection and reconditioning centers, IRCs. They own the financing. And, and yeah, and so, and clearly they're the, the leader of the bunch. So, yeah, I mean, like, fun fact here, like, CarMax sells far more cars in a single year than Carvana has sold in its entire lifetime. And yet it has, like, half the enterprise value. How many cars will CarMax on a year? There are about less than, there are about less than 2% market share of, you know, 40 million cars. So 800,000.

Yeah, about 800,000. So, I mean, there's still, despite being the largest player nationally, it's still an incredibly fragmented market. So I think like the 100 largest auto dealers account for like less than 9% market share or something like that. And then you have Carvana here with like, you know, 0.5% or something like that. National. Just to compare, if my numbers are right, they're going to do, Carvana will do a little over 200,000 units this year. Is that right? Let me see.

Yeah, that sounds about right. Yeah. Okay. So that's the industry of review, Carbana, online selling of use cars. Yeah, so walk me through, like, this stock has been, you know, you wrote it up as a potential platform, like, what really separates an online selling of used cars from kind of the car max model? Yeah, I mean, I think it's more just, it's just more kind of like shifting to the way consumers want to buy things now. It's just presumably an easier way to buy cars. So, like, I mean, I have a soft spot for CarMax because I, you know, I've purchased three cars in my lifetime and all three were purchased that CarMax.

And I'm also in the process now of purchasing a minivan from CarMax. And so I got that process started like maybe a month ago. But, you know, that was like I had to, CarMax didn't have the van on the lot. And so they had to ship it from Reno. I had to pay the shipping costs for them to transport the van from Reno to the lot in Portland. And that required like a 15-minute phone conversation with somebody. And then once it's here, I've got to like, you know, drive to the car max lot. You've got to fill out the paperwork, test drive the car.

And, you know, it's not a huge deal. It's like a massive improvement probably over like the 40,000 other dealerships that just have that limited selection. there are lots, but, you know, it's arguably has a little bit more friction than just going to Carvana and doing everything online and having them ship that car to your house and you having seven days to drive that and then picking it up if you're not satisfied. And yeah, I mean, there are just these, there are these small, what seem like small differences in sort of user experience, I think with these kinds of businesses, can lead to big disparities and outcomes.

Because it's like you're funneling traffic down into what is sort of like a reflexive process like down below and over time that's kind of like compound. And so it's like, yeah, I mean, if you compare like Carvana's website to like shift or room, you'll notice like a pretty big difference. Like Carvana's got the 360 degree thing. They've got everything like annotated. They point out like all the dense and imperfections. they've got all this detailed information and essentially they're just trying to make it easier and easier for you to make that step of actually buying the car without feeling the need to you know go somewhere and test drive it and it seems like a small small thing compared to like room or shift but but yeah I mean once you kind of have the scale economies in place like further down the funnel I think those small differences in ux can possibly like compound over

time. So you mentioned scale economies a few times, right? And CarMax sells 800,000 cars a year, while Carvana is going to sell maybe 200,000 cars this year, right? So why can't CarMax do everything that Carvana is doing, right? Like CarMax has the 800,000 per year scale. They've got way more scale. They should have way more inventory because they're selling more cars. They should be, there should be a lot of advantage there. So what separates Carvana from CarMax? Yeah. I mean, I think it's, I think, I think, part of it is just like you know if they if they wanted to replicate what Carvana is doing they would have to sort of reconfigure the way that their

supply chain works in a sense because right now they've just got I don't know like over 200 CarMax stores spread around and then they're selling maybe like 6,000 cars per store something like that it's still like a pretty like local type of business whereas you know Carbana's inventory it's more kind of like nationally pooled so like a typical Carbana IRC Well, an average one might sell like 50,000 cars. It's an inspection in refurbishing center, I guess what it's called. But yeah, I mean, these are huge, massive, like 40-acre type monstrosities with like hundreds

of thousands of square feet. And so they get the car in there, they drive the car into like this big dome-shaped photo booth and they do the 360-degree photos and then they do the 150-point inspection. It's a process that takes like, you know, a few hours and then it's like up on the site, right? But yeah, I mean, it's – but yeah, so they've launched, I think, I don't know, maybe 11 or 12 of these things right now. So it's – so they've got kind of this more – a more kind of concentrated national inventory pool, and then they've got sort of like the last mile logistics around that. And so CarMax, they're trying to do, like, it's not like CarMax is, is totally blind to this.

