Carvana Co. (CVNA) Earnings Call Transcript & Summary

So we're getting some questions on how you envision Carvana's business evolving over time. And the questions have to do with marketplace consignment, service or any of those things of interest.

Sure. So let me start with this. Last quarter, we sold about 90,000 units. There's about 40 million transactions per year. That means about $10 million per quarter. That means we are about 0.9% of the market last quarter, right? So the opportunity in front of us when we think about it that way, we're kind of a little less than 1% is still 99% as big as it was on day 0. So I think that one thing we've got to stay really focused on is just the core opportunity of what we're doing right now sitting in front of us is really, really, really, really large, and it's going to be a lot of work because It's not trivial to expand a business that is -- that requires the operational complexity across the chain that our business requires, right? That's the same thing that generates the moat, which is good, makes it hard to expand, which [ all those content ] is bad. And so I think that needs to be our central focus for sure. Now that said, it's not -- it's also true that for customers to transact in cars, you basically need -- you need a customer, you need a transaction platform, you need a vehicle, right? Those are like the 3 fundamental components. And so for us, those things are the brand that we're building, it's the transaction platform of Carvana.com and then it's our underlying supply chain. And all of those are kind of valuable assets that can in theory be separated to some degree. You can kind of unplug one portion and plug in something else. When you do that, there are impacts to customer experience, there are impacts to unit economics, there are impacts to scalability. Oftentimes, those impacts trade off in ways that oppose each other. And so you have to kind of think about what you should be doing. I think that we are aware of that reality and the opportunities that arise from these 3 assets that we've built. And so I think we'll pay close attention to what those things are. But we also need to make sure that we stay really focused on just doing the thing that we know expands to a size that could be extremely large. So I think there's going to be a lot of opportunity, I think, we'll around the edges be testing for sure because we're aware that, that opportunity exists. But I think we have to balance that with just the focus of growth.

So there's also a question on -- I guess I would phrase it as customer engagement. So recognizing you're a young company and the length of the purchase cycle in this category, you may have seen some customers twice, right? But you're not seeing them like Chipotle once a week. So the question is really along the lines of, look, if I bought a car from you, but I'm getting a service by some other dealer, but how do you maintain engagement with me as opposed to me starting to develop that relationship with whoever is servicing my car?

Sure. I think that's a good question, and I think it's a really important question. And I think when you think about kind of the long-term opportunity here, I think getting customers into our ecosystem and kind of keeping them there because we maintain a great relationship and we deliver a great customer experience across everything that we do over time. It is really, really centrally important to achieving our ambitions. So I think that's very important. I think the most important thing that we can do upfront is the most memorable part of buying a car is the experience of buying and getting that car, right? And that is the moment in time that when you think about buying your next car, you're kind of reference point and what you're thinking about is all the times when you've bought cars in the past. And so I think delivering a great experience is the most important thing that we can do there at that moment in time. I do think maintaining a relationship with the customer across their life cycle is also important. And I think there's a number of things that we already do there. Through our financing platform, we maintain contact with our customers through managing our warranty. We maintain contact with our customers. So there are a number of things that we do there. There's undoubtedly more that we can do there as well. Now the sum of what we've done so far, it's very early in seeing repeat customers, because going back to that kind of mental model earlier, if customers are buying a car once every 5 or 6 years, we're only an 8-year-old company. So -- and we had really low sales relative to today in the early years. So you're not seeing that much of the distribution of customers that have kind of got to their second purchase cycle yet. But the early indications there, I think, are very good, and I think we're excited about that. And there's some even kind of earlier indications and just customers swapping another car such as households buying multiple cars from Carvana. And so I think there's -- I think there's some really good signs there, but there's undoubtedly more that we can do over time.

So used car prices have been quite a hot topic this spring. We've actually seen a lot of unusual volatility over the past 18 months. But can you talk about what the escalation in used car prices means for Carvana and how do you position the company for an eventual normalization?

