Carvana Co. (CVNA) Earnings Call Transcript & Summary

Good afternoon, everyone. Before we get started, if you are a member of the press or media, please disconnect at this time. This is a restricted line. Any unauthorized party in this meeting or any unauthorized use of the information communicated in this meeting is subject to prosecution to the fullest extent of the law. Any unauthorized person, including the media that is on the line at this time, please disconnect. Please note, today's call is being recorded.

All right. Good afternoon. Thanks, everybody, for joining us. I'm Zach Fadem, retail hardlines and e-commerce analyst here at Wells Fargo, and I'm happy to welcome Ernie Garcia, Founder and CEO of Carvana, as well as Vice President of Investor Relations, Mike Levin. So before I kick it off with the Q&A, a quick note for those of you listening in today and would like to ask questions, I'm happy for this to be interactive. Feel free to e-mail them to me at [email protected]. That's Z-A-C-H-A-R-Y.fadem, F-A-D-E-M. With that, I'm going to hand it over to Mike for some opening comments, and then we'll get to Q&A.

Thanks, Zach. Today's discussion may include forward-looking statements within the meaning of the federal securities laws, which are subject to risks and uncertainties that may cause our actual results to differ materially from such statements. A detailed discussion of the material factors that may cause actual results to differ from forward-looking statements can be found in the Risk Factors section of Carvana's most recent 10-K and 10-Q filed with the SEC. The forward-looking statements and risks in this presentation are based on current expectations as of today, and Carvana assumes no obligation to update or revise them whether as a result of new developments or otherwise. I'll turn it back over to you guys.

Levin always does that. He really does an excellent job. It's pretty impressive. Every earnings call, he goes into that same commentator voice. He just nails it.

He's like the Michael Buffer of IR safe harbor statement reader, so very strong. So with that, Ernie, let's take it off with a high-level question as we get ready to close the book on 2021, it's obviously been a year of great milestones. And as you think about performance year-to-date, maybe you can level set us on how you think you're positioned into 2022? What are the biggest near-term opportunities, threats, challenges? And how should we think about the next chapter for Carvana from here?

Sure. I think we're very well positioned. I think we could be better positioned, I guess, just to be totally candid. The way we can be better positioned if we weren't facing the constraints that we faced especially through Q2 and Q3 and are working through now. But I think that we've got a very clear plan. I think that we've laid the foundations for great year next year over the last couple of years. I think that we've got a great team. It's reacted really effectively to the constraints that we've seen over the last 6 months and is working quickly to alleviate those. So I think, yes, at a high level, I just don't think anything is that different. It's just we still have a ton of work to do to try to grow into our opportunity, and it's going to be hard, and there's going to be a lot of execution challenges. But I think we're up to it and feeling great about where we are and just have to do all the work.

Yes. In Q3, you made the decision to use your term meter, your unit growth or pull back a little bit, despite what looks like an accelerating demand environment. Maybe you can start with walking us through that rationale in a little bit more detail. And also, is there any way to quantify the impact to sales or profit?

So what I would say is, I mean I do think that there's 4 major operational functions inside the business. You've got kind of customer care, which is answering any questions the customers have, verifications, title and registration, all of that, you've got the inspection centers, you've got long leg logistics and last-mile logistics. And so I think those are all the different areas where constraints are likely to show up because there are areas where you have to physically things and you have to plan ahead of time for the volume of work that you expect to show up and then you've to manage through that volume of work. I think in many of those groups, if we find ourselves constrained -- at least in some of those groups, if we find ourselves constrained, the way that they kind of automatically like kind of self-correct ends up not necessarily impacting customer experiences in a way that's super difficult. So for most of the pandemic, we've been most constrained in the inspection centers. And so what that means we're producing fewer cars than we would like. And so our inventory is smaller than we'd like. So customers that shows up at the website is less likely to find the car that they want, which isn't a perfect customer experience any way shape or form isn't what we're aiming for, but it's not a particularly friction-filled customer experience either. It just kind of is what it is. The car wasn't there, and I move on with my life and go to the next website. I think as the constraints move to other parts of the chain, the customer experience that the kind of the impact there that then kind of self meters your overall demand can be a little bit worse because all of a sudden now, it means that your delivery times are longer and that doesn't feel as good. It means your risk of missing a delivery window are higher. It means that when you call in with a question, there's a longer wait time to talk to someone and maybe a longer hold time in between. And so those constraints, I think, can feel worse for customer experiences. So as we move through Q2 and saw a huge increase in the demand for selling cars to us and that started to constrain these other operational portions of the business aside from inspection centers. The impact to customer experience that would kind of automatically self reduce overall volume they just -- they were -- they were impact to customer experience that we weren't as excited about, right? And so then you're in a place where you have to choose, okay. It doesn't really do you that much good when your system is out of balance anyway because your system -- your total transactions are going to move at the speed of the slowest part of your system. And so by electing to reduce visibility of certain statuses of inventory, certain types of inventory, certain locations of inventory, we were able to push kind of the customer experience feeling more in this direction of I don't see the car I want, and less in the direction of there's friction in the rest of the experience. And so thought that in light of the options that we had, that was a better choice. And so we chose to kind of proactively pull that lever. So that's, I think, the that's kind of how the whole system unfolded. I'm not sure that we would want to try to engage in the hypothetical of what would transaction volumes and economics had -- have been in any super detailed way. We do think that we -- if we were better prepared and didn't have to make those choices, we think we would have sold more cars. We think we would have bought more cars. We think our expenses per transaction would have been lower, but I wouldn't want to quantify that precisely.