And so they're trying to do sort of like an Omnichannel experience and leverage sort of their store base and what they've got. And I don't know, oftentimes when I hear about like companies trying to pull off Omni Channel, it sometimes appears more like splitting the difference between. what incumbents good at right now and the way trends are going, right? So it's sort of like a middle ground that they're taking instead of like a full-on leap. And I don't know. It's still honestly kind of like an undetermined, we don't know really the answer. Like maybe they will sort of be able to pull this off.

But right now, I mean, they kind of launched the Omni Channel back in like 2018. I would say the traction has been, I mean, not outstanding. It's been pretty limited and most of it, I think now there's like maybe 10% of their units are sold through either curbside pickup or home deliveries. But most of that 10% is like curbside pickup. So it's still like it. Yeah. If I just had to boil it down, I would say car mats is an improvement on the used car dealer model, right? Like they still have the local lot and you still go there to buy your car, but they've improved on a lot of the different things, right?

They've bought some better buying practices. They bought a lot of better selling practices. They brought a lot of data to the thing. I'd say, so that was an incremental improvement. Whereas Carvana, and I think you've used the term zero to one, they're not even close to the old kind of car lot thing, right? They are used cars, yes, but they take a nationwide used carpool, you go online, you look at the car. they've got great photos and stuff so they've shown how to do the selling online properly and then they'll kind of deliver the car to you so it's not even a used car salesman it's not a used car a lot

anymore it's just getting you a used car and cutting out a lot of kind of the local selling calls would that sound right to you yeah yeah you said that much better than much better than I did so yeah no that was good um but yeah so I mean basically so the way this model works and what they're going for here is you know so to get the most to get sales you need to have a wide selection of cars and you need to be able to deliver those cars pretty you know as quickly as possible and so and so you need to invest up front in logistics and these IRCs and inventory and so by doing so you're you're sort minimizing the distance between these inventory pools and and the consumer and so idea is like by doing so you can convert your sales faster your customers are happier because they get

their car quickly they there's word of mouth there so Carvana has excellent you know NPS scores and so that sort of draws in sort of more sales volume and with that you're able to leverage fixed costs you invest reinvest some of that into price but then you reinvest some of that into into more IRCs and you get you know and so on and so forth And so, yeah, and so this is sort of like this pattern that, or this process that's been playing out for a few years now at Carbana where, you know, they launch into new markets. They gain, they advertise, they gain more and more penetration for market. Yep. And so what you've seen over time is gross profit per unit just trend higher and higher.

And so there are basically like three things going on going on there. You know, first they've been able to like cross sell a lot of like ancillary revenues onto the used car sale. So they'll cross sale like gap insurance and in service contracts. They'll take a commission on that. Financing, I guess is the big, the big one. And then the second thing there is, is they're getting leverage on the IRCs that they open. And then the third thing that they've been able to do is just source more of the inventory

for, from customers instead of through auctions. And so, yeah, they've gone from like less than 20% of their inventory source from customers from trade-ins a few years ago to maybe more like 40% now pre-COVID. So they actually said in July they bought more than 100% of the cars they sold from their customers. So can you, why is that a big deal? Why does it matter if you're like to me, from the outside, I hear you bought more cars from your customers than you sold to them.

I say, oh, you've got an inventory problem. Like you're buying more than you can sell. But why is that actually a good thing? Well, I mean, compared to like sourcing inventory through auction, you're, you're probably better off by like $500 to $1,000 per unit, just sourcing it through customers. But also, like, there's sort of a feedback effect between the amount of inventory you have and the leverage that you get from advertising. Like, it doesn't make sense to advertise in a local market if your inventory is kind of running low. And so you don't want to draw customers to your site. and then have them realize that there's not live in the Twitter there.

So there's a little bit of like a feedback effect running that way as well. And so, so yeah, I mean, they've done an excellent job, just improving their GPUs over the last few years. And also, obviously, as they're running more probably through the system, they're getting leverage on their operating expenses, like advertising and the vending machines that they've got up and everything else sort of in that bucket. And so, yeah, I mean, if you look over time, I think if I, you know, they've got gross

profit per unit that's climbed from $177 in 2015 to $2,323. And then if you look at SG&A per unit, that's falling from, you know, $4,500 per unit to about $37.00 today. So you've got, you know, gross profit per unit going this way. And you've got like SG&A per unit going that way. And the idea is like, you know, eventually those two lines will converge. And then gross profit will overtake SG&A. And yeah.

