Good day, and welcome to the Carvana Second Quarter 2024 Earnings Call. [Operator Instructions] Please note, this event is being recorded. I would now like to turn the conference over to Meg Kehan, Investor Relations. Please go ahead.

Thank you, Brenda. Good afternoon, ladies and gentlemen and thank you for joining us on Carvana's second quarter 2024 earnings conference call. Please note that this call will be simultaneously webcast on the Investor Relations section of the company's corporate website at investors.carvana.com. The second quarter Shareholder Letter is also posted on the IR website. Additionally, we posted a set of supplemental financial tables for Q2, which can be found on the Events and Presentations page of our IR website. Joining me on the call today are Ernie Garcia, Chief Executive Officer; and Mark Jenkins, Chief Financial Officer. Before we start, I would like to remind you that the following discussion contains forward-looking statements within the meaning of the federal securities laws, including, but not limited to, Carvana's market opportunities and future financial results that involve risks and uncertainties that may cause actual results to differ materially from those discussed here. A detailed discussion of the material factors that cause actual results to differ from forward-looking statements can be found in the Risk Factors section of Carvana's most recent Form 10-K and Form 10-Q. The forward-looking statements and risks in this conference call 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. Our commentary today will include non-GAAP financial metrics. Unless otherwise specified, all references to GPU and SG&A will be to non-GAAP metrics, and all references to EBITDA will be to adjusted EBITDA. Reconciliations between GAAP and non-GAAP metrics for our reported results can be found in our Shareholder Letter issued today, a copy of which can be found on our IR website. And with that said, I'd like to turn the call over to Ernie Garcia. Ernie?

Thanks, Meg, and thanks, everyone, for joining the call. The second quarter was another landmark quarter for Carvana. In the first quarter of this year, we were both the fastest-growing and most profitable public automotive retailer for the first time. In the second quarter, we did it again. And this time, we extended our separation in each category. Historically, when those two things are true at the same time, it bodes very well for an extremely successful future. We have every intention of working hard to validate that pattern. And in this case, a promising pattern is paired with tremendous room to run. We are a company with just a 1% market share in a $1 trillion industry with highly fragmented competition and barriers to entry that have recently been proven to be extremely hard to overcome. Our position is unique. And it is important to remember why. We are in this position today because 10 years and $10 billion ago, we set out to build an entirely new way to buy and sell cars. We thought through everything from scratch, starting with what our customers wanted, and then pairing that with what we believe was possible with new technology and new operations. We also discarded what others told us was impossible and all the reasons they provided us. We were stubborn and ambitious. I'm grateful for both. And I'm also grateful that we had no idea how hard it would be to get to this point. Being right about outcome and wrong about the path may be the most productive combination there is. From here, we believe the outcome is clear and exciting. And we are weather-worn enough to know the path will be harder than we think, but we are still ambitious and stubborn enough to keep pushing…

Thank you, Ernie, and thank you all for joining us today. The second quarter was an exceptional quarter for Carvana and reinforce the significant and sustainable progress we have made and continue to make in our current multiyear phase of driving profitable growth. For the second consecutive quarter, we generated positive net income and we set new company records for adjusted EBITDA, adjusted EBITDA margin and GAAP operating income. For the first time, quarterly adjusted EBITDA margin approached the midpoint of our long-term financial model EBITDA margin range of 8% to 13.5%, and we see meaningful opportunities for fundamental gains to drive towards the higher end of that range over time. Moving to our second quarter results. Unless otherwise noted, all comparisons will be on a year-over-year basis. Q2 again demonstrated the strength of our differentiated business model. Retail units sold increased 33% despite continued focus on unit economics and profitability initiatives as the strong demand we experienced in Q1 continued into Q2. Revenue increased by 15%. Revenue grew less than retail units, primarily due to industry-wide declines in retail and wholesale vehicle average selling prices. In Q2, our operational teams focused on increasing production capacity to increase selection to more optimal levels for our customers. The teams met their production targets in the quarter, but we still remain below our target available website inventory due to continued strong demand. In the near term, we will continue to increase production across the country. Our strong profitability results in Q2 were driven by meaningful fundamental improvements in GPU and SG&A expenses. In the second quarter, non-GAAP total GPU was $7,344, an increase of $314 and a new company record. Non-GAAP retail GPU was $3,539, an increase of $677 and a new company record. Our strength in retail GPU continues to be…

[Operator Instructions] The first question comes from Chris Pierce from Needham. Please go ahead.

