Copart Inc (CPRT) 2019 Q2 法說會逐字稿

Good day, everyone, and welcome to the Copart, Inc. Second Quarter Fiscal 2019 Earnings Call. Just a reminder, today's conference is being recorded.

For opening remarks and introduction, I would like to turn the call over to Mr. Jay Adair, Chief Executive Officer of Copart, Inc. Please go ahead, sir.

Thank you. Good morning, everyone, and welcome to the second quarter call for Copart. On the call with me today is Jeff Liaw, CFO; and Will Franklin, Executive Vice President. I'm going to turn it over to Jeff Liaw for opening comments and then Will Franklin will give us an update on operations. And then we'll be happy to answer any questions that we have at that time.

All right, thanks so much. Jeff?

Thanks, Jay. I'll start with the Safe Harbor. During today's call, we'll discuss certain non-GAAP measures, including non-GAAP net income per diluted share, which includes adjustments to reverse the effects of disposals of nonoperating assets, foreign currency-related gains and losses, the impact of income taxes on the deemed repatriation of foreign earnings, net of deferred tax ranges and certain income tax benefits related to accounting for stock option exercises.

We've provided a reconciliation of these non-GAAP financial measures to the most directly comparable GAAP measures on our website under the Investor Relations link and in our press release issued yesterday afternoon. We believe the presentation of these non-GAAP measures, together with our corresponding GAAP measures, is relevant in assessing Copart's business trends and financial performance. We analyze our results on both a GAAP and non-GAAP basis described above.

In addition, this call contains forward-looking statements within the meaning of federal securities laws, which are subject to substantial risks and uncertainties that could cause actual results to differ materially from those projected or implied by our statements and comments.

We do not undertake to update any forward-looking statements that may be made from time to time on our behalf. For a more complete discussion of the risks that could affect our business, please review the management's discussion and analysis portions in our related periodic reports filed with the SEC.

Now I'll turn our attention to the second quarter of our fiscal 2019. We're pleased with our operating results. I'll start off with a reminder that the first 6 months of fiscal '18 were distorted by Hurricane Harvey. Over those first 2 quarters of fiscal 2018, we incurred losses of nearly $10 million on an operating basis. Q2, in isolation, would reflect a gain because the first quarter of 2018 disproportionately captured storm-related costs, while the second quarter would disproportionately capture storm-related revenue. I'll make it a point over the course of this call to communicate our key metrics with and without the effects of Hurricane Harvey.

We achieved a record second quarter in revenue gross profit and operating income, starting with our nominal global revenue growth of 5.6% with an unfavorable year-over-year currency effect of $4.9 million from foreign operations, primarily due to the relative strength of the dollar in comparison to the pound and the Brazilian real. Excluding the effect of Hurricane Harvey, our revenue growth was 17% even.

Our global service revenue growth was 3.7% year-over-year and again, excluding Hurricane Harvey, was 13.7%. We typically suggest looking to service revenue and service revenue growth as the more accurate indicator of underlying business activity.

Our purchased car growth of 19.1% year-over-year, driven largely by our international businesses, split approximately equally between the U.K. and Germany. A quick reminder that our Copart-owned inventory remains relatively small at $28.3 million at quarter end, which is small in the context, of course, of the overall business.

Turning to unit sales. Our nominal global units declined slightly at 0.6% year-over-year with a slight U.S. unit decline of 3.7% and international unit growth of 18.7%. Again, excluding Hurricane Harvey, our global unit sales grew at 7.7%. U.S. units grew 5.7%. And excluding Harvey and charities volumes from that U.S. number, our volumes grew 7.5% year-over-year. Our U.S. unit growth was driven by both our insurance and noninsurance segments, which Will, will describe in greater detail in a few minutes.

Then turning to inventory, our nominal global inventory grew at 6.6% year-over-year. Excluding the effect of Hurricane Harvey, our inventory grew at 7.7%.

For the quarter, our gross profit grew 8.7% year-over-year. Excluding Harvey, that same number would be 13.6% year-over-year growth. Our gross margin rate increased from 41.7% a year ago to 42.9% this year, so approximately a 120-basis-point lift. Excluding the effect of Harvey, our gross margin compressed slightly, approximately 1%, which is explained almost entirely by the slight mix shift to purchased car revenue.

Our average selling prices for vehicles at Copart auctions in the United States, excluding Hurricane Harvey, grew 16.1% year-over-year. This -- pardon me, 15.4% year-over-year, that compares to 16.1% growth a year ago. So 15.4% this year and 16.1% a year ago for the same quarter. Will, again, will provide more context on this phenomenon, a reflection both of the type of cars that are consigned to Copart as well as the expansion and marketing efforts that we pursue with our Copart member base.