And then they'll kind of be profitable at some kind of scale. Now, of course, like there's, yeah. No, I'm just kind of angry because I've got the gross profit per union going from 206 in 2015 to 3K in 2019. So I'm just kind of angry you front-ray in my next question. But two things, you mentioned excellent customer service scores. I mean, I think that's almost an understatement. Correct me if I'm wrong, but their NPS is along the same lines of like Apple, right? Their net promoters score NPS.

Yeah, it's along the same line of Apple. And then can you just dive a little bit deeper into? We talked about the gross property unit 200 or so in 2015 to 3,000 in 2019 or so. That is a mammoth, mammoth increase. Can you just dive a little bit more into what's driving that? Because it's not all just sourcing more from your customers. Yeah, no. I mean, I think it's a combination of sourcing from customers, leverage on the IRCs,

and then just the cross-selling of these ancillaries. But I think one interesting thing is like when you break, so, you know, if you compare the GPU at like CarMax versus Carvana, I think like CarMax, if I'm looking at this, almost 4,000. Well, okay, so if you blend retail and wholesale GPU, they're kind of like 24 or something, and then Carvana is at like 23 something. So it doesn't look like the disparity is that wide.

But yeah, but one thing to note is that like Harmex, there's a lot more wholesale volume. And so that's kind of a lower GPU and that drags things down. But when you like break things out into like the used car GPU, the ancillary GPU and the wholesale GPU, the two areas where you see like a huge disparity between CarMax and Carvana is on the retail GPU side. So CarMax is running like 2,200 and Carvana right now is running like 1,400. And then the wholesale side, CarMax is doing close to 1,000 and Carbana is there at like 385. And so if you had to like point to a bull case, it would be like that would probably be it for as far as the new profits go is just that well if these things can converge

then or get close to converging they won't converge all the way especially in wholesale GPU because CarMax owns its own auctions and so they don't have to pay like certain fees on on that on those sales but but yeah the general point though is that there shouldn't be anything stopping the convergence between between those GPUs. push back there a little bit because I mean Carvana in your you know the company more than me I've just but one of the things they constantly say is hey we invest a lot of our savings back into our product by giving our customers lower prices right and that that's kind of how we get this keep this flywheel spinning we we have a little bit lower prices so we sell more volume so we can lower our price a little more

and that's kind of how we keep the father swimming so isn't saying hey Carvana is going to fully close this gap with CarMax when Carvana has lower price model isn't that like saying Walmart is going to close the average price gap with, I don't know, what's a, I don't know what Louis Vuitton or something, you know? It's two different models and they can never fully close that. Yeah, yeah, you may be right. I don't know if it'll fully close, but the idea is that they're so under leverage right now. Like their IRCs, I mean, are like 50% utilized. So there's still a long way just to like fill out like their current capacity. And yeah, so it probably, maybe it won't Fully close, I don't know, but certainly there's a lot more, there's a lot of, like,

there's a big gap there between where CarMax is in Carvana. So let's see on the IRCs for a second. By the end of the year, they're going to have IRCs enough to do over 600,000 units per year. And you and I mentioned earlier, they'll do a little bit over 200,000 units this year. 600,000, I mean, that's a triple in size, right? They're actually starting to rival CarMax's size at that point. Is that reasonable? have they built out too quickly or do you think like the scale advantages are so good

longer term they're even bigger yeah i mean so this this is probably like one of my single biggest questions or doubts with a stock because i mean i think in theory like one case you can make is that so you look at like use the publicly traded auto dealers they're probably doing like mid single digit ebara margins and car max which is not totally i mean they only have like 2% market share, but they're about, but they're, you know, they've been around for 27 years. They're kind of like a steady high single digit grower at this point. So pretty mature nonetheless, and they're doing maybe like 8% EBIT margins, but if you kind of do like make some assumptions about, you know, about Parvana closing the gap in an SG&A

per unit going down, you can sort of like work your way to this scenario where, where Carvana's doing like 11% margins. I think like management has an expectation of doing between 8 and 14, something like that. And so I think to some degree I can buy that. Their margins will probably be, could be like structurally higher once they're fully mature because they're, no, they're not hiring like salespeople at these car lots. And they've got, you know, each of their IRCs is essentially selling like eight times more cars than like a store unit.