Chris, you there?

Sorry. The next question comes from Michael Montani from Evercore ISI. Please go ahead.

Hi. Congrats on the quarter. And thanks for taking the question. Just wanted to ask if I could. Can you discuss a little bit the trends that you're seeing for the 100K plus income consumer versus those who are below? And then also, Ernie, can you give us some insight into thoughts around credit tightening from here? Or do you feel like you're at an appropriate level now?

Sure. I think from a demographics perspective, I think, clearly, affordability was impacted pretty heavily over the last couple of years. I think there's good news there, though. We have seen kind of higher levels of depreciation over the last 1.5 years. I think relative to CPI, car prices are now only probably about 3% higher than they were pre-pandemic. So I think we've closed a lot of that gap. Rates are obviously higher. So if you look at payments, payments are still about 10% higher than they were pre-pandemic for a similar car. So there's probably a little bit of room for that to continue to improve. And that, of course, is impacting the lower end of the demographic spectrum probably more than the higher income end. But I don't think there's anything too notable to call out there. I think we're just focused on continuing to buy the cars that our customers are demanding on our site, getting those up, getting those reconditioned, delivering them to customers and giving them great experiences. And I think that's what's driving our success right now without too much focus on one group or another. From a credit perspective, I think credit clearly was kind of slowly moving back toward pre-pandemic normal after being very good in '20 and '21, and then probably crossover was a little bit worse in '22 and parts of '23. And I think many lenders ourselves included, started to tighten credit in the fourth quarter of '23. What we've seen so far from that is performance that's definitely in line with what we would have hoped to see. So I think at this time, it doesn't feel like there are other moves that would be super material. But obviously, we're paying attention and we may adjust that over time.

Thank you.

The next question comes from Adam Jonas from Morgan Stanley. Please go ahead.

Thanks, Ernie, I never say congratulations on a quarterly call. And believe me, you don't want me to. But I mean...

I think we do want you to actually if we...

No. Believe me, you don't. Wow, dude. It's a hard wow. All right. So Carvana grew retail units 33%, with mostly flat SG&A, right?

Correct.

CarMax units fell 3% and had SG&A rise 13% year-on-year. Again, I don't want schadenfreude aside, what's the forward view on SG&A dollars on a per unit basis or either gross dollars or per unit, like how much longer can Carvana [technical difficulty] while keeping SG&A flat? Because the facts that you're mentioning, the capacity that you're burdened with, it might give the impression that you can keep SG&A really relatively unchanged to the dollar amount, but I don't think you want us to actually believe that. Or do you? That's my first question.

We've broken down SG&A in the past into three categories: overhead, marketing and operations expense. And I think it's probably useful to hit them all separately. On overhead, we've held that pretty flat for the last five or six quarters. And I think that that's largely our plan for the immediate future, is to continue to hold that flat. On marketing expense, we've moved down from a kind of long-term average just over $1,000 to 542 this quarter. That's something that we're extremely excited about, obviously, and that reflects a lot of gains across, I think, many different parts of the system. I think from a long-term fundamental perspective, I think debatably the most exciting part of that is what's going on with operations expense. We, this quarter, we're at $1,696 that's a number that we're extremely excited about. It's a number that's several hundred dollars less than our kind of pre-pandemic numbers that were pretty good at the time. But that's happened at a time when the average of other automotive retailers have probably seen their SG&A go up by about $1,000 with inflation and everything else. So I think the gains look like a couple of hundred dollars there, but I think inflation-adjusted and, most importantly, I think, relative to what else is out there, I think those gains are likely over $1,000. And then further, I think that inside of that operation expense, there's a warranty line item that today is around 350 and was kind of closer to 150 to 200 pre-pandemic when inflation rates were quite a bit lower. And so I think when you look at kind of the operations expense ex warranty, the gains versus pre-pandemic are more like $400 per unit, again, at a time when costs across the industry have gone…

Thanks, Ernie. Just a follow-up. If I take the midpoint of your guide, that would imply the second half quarterly run rate EBITDA of maybe $100 million, I'm rounding here, Ernie, bear with me, about $100 million less than what you did, kind of a mid-200 number rather than a 355 number. So what would you say is kind of baked in there, including seasonality and then other stuff beyond seasonality that might have maybe made the 355 a little on -- and again, maybe I'm stupid for using the midpoint of your guide, and you're being very conservative. But just if it was a genuine guide, what's embedded and why is $100 million lower run rate? Thanks.