Turning to our general and administrative expenditures. Excluding stock compensation and depreciation, we're up from $29.7 million a year ago to $33.2 million this year. We're down $1.6 million sequentially versus the first quarter of 2019. Repeating a mantra that you've heard before, in general, G&A expenditures will vary from quarter-to-quarter and will grow over time with inflation and complexity. We continue to believe we can achieve operating leverage given the top line growth we have experienced.

Our GAAP operating income growth was 9.1% for the quarter. Excluding Hurricane Harvey, that same operating income would have grown at 15.5%. Our net interest expense you can see is down slightly year-over-year given our lower average net debt balance. Our other income of $4.8 million is largely a gain on sale of an asset, specifically a now replaced data center. You also see that adjusted out in our non-GAAP reconciliation.

Our second quarter income tax rate was 20.4%, a reflection of the lower U.S. federal tax rate we discussed on prior calls as well as onetime benefits from stock option exercises, which we again, reflect in the non-GAAP presentation. Our GAAP net income increased from $103.3 million a year ago to $131.4 million this year for the second quarter, an increase of 27.2% year-over-year.

On a non-GAAP basis, our net income grew 11.5% year-over-year. Excluding the effects of Hurricane Harvey and including an assumption for tax rates, that same growth rate would be 18% plus or minus for non-GAAP net income year-over-year.

The non-GAAP schedule, we've already talked about the major adjustments on that page for the second quarter for this year, which include both the disposal of nonoperating assets as well as the excess tax deduction for stock option exercises. I'll just quickly remind folks that the major adjustment for last year of the second quarter was a $10 million adjustment of the onetime transition tax charge that's a reflection of the tax reform effect in that quarter.

I'll turn our attention briefly to Germany, -- we would encourage -- before coming back then to the balance sheet cash flow, we'd encourage folks for further background to review the transcript of our first quarter earnings call where we've described in much greater detail the nature of the market and our efforts there, and it represented a onetime deep dive, so to speak, into the business. But as a quick substantive update, we now have 12 locations up and running in pursuit of a critical physical footprint across the country.

At Copart Germany, we're now running daily auctions across those locations with a strong majority of our volume sold to buyers outside of Germany. We think that reflects the power of the Copart brand name, our buyer network and technology platform and frankly, is a strong testament to the inefficiency of the current market model for total losses in Germany. Our unit sales in Copart Germany are approximately 8x the same volume for a year ago for the second quarter.

We also continue to demonstrate our ability to purchase cars through Wreck Online, the listing service that we own and to sell them at a positive margin at Copart Germany auctions. As you know, we also continue our dialogue in parallel with insurance carriers in Germany and believe that a Copart model, akin to what we have in the U.S. and in the U.K., is ultimately the right answer for that market as well for a host of reasons, including both total loss and claims cost for carriers as well as the claims experienced policyholders.

Turning to the balance sheet and the cash flow statement. Our operating cash flow for the quarter was $107.5 million with CapEx of -- a gross CapEx of $74.4 million. About 60% of the CapEx was attributable to capacity expansion and lease buyouts with the balance attributable to maintenance and other activities.

We also purchased 7.6 million shares of Copart stock in the open market at a weighted average price just below $48 for total outlays of approximately $365 million. We've funded these purchases with cash-on-hand and a revolver draw of $93 million as reflected at the end of the quarter. We still have available liquidity of more than $750 million.

With that, I'll turn the call over to EVP, Will Franklin.

Thank you, Jeff. Let me provide a few more comments about our operational performance for the second quarter.

Copart, once again, delivered another strong quarter. Our U.S. volume, when adjusted for the Harvey activity in the same quarter last year, grew by 5.7%. Our volume growth continued to be driven by organic growth and market wins within the insurance market and our continued expansion into the noninsurance markets.

Organic growth in the salvage market is driven, we believe, by an increase in total loss frequency as high repair costs and our elevated auction returns are leading to a higher percentage of claims resulting in an economic total loss.

The growth in our auction returns has significantly outpaced the growth in used car values. Using January 2017 as a baseline, the Manheim used car index has grown 8.4%. Within the same baseline, ASPs we're generating at our U.S. auctions for only insurance cars has grown by 35%.

While the rest of the industry is quickly moving toward the digital remarketing convention, we have been completely digital since 2003 when we introduced our VB2 platform. We have continually improved our auction platform, now VB3, over the last 16 years and is commonly recognized as the standard in the industry. The efficiency of VB3 and our elevated marketing focus on international buyers, have led to a significant growth in bidding activity from those buyers. Our full U.S. website is now translated into 7 languages. In addition, we have elements of our website that accommodate languages native to 135 countries. And currently, we sell from the U.S. into 147 countries.

In the quarter, 38.4% of all U.S. units sold were to international buyers and 46.9% of the value of the units sold were to international buyers. In total, over 70% of all the vehicles sold on our U.S. website received at least 1 bid from an international buyer.