And these IRCs are located kind of out in the boonies. nowhere, in the middle of nowhere. And so you just like, and so it's sort of like kind of makes sense. I think my, my bigger question is sort of around, sort of like, sort of like the growth rate. Like what is like kind of an appropriate fade rate for growth here? Because I think it's, because you can, you can sort of like work out this scenario. And I think I did this really quickly yesterday where it's just like, you know,

if you were to grow their units by like 40% per year, year out over the next like seven years and maybe they get to like a 5% market share. And you were to, um, you know, put that 11% keep it on margin on, put like a 30x multiple on maintenance free cash flows on that out here. You can maybe like compound at like high teens or something like that. And now like if you're a Carvana bull, you might look at like 40% annual growth as almost conservative in light of the fact that they've just been like five, they've like five their unit sales over the last two and a half years.

But, you know, one thing I wonder about it is like in their oldest market, which is Atlanta, where they've got only like 2% share, their sales growth there is like only 18%. And if you look at the markets where they opened between 2013 and 2016, it's only like 50%. And if you look at like CarMax back in the late 90s, they were growing by like 70% a year. But like less than a decade later, that was all the way down to like, you know, low teens, right? And so I just wonder, it's sort of like, it's a little bit of a mystery. I don't know the answer to this, but I just wonder if there's some kind of like constraint, whether it be like sourcing inventory or what that where like maybe the fade rate for this growth is a lot more precipitous than maybe a bull might believe.

because it just seems easy to just say, well, they've just, look, they're growing triple digits right now. Like, what are the chances of that? Let me, I agree with everything you said, but let me flip it around because I think the company's bull thesis would partly be, the CEO said this early last year, I believe, he said, hey, right now there's 40 million used car sales in America. We think we can get to 5% market share, right? But right now, it's such a pain to sell your car that you the average car only trades hands once every 6.75 years if we can get that down to once every six years the market's 15% bigger and we can go from you know 2 million to 2.3 and by the way we'll probably take more than our share of that because it's getting the market's growing because

we make it easier right uh something like uber the parallel i would kind of draw loosely is the taxi market was 100 million but the taxi market sucked it was tough to get a taxi payment sucked all that Uber comes along. It's super convenient. You tap the button, you get it. You know, the market turns out to be 20 times as big. With Carvana, hey, it's easy to, easy to buy and sell. I get a better price because Carbana, they turn it over faster. They might be able to pay a better price on my sale, and I can better get a price buying. I can kind of flip this a little bit faster. What do you think about that? Yeah, I don't have any strong thoughts on that. I mean, like maybe. I don't know. Yeah, I don't know. I don't know. It's not something I would like hang my, hang the thesis on. I mean, it just seems like one of those like how far do you want to take the, the bull thesis? Because there's a lot of different ways that you can, yeah, you can go as crazy as you want with it and still sound like semi sane. You know what I mean? So the most underrated answer in investing is I don't know. So completely agree with that. It was a little bit out there, but it was just something that came on. I've got. And I think also there's just like, like there's sort of this inherent, because I mean, if you take management's expectation of like eight to 14 percent, you know, margins at scale, I mean, if they have like 10 percent in the market, that's like 80 billion of revenue approximately. And like six points of margin

difference, that's like five billion of EBITO, like you put a 20 multiple on that. That's like $100 billion of enterprise value, like between the low and high end. And like their enterprise value right now is like $35 billion. And so it just speaks to this notion that there's just a huge amount. I don't blame management for putting such a wide amount of fur frog making their guard rail rail so wide. But it just speaks to like there's just a lot of uncertainty as far as just how at the end of the day, unit economics sort of drop down and sort of what like a sustainable growth rate is. So there's just like a lot of questions around that, I think. I have two or three more and then I'm going to let you go.

COVID, I think there's been, you can't try, you can't take flights pretty much anywhere right now, right? I think there's been a huge use car boom over the past couple of months as people readjust to, hey, Uber's are out, don't want to transport, don't want to fly anywhere, we're going on vacations, we're driving, right? We're fleeing the city. We need a car for our suburban or rural lifestyle. Do you think there's a possibility, you know, obviously Q2, they started talking, they saw real momentum into the end, July seems like it's going to be good. Do you think there's a possibility they've pulled forward a lot of demand, but actually, unlike a lot of kind of the e-commerce plays, like they pulled forward so much use car demand that there's kind of a barren period on the back end of that, if this makes sense? Um, yeah, it's possible. Although, like, I mean, one of the remarks, I think, so I think the way the, the, the quarter sort of panned out is, like, April was really ugly for them, um, where they're down like 30% units. But then, but then. the following two months they were up like 20 and 30 not sequentially but like year over year I think like one of the things that they were they kind of pointed out

was that they were more so like constrained not by demand but just by by by the speed with which they could put cars on the website like it didn't seem like they were really trying to fulfilling the true demand that was like out there they seemed even constrained on that on that end so i don't know perfect uh there has been a this ties into a lot of the stuff we've seen but i believe carvana for years to set a short thesis around it it's been one of the top five most shorted stocks uh pretty consistently over basically sent the ipo uh and over that time the shares look like 15 or 20 times so you know can you just quick what was the short thesis and kind of what did the shorts miss when it came to carvana