Yes, sir. So I think first and most importantly, we see gains in front of us, in growth and in the various fundamental areas we just discussed, and we're going to go out and try to get them. And I think that that's going to be something that we're going to continually do over the next several years. So I think that's probably the most important thing. I think as it relates to the guide, I think you're bringing up all the appropriate points and doing all the math correctly. We did sell a little bit more in loan production this quarter than we originated. So we provided a bridge in the Shareholder Letter that that was probably about 0.4%. So probably a more normalized number would have been about 10%, in dollars that would have been about $12 million give or take. So that positively impacted the quarter. And then there is seasonality heading into the back half of the year. But obviously, we're extremely excited with the results we just had. We're extremely excited about the outlook for the rest of this year. And I think most importantly, we're just extremely excited about the opportunity that we've got, because I think part of the goal of building out this dream is to take something that was very nonobvious and slowly but surely turn into something that's obvious. And we think this quarter is probably the biggest discontinuous step in making what we're trying to achieve, obviously, we've ever taken. So we're very excited about it.

Thanks, Ernie. It's impressive.

Thank you.

The next question comes from Chris Pierce is with Needham. Please go ahead.

Hi. Can you guys hear me this time?

We can.

All right. Perfect. I just wanted to kind of - can you talk about where the upper bound of retail GPU might be? And can you kind of touch-base on the standalone reconditioning? I guess, CarMax was just mentioned the previous question. We were kind of told that $2,200 in retail GPU was the efficient frontier as far as units. But that was in a, hey, the conditioning centers in the store type of model. So like what's the right way to think about where retail GP can go? And how excited are you as far as pushing retail GPU even higher from here?

Sure. Well, so first, I would say we believe there are significant fundamental gains to be had in every GPU line item, retail GPU included. And so we think that we can deliver to customers the same quality of offering and cause retail GPU to go up. I think it's important to reevaluate what GPU might mean in this environment versus pre-pandemic. And I think a lot of heuristics over time have been established pre-pandemic that maybe warrant some revisiting now. And so for perspective on that, I spoke in the last question about how most dealers have seen SG&A per unit go up by about $1,000. I think the right kind of first order mental model for how the automotive industry works is that, generally speaking, it's cost plus. Many dealers share a cost structure and they kind of -- they're at auction, holding up their hand to buy a car at a certain price, and they're pricing it with a profit in mind, and many are acting in similar ways. So if you look at the average retail GPU across many automotive retailers now versus pre-pandemic, it's also up around $1,000, as like a good first order estimate. And so I think debatably, kind of all the heuristics from before, you could kind of move up by $1,000. And I think that that's also buttressed by just looking at wholesale retail spreads today, which is basically kind of you're just looking at the aggregate market that is impacted by the sum of dealers. It probably is supporting about $1,000 more in retail GPU than it was pre-pandemic. So I think all the heuristics -- first order, I think you kind of moved them by roughly $1,000. And I think that's helpful in explaining the significant changes that you've…

Okay. And then on adding production, Mark's comments. Should we think about that as adding a second shift in facilities? Or based on kind of what you talked about in Rocklin, you have production capacity that can increase as you increase the vehicles flowing through the one shift that you have now? Or like what's the right way to think about what could happen to the IRC staffing?

Sure. Yes. I think there -- I think we have a lot of flexibility to increase production. I think that takes three primary forms. So first, adding -- continue to add lines in existing Carvana facilities that have excess capacity to add lines. I would say it's category number one, and we definitely have capacity to continue to do that. Category number two would be more what you suggested, which is also a slightly different form of adding lines, which is adding shifts to IRCs where all the production lines during the day shift are full, you can add a second shift to increase production capacity. And then the third that I would layer in is increasing production at ADESA locations. So we've had a lot of success adding Carvana reconditioning software and processes, in two locations so far, in Buffalo and Portland. We plan to integrate Carvana auto production into a third location in Kansas City that we call the Megasite, which is a site where we are continuing to serve very effectively physical auction customers at the site, while also implementing Carvana reconditioning processes and systems. And so integrating more ADESA sites with the ongoing physical auction services, combined with Carvana reconditioning, is the third avenue for production growth that we're excited about. So all of those together give us a lot of flexibility. I think one of the things we're excited about in this next phase of growth is we do have more flexibility than we've ever had before for growing production efficiently, at the right cost, at the right level of quality over time.

Okay. Thank you, both.

Thank you.