Sure. So this has been a very odd time undoubtedly. And so I think some of the historical patterns that I think remain the most predictive of what will happen over the medium and long term have kind of changed a little bit. So let me start with what I think the patterns are that matter over the medium and long term because I think that, that's the most important thing by a long way for how we manage the business and how an investor thinks about us over time. But generally speaking, there's just -- there's massive ballast in the market in terms of kind of what the gap is between the total revenue that you can get when you sell a car, whether that revenue takes the form of vehicle price or if it takes the form of vehicle price plus F&I income or whatever it is, the total revenue and the underlying cost of acquiring that car in wholesale market. And the balance that kind of ensures that those 2 things stay connected with kind of a rope of relatively consistent length is just that all dealers out there -- I don't want to say all dealers, but the vast majority of dealers have very, very similar processes and cost structures and profit goals. And then they're connected even more deeply by the fact that most dealers don't have cash reserves to take losses over any sustained period of time. And maybe most importantly, but least concretely, there's a very strong culture of not losing money on any given vehicle transaction across all of automotive retail. So generally speaking, what that does is when there's tens of thousands of dealers that have the exact same cost structure and culture and profit goals, it keeps like a really stable gap between those 2 things. Now that kind of is true regardless of the level of prices. If prices are higher or lower, the gap generally remains the same. When we've historically talked a lot about seasonality, what seasonality generally means is its variation in vehicle depreciation rates. Vehicles are depreciating assets, but the rate at which they depreciate on like a daily basis can vary based on the kind of relative balance of supply and demand across time. And I think that, that has led historically to some fluctuations in our GPU. Where like if you average it out across a year, it's -- your connected by this rope, but when you have a whole period where you're holding a car between the moment that you acquired wholesale and the moment that you retail it, that depreciation that occurs during that whole period impacts the margin that you ultimately see. And so I think in these moments of abnormal vehicle price movement, you're effectively seeing like a magnified seasonality is maybe a way to think about it. So it's -- that's undoubtedly going on. I don't think that we should expect that to be persistent. I think that the fundamentals of the balance that exists between the wholesale market and the retail market, I think we should expect to be very persistent. So I think that like when you're thinking about a long period of time, that's what matters when you're thinking about dynamic moments of transition. I think that you can have some expansion in that gap. Historically, as you normalize, you don't see a huge collapse in that gap, right, which is interesting. Now I don't know if that will always be the case. But I think the reason that, that occurs goes back to the fundamental ballast that sits between those 2 markets where there aren't cash reserves and there is a culture of making money. So for example, when we saw something that would kind of maybe look like what an unwinding of this unique time would look like if we go back to the onset of the pandemic in March of 2020, we saw wholesale prices fall dramatically. We didn't see retail prices fall dramatically because you have all these dealers in a position now where I paid a certain price for my car, they're sitting on my inventory line at that price. If I take losses, I got to come out of pocket on my inventory line. If I just leave them there, I don't. And I also have got the same expenses that I've always kind of had. And so you tend to see dealers hold prices flat for a long time. You didn't see retail prices start to drop down until dealers clear through that inventory. So I think historically, when there's less depreciation, you can see some expansion in that gap. When there is more depreciation, I think you either don't see a collapse in that gap or you do, but it historically at least hasn't been as pronounced as the expansion can be. So I don't know. I think we're in a unique time, and it's hard to forecast the future, but those dynamics that I think are at play. And again, most importantly, over the medium or long period of time, it doesn't matter. What matters is kind of what your customer experience is, what your cost structure is and what your variable revenues are compared to the rest of the market. That's what matters because that's what creates your gap versus the rest of the market's gap.

The used car space is so unique in that up until the last 5 years, really the only national brand was CarMax. But certainly over CarMax's time and business, we've seen them be able to take what I call a trust premium on their pricing, right? They have earned consumer trust and what's a untrusted transaction and they basically priced at or above the market. You're in a position now where, obviously, you, too, are a national brand with Carvana. Have you been able to kind of start to get that trust premium in your pricing architecture? And has that been part of the benefit we've seen the GPU over the last several years?