Yeah, sure. So if I open that, so you send me that link, I think Spruce Point put out a short report back in 2019. And so let me just open it up here real quick. Well, anyway, I can't find it. But one of the points that they were making was just that, Look, if you look at their GPUs, it's, they're kind of like pulling a fast one on you because most of it comes from like financing income. And if you look at their core auto GPU, it's very low compared to CarMax, right? But I looked at that and said, well, that's the bull case. I think the biggest difference is just that they were kind of looking at this as a static entity.

So it kind of goes back to just looking at things as part of a process, right? So like, yeah, you could have taken a snapshot in time and said, well, their GPUs are a lot lower. But I think the smarter thing to have done would have just been to look forward and say, well, once they're kind of fully skilled, what does that GPU then look like? And so, so yeah, I think maybe it was just kind of just kind of like static analysis and not like kind of extrapolating ahead. I think that makes sense it's the old Netflix thing that's right? You're looking at the historics and you're saying, hey, these guys are way below their peers on margins. And then it makes sense. And then the real thing is, hey, you know, we run this four or two years. You throw another 50% of units on it. All of a sudden, the things, it looks like it's going ganglusters.

Yeah. The only other thing I would say, and this is probably just a mental bias for me, the company does a heck of a lot of equity offerings, right? They issue stock. Yeah. And that does make sense for a company with great unit economics to grow as quickly as possible and capture a big potential opportunity, right? But I don't really have anything there. It's just something that really sticks out to me as something I don't like to see in companies. Yeah, I mean, look, they're burning a billion dollars of cash a year to, like, build out this network. It's not like a riskless proposition. I think maybe sometimes that gets forgotten when we're looking at some of these like high flyers is that, yeah, there's like a fat tail maybe

on the right side, but there's also possibly a fat tail on the left as well. And so, yeah, I mean, look, they had to issue like a billion dollars of equity at like 60 bucks a share or something during the COVID to kind of reshore their balance sheet. And so it's not, so yeah, I mean, it's just, yeah, I sometimes wonder about this because I was like thinking about, like, if you take a, we don't have to go here, forget. It's one of the things I think I've struggled the most with in the financial crisis and with companies in general. Like, in the financial crisis, there were a lot of companies, travel related to everything, that it was clear three years from now, once the virus behind them, these companies had value. But there was a

big question of how do they get to three years from now? They're going to need financing. And you could tell me that financing comes in a bankruptcy where equity is wiped out. Or if the equity trades where it was three months ago, they issue 2% of shares and they're completely fine. So it's one of the things I've struggled with with those, with Carvana, where we're going to dilute shareholders like crazy, but the returns on that dilution are going to be so good. It's actually good for you. You know, exactly. It's like it's a little bit binary in a sense. But, yeah. Yeah. Last question.

Is there anything that I should be asking you on Carbana that we haven't covered or any point that you think is kind of important that we kind of missed here? No, I don't think so. Not really. I think we covered the main points. Well, hey, it's a fascinating company. You know it well. It was a great conversation.

Last question before we go, is there any company that you'd like to hear a podcast on or any investor that you think we should have on the podcast that would make for kind of an interesting. interesting conversation? Yeah. Yeah, you know what? You should get Lyle Taylor on the podcast. Do you know Lyle Taylor? I believe he's in Singapore,

so the timing differences might be different, but I will try to get him. Yeah, he's a really interesting guy. I think he does, he writes really interesting articles on his site. I forget the name of the site, LT-3000 or something like that. But yeah, he puts out like thought-provoking

essays. And so I think it would be interesting to have him on the podcast. Great. Well, we'll try to get him. Well, David Kim from Scuddleblur. This was so great. Everyone should subscribe to his work. It's awesome. This conversation is great. Thanks for coming on, man. Cool. Thanks a lot.