The next question comes from Brian Nagel do from Oppenheimer. Please go ahead.

Hi. Good afternoon. I want to add my congratulations. Nice quarter here.

Thank you. Appreciate it.

So the question I have, just to understand better, I guess my first question, maybe to understand better the demand dynamic and the sales trajectory for Carvana. You've talked about still, I guess, a lack of a reconditioned -- a limited supply of reconditioned cars holding back sales. So the question I have, as you look at the business, I mean - how much is that number? If your reconditioning were operating where you wanted to, what would sales be running at?

Sure. So I think we talked last quarter about kind of moving into this transition to growth. And I think that that's where we remain. I think in the first quarter we sold more cars than we anticipated. That caused our inventory to be a little lighter than we expected. We turned up our production plans, and the operating teams did a great job, and we're able to achieve that. That enabled us to sell more cars than we expected again in the second quarter. But we did build our inventory much less than we would have liked to. So I think we're continuing to lean in that direction. We haven't invested in marketing yet. There's -- our marketing expense relative to our gross profits leaves a lot of room. And so I think we kind of put this outline in the Shareholder Letter of the things that we believe have driven our growth since the beginning, which we think have been consistent but have just shown up in kind of different weights across time. And that's continuously improving customer experience, that's highly differentiated, increasing awareness, understanding and trust. I think those are three different steps to brand, but every step matters there. And then increasing inventory selection and other scale benefits. And I think that so far, we've kind of gotten to a place where we've gone from negative effects in the second and third area over the last couple of years as we were shrinking inventory and as we were investing less in marketing and as we were getting beat up a little bit in the press, to more like still wins, but those still wins still have enabled pretty exciting levels of growth. So I think this transition period is about figuring out where exactly that healthy…

That's very helpful. My follow-up question, just with respect to ADESA, I mean one of the questions we get a lot is you talk to clients about carbon, just how should we think about the timing of the conversion or the, I guess, reconfigurations, if you will, some of these ADESA facilities to be reconditioning type centers for you?

So I think that that's also something that we're figuring out in this transition period. I think Mark spoke about our next site in Kansas City, which is going to be a site where we're calling it a Megasite. It's going to have all the capabilities at once. We've done it very efficiently so far with almost no CapEx. I think the more of these sites that we choose to open, there are many more that we could integrate and not have a huge CapEx investment. But I think the more of them that we choose to open, the more that that will lean a little bit in the direction of CapEx. But a lot of other offsetting benefits. You get access to a lot more inventory pools, you get closer to your customers, your inbound transport costs are less, your outbound transport costs are less, your transport times are faster, which leads to higher conversion. And so, I think we're trying to find the balance right now between adding inventory pools and logistics capabilities to those sites and being really efficient. And so far, that's led to of adding reconditioning capabilities in Buffalo and Portland and Kansas City. I think over time, we will certainly do more. And I think the speed at which we do that is something that we're figuring out as we go right now.

Got it. Right. Appreciate the color. Thank you.

Thank you.

The next question comes from Michael Baker from D.A. Davidson. Please go ahead.

Okay. Thanks. Two questions. One, you talked about giving back to customers, giving some of the profit back to customers. That show up in sharper pricing, more choice, faster speed of service, all of the above? And then second question, do you think you gained any share this quarter from the - from some of your competitors being impacted by CDK global cyberattack? And if so, how much? And does that mean share gain would be a little bit less going forward? Thanks.

Sure. I think that - so I think - we provided in the bridge and we put in the Shareholder Letter, it gives you a walk from kind of where we are to where we expect leverage to be and how we're doing from a marketing perspective in some of our more mature markets. And I think that gives a pretty clear walk, to the very high end of our long-term financial model from an EBITDA perspective without the need for fundamental gains. And then we do believe there are very significant fundamental gains to be had. And I think the areas where we could invest them were all of the areas that you suggested, and a couple more. And I think as we move forward, we will try to do that as intelligently as we can. I'm not sure we want to tip our hat too much on that, because I think we've got some creative ideas. And I think there's also some lessons that we've got to learn between here and there as we start to give that back to see where it's most efficient. But we - as long as we execute, we anticipate having gains to give back. And then, we'll try to do that very intelligently. As it relates to CDK, I think there's no question that that was an impact the industry broadly. I think there were significant impacts on many automotive retailers. The impact to us, I think our best guess at that is that it was pretty muted. We didn't see huge impacts either when it started, when it was ongoing or as it was alleviated. Maybe there were a little directional things that you could have picked out, but nothing that I think warrants mentioned. So I'm not sure there was a huge impact there.