Makes sense. So in terms of the operating constraints in Q3 and how they carry forward into today, we've talked about throughput. We've talked about hiring both across the system. What do you think needs to happen in order to take your foot off the brake here? And do you think we're nearing a good place for you to do that?

Sure. Well, so I think constraints are a function of there being a requirement for more work in the system than we're prepared for. And so that's what generates the constraints. And then I think the first order way that you can alleviate those constraints most quickly is always to hire more people. And so that means that you've got to -- and you can hire people in the right areas of the system that are most constrained because you're not always perfectly in balance and so you want to hire people in the areas that are most constrained because that's where you get kind of your lift. So I think that's the immediate-term focus is just how do we do a good job kind of recognizing the areas where we need people the most, prioritizing that and then putting more effort into hiring and training more quickly in those areas. And then I think that there's always kind of the medium-term goals that we're working on, which are more driven by technology and process, which are about reducing the amount of kind of variable work necessary to complete any given transaction. And I think that those efforts are always ongoing, but they're never the things that kind of solve an immediate-term problem. So I think we've got efforts going in both those directions.

Yes. I mean it seems like there's actually some opportunities here. Like for example, you're listing different levels of inventory in different cities, different locations, which means if you have a like-for-like vehicle in California and Philadelphia, it makes sense to sell the Philadelphia part to the New York car. To what extent have -- how long have you been doing tests and experiments like that? And ultimately, do you think that there are profit throughput, customer experience opportunities from those types of tests?

I would say we're running probably dozens of tests at any point in time. Generally speaking, those tests are at a scale like we try to optimize our testing size to maximize speed of learning but to minimize customer impact of the test. And so a lot of times, when we've done those tests in the past, they have things that have kind of risen to the level of generating a ton of discussion like on an earnings call because they're just -- they're not impacting the results in a super material way. So I think we feel like we have a pretty good understanding of how those things work. And I think that's why we were positioned then to pull those levers when we felt like it was the right choice in light of the constraints that we were seeing. And in general, the way that, that has worked out has been very similar to what we would have -- how we would have expected it to work out given our preexisting understandings. So that's like how it is, that question. I think like you're going from here, I do think that that's all kind of exciting in terms of like the potential gains because the first order goal that we have to our priorities, right, customer experience, volume, gross profit and then expenses, we're focused on driving volume first for a myriad of reasons that we can discuss if you feel like it's useful to head into that. As it relates to inspection centers, that means getting inspection centers up and running and producing cars sort of first order kind of regardless of where they are because you can kind of fulfill transactions with just longer logistics legs. But then the second order goal, once you're trying to optimize your kind of -- your expenses and so your kind of variable unit economics is you do want to minimize the distance that any given car is shipped, that sort of naturally occurs as you expand your footprint and, therefore, have cars in more locations. So I think that first order as it relates to inspection centers, the goal is to get more up. And second order the goal is to get them in locations where there is both supply and demand because it minimizes inbound transport cost, it minimizes the outbound transport costs and then it minimizes time on both sides as well, which preserves depreciation cost and improves customer experience, but that should kind of happen on its own as long as we fill out the footprint.

Got it. And you've talked about opening 8 new IRC facilities over the next 12 months. First of all, can we talk about how this improves your capacity and your customer experience? And is there anything you can share about how the process is going in terms of securing locations and staff?