So I like this framework. So let me expand on it a bit and then maybe interpret it in a slightly different way. So like the way that I like to think of brand, brand is customer-facing information. It's like what do I know and feel about this logo is kind of what brand is. And so to me, like I like to decompose that into 3 steps. The first step is, do I know that you exist at all? That's kind of unaided awareness is like the label that we give to that one. The next one is do I actually know what you do? I may know that you're in the car industry, but do I actually know what you do? I go to your website and there's tens of thousands of cars and you deliver them as soon as tomorrow, and there's a 7-day return policy and prices are good and I can get approved for financing. And oh, by the way, I can actually trade in my car. Do I actually understand that all those things exist? Then I think the most important and kind of final step in that is trust, which is tremendously valuable. And I think the value of that is proportionate to the anxiety of the customer feels in their transaction, which I think is high when the price is high. And I think it's high when you're in an industry that traditionally hasn't had a lot of trust. So I think that what we're trying to do when we build brand is we're trying to accumulate a lot in that kind of bucket that has those 3 steps. And then when you do that, I think that, that kind of converse, you can think about that as saying, like, okay, you can charge higher prices because of a particular label, which we'll call trust or you can kind of abstract from that in a slight way and say, what filling up that bucket effectively does that increases demand. And once you have more demand, you have options, right? You can sell more cars at the same price, you can sell the same number of cars at a higher price. And so I think that what we want to do is we want to build that bucket up as quickly as we can, as efficiently as we can. By the way, a huge part of that too is delivering great customer experiences one at a time over and over again and getting the word of mouth that, that kind of drives that out there. And then we want to try to utilize that in a way that is strategically intelligent given the sum of our goals. But I -- what I'll say to answer your question slightly more directly is, if we look at our pricing over the last several years and kind of the -- if we look at the underlying elasticities that customers exhibit on our website, given all the testing that we do, it would have, on average, suggested that we should have raised prices over time if we were optimizing for real-time cash flow. But generally speaking, we have been optimizing for that. We've been optimizing for building a really big high-quality business and delivering great customer experiences, and that's a slightly different thing. It's optimizing for maybe present value to like turn it back into economics, but it's a slightly different thing. And so I think that we've kind of lived in a world where if we were optimizing for real time, we already should have been raising prices. And I think if we get ourselves to a spot where we're building a really, really strong high-quality brand and our bucket is much fuller, I think it's going to give us more options in whatever direction we choose to move in the future.

And I think we have time for one more question. So I wanted to end on CAC, which I know in this industry is really interesting to analyze because you're talking to customers for things they might buy 3 or 5 years from now. So it's not necessarily simultaneously matched in real time. But if we look at the cohort analysis, it looks like Atlanta, which is your oldest market, has kind of bounced around roughly $600 for the last few years. Is that kind of the low-water mark, if you will, on where you think CAC can go for Carvana? Or is there something kind of that's keeping Atlanta in that kind of stable range?

Definitely not the low-water mark that we expect to be able to hit. I think our long-term financial model remains a good estimate of what we're aiming for. And so that long-term financial model has a goal for customer acquisition cost of around 1% to 1.5% of revenue. If we look at Q4 of 2020, which is the last time that we reported of that data, I think Atlanta was around $520. And I think our average revenue per transaction was just shy of $25,000, so it was barely over 2%. So kind of versus our long-term goal range of 1% to 1.5%, which is informed by all kinds of things, we were actually somewhat close to that, but we were also achieving that at a time where we disclosed on our Q4 call that in January, Atlanta ruled this market, had grown year-over-year about 45%. Well, that's a lot, and that's comparing to a pre-pandemic period. That's not comparing to a pandemic period. So there's clearly a lot of investment in brand that is still occurring in our market that we think is still very immature, even though it's our most mature market. And that we've also been running at inventory levels that are well below where we would like to be. So we think there's a lot of relatively simple drivers over time that will push that lower. But we also think that we're close enough to our long-term financial model toward that's not -- that isn't our biggest area of focus is driving leverage in that line item today. Our biggest area of focus is positioning the business to be really large, deliver great customer experiences in the future.

All right. So I want to thank you both, Ernie and Mike, for joining us. We are out of time. And everyone, have a great rest of the week.

Perfect. Thank you.

Take care.

Thanks everyone.