Okay. Thank you. Appreciate the color.

Thank you.

The next question comes from Sharon Zackfia from William Blair. Please go ahead.

Hi, good afternoon.

Hi Sharon.

I think, hi, I think one of the underlying questions as you're kind of balancing profitability with growth, is just kind of how much unmet demand you're leaving on the table. So when we think about kind of inventory and being under-inventoried, obviously, when you're selling non-commodity items, that's constraining sales, right? So I'd assume conversion is a bit below, where you would like it to be. It sounds like there's going to be investments in things like recon. And at some point, I expect marketing to kick up again. Can you talk about kind of where underlying demand is relative to your sales? And it does feel like you are in growth mode now, just given the sales metrics that you're putting up, what kind of testing and iteration are you doing to think about things like turning back on marketing to a greater extent?

Sure. So I think - I'm not sure my answer is going to be as satisfying as you would like. I think all of your observations are correct. I think we're in this transition period and I think we're moving in that direction. As discussed, we're trying to balance growth and its benefits with the benefits of being able to keep things a little more stable and continue to make gains. I think there's no question that we are very, very small compared to what we want to be, at any point in time, measuring the exact amount of latent demand is complicated because I think you even have to frame what that means kind of carefully. If you believe you have a business model with positive feedback, then as you unlock more demand, kind of that more demand showing up leads to things like faster delivery time and broader selection and unlocks more demand again. So I think our eyes are on selling millions of cars and being the largest, most profitable automotive retailer, and doing it as quickly and intelligently as we possibly can. We think things look very, very good right now, and it's very, very exciting. But we always remain aware that there's always bumps along the way as well. And so, I don't know. Things feel very good. They feel very good, and I think we just got to keep moving forward, and we'll see where it takes us.

You're right, Ernie, it was unsatisfying as an answer. So I will ask something maybe...

Nice that you call me out on that.

Maybe for Mark instead. Seasonality was mentioned earlier, and I just want to level set kind of GPU expectations. I mean typically, we would expect lower sequential GPU in the third quarter. Is that in keeping with what you're thinking? Thanks.

So on the question about seasonality, I think where we typically think of seasonality as being most prominent is really in the fourth quarter, first and foremost, and then maybe to an extent in the early first quarter as well. I think the form that that seasonality takes is typically softer overall used car market demand as well as higher depreciation rates. And so that's industry seasonality that we see year-after-year. And I think where we see that most acutely is in the fourth quarter. and sort of the early part of the first quarter.

Thank you.

The next question comes from Seth Basham from Wedbush Securities. Please go ahead.

Thanks a lot and good afternoon. My question is around other GPU. If you could give us some sense of where you are today but also where do you think it could be in a couple of years, and the key drivers to get there, that would be helpful.

Sure. I can take that one. So I do think we've had a strong quarter -- we've had a strong first half on other GPU. I think there are many drivers of that, including all the work that the teams do internally to optimize our platform, scoring, pricing, monetization, as well as we've -- I think, over time, had some nice gains in the ancillary products part of that as well. I think as we look forward for opportunities, we do see opportunities to make fundamental gains in other GPU. I think some of the areas where we see opportunities are really the areas that I just listed. So we're always seeking to take in more data and improve our scoring and pricing algorithms. We're always seeking to prove and streamline the customer experience, to make using Carvana financing as easy as possible. We're always working to develop the most efficient funding sources for our loans. We're always looking to continue to improve attach rates on our ancillary products. And so, those are areas where we've made gains historically, but absolutely areas where the teams are focused on continuing to make fundamental gains over time.

That's helpful. Direction from here, could we see a step back, because of some of the timing elements associated with the strength you recently experienced, before you get to this much higher level that you expect in a couple of years?

The callout that we had on Q2, I think Ernie mentioned as well, we saw about a $12 million favorable impact just due to timing effect - loan sale timing effects that rounded up to 0.4% of revenue. So that would be the primary thing that I would call out as it relates to Q2.

Thank you.

The next question comes from Nick Jones from Citizens JMP. Please go ahead.