Sure. Well, I think the first thing that happens is you just increase production capacity, right, which then -- when production capacity goes up, then like those cars go somewhere. They're either going to incremental sales, they're going to inventory builds or both. Incremental sales is great because the business is getting bigger. Inventory build is great because it's improving customer selection and therefore, customer experience. So we think that that's obviously really important. And then I would say we're tracking well. I think there's always execution risk on all of these -- anything real that is done with -- that has like real operational components to it, those things tend to go according to plan or worse than planned. But I think that we're tracking really well, and we try to be thoughtful in the way that we communicate to investors that we try to have a pretty high degree of certainty before we start communicating things. So I think we feel like we're tracking well, but nothing is a done deal until it's done.

So just also on that to just clear the air for those that are concerned, should we be worried about the labor environment when you're looking to staff these facilities?

I mean I would say all else constant, it's not helpful, right? So it's certainly not a tailwind, and it certainly is some degree of a headwind. I do think that our experience so far -- and I want to be careful to not be overly confident about these things, but our experience so far is that we've been able to perform pretty well to our updated plans as we've worked hard to try to alleviate the constraints that we've seen over the last 6 months even in this environment, and that's taken work, that's taken effort on the part of the marketing team to try to put more effort into our employer brand. It's taken effort in people ops and the different operating groups to try to make sure that when someone applies, we interview them much more quickly than we have historically averaged. It's taken investment in building out career pathing that is very clear for all these different operational groups. So that when we talk to candidates, we've got a compelling story to tell. And so I think it's taken work, but I think that we feel like we're performing pretty well in the environment today, and hopefully, we continue to.

We have a ton of capacity questions, as I'm sure you can imagine, just to kind of level set there. Let's say you continue on this trajectory. You sell maybe 420,000, 430,000 cars this year, whatever it is, which would put you at 55% to 60% capacity on the 750,000 units. The question is, is that the right way to think about your capacity today? Or does that 750,000 contemplate added production lines that aren't built out yet in that facilities?

So we have to be careful of that questions because it's super interesting. It's -- today, we've had a couple of the smaller meetings, and it's very clear that like market's preferred way to try to estimate next year's sales volume is this number of -- we know what your facility capacity is. Now what percentage of that are you going to be able to produce. And so I think we want to be careful about not kind of unintentionally providing guidance. I do think that you can look at our past. Our past carries information that is real. The percentage of facility capacity at any point in time were able to access is a function of how quickly we ramp our facility up and also how recently we opened any given facility. If you open infinite facilities at one moment, your capacity is going to be 0 because you haven't ramped that up yet. So I think that that's the equation. And then our job is to try to continually get better at ramping facilities more quickly. And that's just another thing that you have to plan for and execute to, but it's not an overnight thing either. We do for many roles in the inspection centers. We hired people more than a year ahead of time. And so you have to have a plan that you've thought of in advance across many different groups to do a good job there. And so we're always working to try to grow the business as quickly as we can. And I think we're always constrained by realities of how hard all those different operational functions are and then we end up where we end up, but we're always pushing.

I think the second biggest topic that we get from folks is just on the used vehicle pricing environment today, up over 20% versus the past year. I guess, first of all, do you think this is good or bad for business? And then second, as I ask you to predict the future a little bit, any thoughts on how this dynamic plays out over the next year? I mean is it a gradual deceleration? Is it a sharp drop? Any thoughts there?

So the answer to the first question is I think it's ambiguous and it depends on what your priorities are as it relates to business. What I mean by that is, I think it's ambiguous for volume. I think it's probably all else held constant, a little bit worse for retail volume. It's all else held constant probably a little bit better for purchasing car volume. It's -- on like the merchant side, it's beneficial or a nonevent. And I think that it depends a little bit on specifics of your business model for where you are there. So I don't know, that's how I'd say it is today. Although I also think those are macro factors that I think in general are our hope and our history is that what matters more is how we perform against the environment more so than the environment itself, just because we're still 1% of the market, and we've been growing GPU rapidly over a long period of time. And so we more measure ourselves by kind of the growth that we're able to execute to and by the progress we make versus peers in the industry as it relates to margin. I think that, that's -- those are the measures that will matter over the medium and long term. And so that's more how we measure ourselves there. I don't feel like I know how prices are going to evolve over the next year. I do think that like the mental model that we have used is -- it does seem like basically international COVID waves that are disruptive to a very complicated OEM supply chain have resulted through many channels, most often discussed is the microchip channel, but many channels can reduce overall volume. And then when you have volume reductions, you see new car price appreciation and that pulls up used car prices. And so I think that if that's your mental model, then COVID is not helpful and some normalization whether through the abatement of COVID or through kind of new processes that are able to handle that better internationally maybe like a major prerequisite for OEMs starting to get production volumes up materially and then that would, at least, in theory, alleviate some of the pressure on used car prices. But I basically think that's been 6 months away for the last 18 months, and I don't know that we really -- or anyone else really knows the answer there.