Thanks for taking the questions. You've delivered a lot of efficiency gains over the last year. You're still talking a lot about a bunch more to be made as you try to deliver a much larger number of units than you're doing today. How are you balancing kind of this relentless focus on efficiency gains given the progress you've made versus, I guess, potentially new revenue opportunities that it sounds like you maybe had ambitions if we go back to the Analyst Day years ago? As Carvana gets bigger, delivers this kind of new consumer experience that's differentiated, are consumers going to kind of increasingly look to Carvana for additional services or solutions? And is that a factor to unlock kind of the volume you aim to get long term? Thanks.

Sure. So I mean I think one of the lessons that we learned over the last couple of years is that focus is very valuable. And I think one of the things we learned about ourselves, I think we've assembled a team that is very ambitious and seeks to take on a lot of things. And I think that one of the battles that we constantly face internally, I think, is trying to make sure that we tackle the right number of things and we prioritize them properly. I think given the creativity that we have inside of the walls of Carvana, I think we're never short ideas. That's never the case. And I think just trying to make sure that we're focusing on the right things in the right order to make the most progress that we can as quickly as possible is what we focus on. I think that that extends to efficiencies, growth and other opportunities where we believe they are exciting areas, all three of those are very exciting areas. So, I think we're trying to be as smart as we can there. Half of saying this out loud is a reminder to ourselves to try to stay focused, because I think we remain in a place where there are more opportunities than we should intelligently take on, and we're trying to be as thoughtful as we can about which we take on and how.

And maybe a follow-up, thanks for that. And I guess on CarMax, has come up a couple of times on the call, I think historically, competitors have maybe not been believers in Carvana's capabilities, do you say there might be a different urgency and a competitive reaction given your kind of recent sustainable results? Thanks.

Sure. So I mean I'll respond to that maybe just generally. I mean I think the view that we've always had and we've tried to continually share is that, as it relates to ourselves, we try to not be too focused on competition. We try to be focused on our customers. I think it gets very easy for companies to follow each other in circles and believe that what the other company is doing is smart. And then just kind of continually chase each other instead of listening to their customers. So what we try to do is listen to our customers. That's our continual goal, and we will continue to do that. I think that there's no question that whenever results are being put up that are high quality, people are going to take notice. And I think that will be true of probably many in the industry. And that's fine. I think we never expected to be alone or be handed anything. We expected to come out here and compete for it and try to build something differentiated for our customers. And I think that's what we intend to continue to do. I think something that - putting on our very biased Carvana hats, something that we're just extremely excited about is we think that the things that we've built are really differentiated and take a long time and a lot of effort and are associated with a lot of risk. I think there are many examples, 10 to a dozen international examples of companies that have sought to do something very similar to us and put big dollars behind it. And I think it's pretty clear at this point that we are, by far and away, the most successful of those big swings, and success in the middle looked like down 99%. And so I think it's - this takes a lot to build. And I think that you're never competitively alone. But I think the degree to which you are competitively differentiated is a function of time, money, effort and difficulty. And we think there's a lot that separates us there. And so we're excited by that. But we will, in no way, shape or form be complacent. We're going to keep trying to put more space between us and everyone else, and keep delivering for our customers.

Thanks Ernie

Thank you.

The next question comes from John Colantuoni from Jefferies. Please go ahead.

Great. Thanks for taking my questions. Just wanted to go back to Ernie's comment about passing additional gains on to the consumer. Can you just detail what you mean by that comment? I'm probably thinking about this wrong, but it sounded to me like you might - maybe you expect GPUs to be more consistent with what they are today, and over time, and then you can use any additional efficiency gains to drive faster growth. But please help clarify that for me. And then second question, the comment about the industry supporting, I think it was - the comment was the industry can now support 1,000 higher GPU. Can you just talk about what specifically is driving that and whether those underlying drivers are transitory or sustainable? Thanks.

Sure. So I think on your first question, I think you interpreted it correctly. I think what we define as a fundamental gain is getting $1 more efficient in any given function. And that means then you have this question of do you keep profitability the same and pass the dollar back to customers in the form of lower price or lower rate, or higher bids on their car, or more investment in their experience in whatever form you want to do? And I think - or you can obviously have that to show up as higher profits. We anticipate in light of the scale of the additional fundamental gains that we see opportunity for from here, we anticipate that a significant portion of that will be passed back to customers. And we'll seek to do that as intelligently as we can. On the GPU comment, I think the point that we're just simply trying to make there is. I think the mental model that would have predicted the way things have turned out over the last four or five very volatile years the best would have been that automotive retail is largely a cost-plus business. And the costs are what do cars cost at wholesale plus what are all the operating cost to deliver to a customer an experience where they get a car at the end of it? And given that we've seen a lot of inflation in many areas of automotive retail, and many of the different automotive retailers out there have expenses that are on the order of $1,000 higher today, it would stand to reason with that mental model that you would expect $1,000 higher retail GPUs. And that is approximately what we observed, if you look at the average of many public retailers or if you look at the wholesale retail spreads. And so, it seems as if there has been inflating costs over the last several years, and then those inflating costs are being passed through to higher gross margins that are leading to somewhat similar EBITDA or operating or like EPS margins. And so I think that was the point that we're trying to make there.

The next question comes from Marvin Fong from BTIG. Please go ahead.

Good evening. Thanks for taking my questions. Congratulations on the results. So apologies if this was asked before, I joined the call late. But in talking to some investors, I think last quarter, you highlighted how quickly vehicles were selling on the site, and that kind of limited depreciation and helped GPUs that way. And I just thought I'd ask about do cars continue to sell ahead of the low 60-day kind of days to sale that you're targeting? Or has that normalized? And then second question, just thinking about your ability to grow volumes, well aware that you have the capacity. But are there any gating factors, whether it's hiring personnel or just getting the reconditioning throughput that would limit you from kind of achieving sequential unit growth better than what you achieved in the second quarter? I mean could we see unit sales go up? Or is there any limiting factor that's preventing you from selling 15,000 units a quarter, or more quarter-over-quarter or along those lines? Thanks.

Sure. I can take the first one. So average days sale was again below sort of our typical target range in Q2. It was also below that range in Q1. I do think we'll - over time, as we onboard production capacity, I think our goal continues to be in our target range versus tracking below that. And I think the reason for that is we prefer to make a bit more selection available to customers on the site than what we had in Q1 and Q2.

And then on your second question, I mean, I think you listed many of the gating factors. I think we buy cars from customers from auction, from partners. We transport them to our reconditioning centers. We recondition them through our reconditioning process there. Customers go on our website. We answer the questions that they've got. That means that we're answering phone calls. We're handling different digital tools to resolve their questions. We attach financing. That means we have a verification function that we go through, to make sure that everything associated with the loan is taken care of. We've got title and registration. We've got a delivery long leg and last mile. Those are the various operating groups. And I think every one of those groups, we scale up together as we grow. I think I made the point earlier that, in Q1, we grew more than we anticipated. That means that all of the groups were not positioned for the level of sales that we saw, but we started to see the demand and we started leaning in that direction. And all those groups simultaneously scaled up. And they did so in a way that was very efficient, and allowed not only our fixed expenses delever, but also our variable expenses to go down. The same thing was true again in the second quarter. We sold more cars than we expected. That means that all those same operating teams were given higher targets, and they went out and they got that done. So, I think that that's exciting. I think that, that's the continual march of building a complex machine that has lots of people and moving pieces and physical space associated with it. But I think that's something that we've demonstrated that we know how to do over a pretty long period of time. Took a little break there for 18 months. But I think for most of our company's life, that's something that we've really demonstrated we know how to do. So, I think, we'll continue to do that as we continue this transition period.

That's great color. Thanks guys.

Thank you.

The next question comes from Doug Arthur with Huber Research Partners. Please go ahead.

Yes, thanks. I just wanted to go back to the really strong retail GPU number. You list 4 or 5 factors that are pushing that up. And I would assume that the -- I guess, question 1 is, how significant are the higher spreads between wholesale and retail market prices currently in that equation? And I know you're buying a lot from your customers, how much can you determine that over time as the market moves around price-wise?

Sure. So I think I would start with kind of the same figure that we discussed earlier, which is I do think that the market is supporting today approximately $1,000 more spread between wholesale and retail. And that's defined specifically as auction prices relative to average prices offered at -- across many dealers. So I think our explanation for that is that dealer costs have gone up, and it's a cost-plus business. Now I think as we grow our business, I think we have many different avenues that we buy cars from a channel perspective. We buy them directly from customers; that's the majority. We buy many from auction. And then we do have partner inventory as well with partners like Hertz where we're getting inventory. And I think then we also kind of, in geography, we buy those cars in different places. And we try to be intelligent about the way that we mix across channel and across location, to go buy those cars as efficiently as possible. That's another area where we think that there are gains to be had. We think there are many benefits that are sort of inherent, to our business model of having a very large nationally available inventory. And we think those show up in many ways. They show up in the breadth of cars that we can buy. They show up in the ways that we can intelligently price those cars given the data that we see. They show up in the ways that we can buy those cars across channel and space. And so I think we're trying to continually get better in all those different areas. But I think our - as far as prediction goes. We think that, generally speaking, the idea that automotive retailers cost' plus has been very predictive over a very long period of time with a lot of volatility, it also just kind of makes sense. And so absent big changes in average dealer expenses, I think our expectation would be that profit margins available on vehicles would be pretty similar over time.