Yes. You touched on GPU a little bit, poised to end the year nicely above $4,000, which is at least $1,000 per car improvement versus last year. So first question for the worrywarts out there. Can you talk us through the structural versus external drivers of that $1,000 per unit improvement? And do you think the current level is sustainable next year?

So again, I think we want to stay away from guidance for next year. And I think we want to be careful to not be overconfident in our precise reads of what's going on because I do think the environment is incredibly unique, and therefore, like your certainty around assessing exactly what's happening should just be a little lower than it otherwise would be. But I think that it's useful to look at what others are doing because I think that the most important attribute in our opinion of this market is that it's driven more by structural forces than by macro forces. And the structural forces are tens of thousand dealers with similar processes, similar cost structures, who generally also don't have cash reserves to be able to absorb losses, they kind of have to coordinate in similar ways as a result of kind of shared economic fate. And so once that's your model for what matters, the big question is just how are you performing versus others. When you're thinking about your gross profit, how is your gross profit per unit versus others is the most important question. When you think about your expenses, how are your expenses versus others is the most important question. And so I think through that lens, we think that, that may be one way to define what structural is, is progress versus others is structural and shared progress may be more circumstantial. The next question there then is who do you compare to because there are different classes of dealers that have fared very differently over the last year or 2. And I think there's a lot of things that are different and different classes of dealers, but I think probably the most important attribute that a class of dealers would have would be the channel through which they are acquiring cars. And I think that franchise dealers today have like this very nice channel of getting off lease cars that are valued incredibly well versus their residuals. And many customers know that and sell the car at a much higher price and many customers kind of drop off their keys and call it a day, and then that ends up being a really nice situation for those franchise dealers. And so I think they clearly have had some benefits. I think those that are reliant completely on auction have generally struggled as the wholesale prices have appreciated much faster than retail prices. So I think that's been difficult. And then I think for those that have kind of access to buying cars from customers and kind of a more diverse set of channels, for the most part, GPUs have been somewhat similar to years past. And so we think that that's probably the best group to compare ourselves to, and we're trying to assess our own progress that we think is structural. But as I said, I do think that uncertainty should be higher in light of the uniqueness of the environment than normal.

Got it. And as we think about your long-term GPU goal of 4,000 plus and all the long-term GPU drivers you've called out in the past, can you talk about where there's still upside on that current goal and then maybe where there's room to take the goal a little bit higher, given some of your newer partnerships and opportunities with Hertz and Vroom in the marketplace model? Can you talk us through that? I know you don't want to give any new groundbreaking guidance here. But qualitatively, I think that would be helpful.

What I would just say is I think the reason that we put out our long-term financial model. I think it was in late 2018, correct me if I'm wrong on that. There's like a variable question like the business has changed so much since that, right? Like a lot has changed, the business scale has changed. We've hopefully proven quite a bit. And so a lot has changed. I think it then begs a question of should that be updated in light of all that change. I think at a high level, we do believe that the most important like descriptive feature of the market is a structural thing that we talked about, which is tens of thousands of dealers. And so because of that, we think that there's a robustness in the way that we thought about what economics could be in 2018 and generally speaking, what we think economics can be on a go-forward basis because we think that economics are more a function of the structure of the industry than anything else. And the structure of the industry has not changed in supermaterial ways. So at a high level, I don't think we feel materially differently about what our long-term economics should look like than we did then. I think across every line item, we have room to get better and more efficient. I think there are certain areas that benefit from transaction size with finance, probably being one of those. So I think that there's -- depending on what happens with vehicle prices, which I think is highly uncertain. I think there's room for things to move around there. But in that line item, for example, there's clearly still room for us to get more efficient. And I think that's true basically in every single line item. So I wouldn't want to quantify it, but I think that there's not immaterial possible gains in all those line items, and we try to do the work to say like what do we think we can do here we're best-in-class or better? And so I think there's room there. And then there's -- there may be other opportunities in the medium and longer term just because customers have many needs when they buy a car and to the extent we do a great job, delivering great customer experiences and then we execute well across all of the things that we do for customers today and then that unlocks opportunities to do more for customers, there may be more opportunity, but I think that that's not something that we would want to bank on today because we don't have the proof points to point to that in a super robust way like we do with everything else. So I think that's why we try to stick with the model that we put out before.