Okay. Great. Thank you very much.

Thank you.

The next question comes from Rajat Gupta with JPMorgan. Please go ahead.

Great. Thanks for taking my questions. I just had one question and one follow-up. Firstly, on seasonality, there have been a couple of questions earlier. If you look at the business from here, excluding 2022, 2023 when you were going through a period of restructuring, you've always grown units, 2Q to 3Q, 3Q to 4Q. And so I'm curious, like, given you're adding production now, you're hiring a lot of technicians, you're hiring market ops, fulfillment standard employees, why would the business, and given where your share is in the market, why should we still expect seasonality, at least on the units, I mean we understand the GPU seasonality, but why should there be seasonality in the units from here for Carvana. And I have a follow-up? Thanks.

Sure. Well, I think we provide our outlook, so we're going to stick with that. I think seasonality, is definitely an industry-wide thing. And then I think we're obviously focused on continually leaning through this transition to growth into more growth, and I think we'll try to be thoughtful on how we do that and the speed at which we do it. But I think we have to stick with our guidance on that one. So apologies for not being more helpful on that question.

No worries. And another one, just on the long-term target. Obviously, a lot of discussion around the cost-plus nature around the GPUs, the $1,000. We fully understand that. But curious with the experience over the last two years within the business, are there other areas where you think the inflation of pre-pandemic that informed your long-term targets are no longer relevant? Maybe areas within SG&A, like advertising, or other areas that you might want to call out that might have changed? Thanks.

I think first order - I think the simplest thing is just that I think the whole business model is kind of inflated, with inflation, is like a reasonable way to think about it. I think we've put out that long-term model five or six years ago, the reason that we were able to put that out and it's been fairly accurate is, because this is a very mature industry and so there's a lot of data that we could look at to see where, were the various pockets of profitability. And we were able to reasonably accurately build up what we thought expenses would be. And so I think from a long-term perspective, I'm not sure that the model has changed all that much. It's certainly inflated and I think, most specifically, it's inflated in retail GPU. I think the next thing that I would say is we acquired ADESA. And we vertically integrated versus what was anticipated at that time. So that's another change. But otherwise, I think the fundamentals of the market are pretty similar, right? Competitive dynamics are still pretty similar. Consumer options are still pretty similar. So we think that that remains a reasonable way to think about it. I think in terms of our own execution, I think this last year has been - well, the last two years really, have just been very, I think, exciting in terms of the speed at which we've been able to execute. And I think also the additional opportunities you see every time you take a step forward, you tend to see a couple of more opportunities. And so, I think that there's a lot of exciting opportunities in front of us, but I'm not sure that we're at a place where that should change our view on kind of like what we think is an attainable long-term EBITDA margin too much for the future. We'll keep you updated if that changes. But I think high level, we remain in a similar spot, and we expect now to be at the high end of our long-term model. But I think we don't have a ton more updates than that to provide.

Understood. Great. Thanks for the color and good luck.

Thank you.

This concludes your question-and-answer session. I would like to turn the conference back over to Ernie Garcia for any closing remarks.

Great. Well, thanks, everyone, for taking the time to be on the call. Thank you to all of our shareholders out there. This is - you don't get to have that many quarters that are this fun and meaningful. And I think this is a pretty fun one for the team internally, and I hope it's a fun one for many of the shareholders that have stuck with us over the last couple of years. I know there's been some ups and downs. But we're excited from here. We hope that you're excited about these results and about the prospects that we have going forward. To everyone inside Carvana, I just cannot thank you guys enough. I hope that you are incredibly proud of the results that we've been able to put forward here. I think it's the coolest thing that I've ever been associated with, is getting beat up that bad by the whole world and then just coming together and fighting all the way back to this spot. And I hope we always remember that. I hope we always stay in that fighting stance. And I hope we keep proving people wrong, because it's been really fun, and I think that we can and I think that we should. So thank you to all of you. Take a moment to be proud. We'll see you tomorrow, we'll be back at it. Thanks, everyone.

This conference has now concluded. Thank you for attending today's presentation. You may now disconnect.