Got you. We got a couple of questions from folks on the line rolling in. The first one I want to put out there is just how your inventory selection compares versus franchise dealers who have a higher mix of newer cars that are off lease or trade in? And that's the first part of the question, and I'll add, do you think partnerships with Hertz and other rental car companies helps unlock a higher portion of new cars for your site?

So I mean what I would say is I think we believe that selection is really important and is one of the areas where we're highly advantaged due to the fundamental of not selling the car in person and therefore, being able to list cars across many geographies for all customers. And we think it's one of the areas that generates positive feedback for us as we grow. And so I think versus any deal we have a very high-quality selection, almost regardless of how you cut it, I think versus like the full coverage universe of all possible vehicle SKUs depending on what your granularity of definition is there, I think we still think that there's a lot of room to continue to improve. So I think we view that as good news also, right? That's like -- that's the channel through which the positive feedback should kind of show up to the extent that we're able to continue growing and execute well. So I mean I think it's a big advantage for us. And I think it's hard probably to find a cut where we wouldn't be advantaged in inventory selection today. Now I do think franchise dealers have always had versus non-franchise dealers a subtle advantage in sourcing off-lease cars. Generally, the way those flows work and it's different for every OEM. But generally, the way those flows work is the customer returns the car to the kind of original selling dealer, and that's called the landing dealer. And then that dealer kind of can take a look at the car and decide if they want it. And then, if not, oftentimes those cars will go to a private kind of regional auction for franchise dealers of that brand. And then it will go to kind of the broader market. And so if you're not in that franchise flow, you're sitting kind of lower in the chain. On average, over history, that has not mattered. And generally speaking, independents have, in many cases, done significantly better financially over long periods of time. In today's world, that clearly does matter, where vehicles are worth 8,000 more than their residual. That's like a super helpful place to be. And at least so far, it seems like captives have been okay, kind of letting the dealers capture most of that excess. So I think there's like a subtle disadvantage there, but I don't know that subtle disadvantage is any different in expectation in the future than it's been on average in the past.

Got it. And then with respect to your pricing, historically, you've charged about $1,000 below your competitors or your brick-and-mortar competitors on like-for-like vehicles. Is that still the case? And when you think about the current price environment, how does that impact the pricing strategy?

Sure. So I think at a high level, yes, it's like the simplest answer to that. We sell cars at a lower price. We don't charge dock fees. We don't sell cars with accidents. You start looking at all those kind of different value adds. And in general, that means our customers are saving on the order of $1,000 per car they buy for a like-for-like car. I do think like especially over the last year that there may have been a little bit more fluctuation in that -- potentially in both directions than normal as different dealers express different views of what might happen to car prices because normally, predicting a car price over the next month or 2 is like not a super controversial thing, and therefore, there's shared views of that. Over the last year, there may have been more fluctuation in that across different groups. But generally speaking, I think that's the right way to think about it. It's a similar place to where it's been for a while.

Got it. We're just about out of time, but just to have a little fun. I want to do a lightning round of would you rather questions.

We can't do this publicly? Okay. I have no idea where you're going. I'm excited.

So first question, would you rather sell 70,000 cars next year and breakeven on EBITDA or sell 600,000 units with a modest EBITDA profit?

Units. And I would just go back to our prioritization structure there, and our view that the long-term economics are pretty clear. So expressing those economics more fully, more quickly is not necessarily hugely valuable in the long run.

Got it. And then second one, would you rather see e-comm adoption at 50% of the category over time with a handful of leading players or would you rather see e-com represent half of that 20%, 30% of the category with 1 liter being you?

Can we pick 100% just us? Is that one of the -- I would say -- what I would say is I think e-comm -- I think there's a very good chance that e-comm penetration for a while is a function of the ability to handle the operational chain. I think like to even say what e-comm penetration is today, I think, has little to nothing to do with what demand for e-comm in automotive retailers and almost everything to do with just what the capacity is built out. So I think we need to -- our job is to keep delivering great experiences that will get people that today might not be open-minded to transact in that way to sort of realize like, I can save money, I can have a broader selection, I can have a great experience, I can return it in 7 days if I if I want to, why wouldn't I do that? And then it's to build out the operational capacity as fast as we possibly can. And I think if we focus on us and we stay ambitious and we keep trying to run fast, I think that we'll be in a good spot. And I think that there's a chance that e-comm penetration in the long run can be an exciting number.

Perfect. Well, let's stop it there. Ernie and Mike, thanks so much for joining us.

Thank you. Appreciate it.

And everybody on the line, thanks for joining us as well and enjoy the rest of the afternoon.