Our marketing efforts have 2 goals: to bring more buyers to our auctions and to get those who attend to place more bids. We have been successful in both efforts. The number of unique bidders was up 13% and the number of bids received per lot sold was up 8%. Breaking down the growth in unique bidders further, we saw an 11% increase in domestic bidders and a 22% increase in the international unique bidders.

The growth in our ASPs has been the primary driver in a 6.4% increase in the U.S. revenue per car. Also contributing to that growth are the additional services we're providing to both the buyers and the sellers. The noninsurance markets continue to be a focus of growth in our U.S. strategy. These markets include franchise-independent dealers, finance and leasing companies, fleets, charities, equipment dealers and wholesalers. Excluding the charity and municipality markets in the U.S., our noninsurance volume grew by almost 20% and by more than 100% over the same quarter last year and the same quarter 2 years ago, respectively.

The growth in volume was spread broadly across multiple seller segments. Volume from dealers was up 14%, finance companies 24%, wholesalers 24%, rental car companies 62% and fleets and industrial equipment were up 41%. We have successfully grown our noninsurance volume as we developed better systems integration into fleets, banks and dealerships as we have developed sales and operational programs targeting individual segments, and as we continue to increase the auction returns that we're delivering to our noninsurance sellers.

Turning to the U.K. We delivered another very strong quarter as we saw growth in volume of 11.8%. In local currency, revenue and EBIT grew by 21.7% and 23%, respectively. The growth in volume came from increases in both insurance business, driven once again by market wins and organic growth and growth in our noninsurance business as we grew our U.K. dealer volume.

We also continued to see meaningful growth in both Brazil and Canada. As the value we offer in terms of technology, processes, land and people has allowed us to expand our market share in those countries. In Canada, we increased our volume in our local currency revenue by 8.4% and 21.5%, respectively. In Brazil, our growth was even more remarkable, increasing our volume in our local currency revenue by 22% and 36.5%, respectively.

Jeff has already provided a commentary on Germany. With that, we note that our operations outside of the Americas, the U.K. and Germany for the quarter remain immaterial in both revenue and EBIT. Globally, we are seeing rising labor, health insurance and sub-haul costs, all of which have led to an increase in our average cost to process each car.

Our inventory was up in the U.S, internationally and worldwide by 5.9%, 11.1% and 6.6%, respectively. When adjusted for Harvey, the growth in the U.S., internationally and worldwide was 7.2%, 11.1% and 7.7%, respectively. The year-over-year growth in our U.S. inventory over the prior 8 quarters has averaged over 9%, and we expect this trend to continue.

To accommodate this growth and to provide standalone capacity along the Gulf of Mexico and the East Coast, we have engaged in a massive capacity expansion initiative. In the last 3 weeks alone, we have announced new yards at Harleyville, South Carolina, serving the Charleston area; Antelope, California, serving the North Bay and Sacramento areas; and Mocksville, North Carolina, serving the Charlotte area. In total, these 3 yards added over 114 acres of capacity. So far, this fiscal year, we have announced 14 new facilities: four in the U.S., one each in Brazil and Canada and 8 in Germany. Currently, in the U.S., we have 17 expansion projects in the construction phase and 29 projects in the engineering phase. These 46 projects alone represent thousands of acres of capacity and will consume several hundreds of millions of dollars in capital.

That concludes my comments. Tod, I'll turn the call back over to you for the Q&A session.

(Operator Instructions) We'll take our first question from Bob Labick of CJS Securities.

A couple of real quick questions on Germany and then one on the noninsurance group, which was very impressive, that you just talked about. Starting with Germany, obviously, over the last, I guess, 6 to 12 months you've really accelerated your pace of roll out in your growth there. Can you talk about what surprised you the most over the last 6 to 12 months during this, kind of, acceleration phase?

I think the biggest surprise was just how well the team performed in terms of opening up so many yards so quickly. When we got there this summer and started looking at changing our strategy in terms of what would make us the most successful. We realized we needed a network. That's what's worked for us in the U.S., the U.K., and other markets, Brazil, Canada is we have a stronger network, we have a better -- an easier ability to pick up cars quickly and provide our services. And really just the team's performance and ability to do that is probably the most surprising thing. Second, I would say, is just the amount of buyers that we've been able to bring in. Once we started to hold auctions, and we're auctioning well over 500 cars a week now. Once we started that process and had regular auctions available to the members, the marketing team has just done a really great job on getting our brand out there and cultivating buyers for our auctions. So it's been very positive. We've talked about it in the past. I liked Jeff's comments, which are, let us succeed and perform, and then we'll report on those results, but we're very happy with what we see in Germany.

Okay. Great. And I know you want to succeed and perform first, but I just wanted one more question, then I'll move off. I know you talked about, potentially over the next several quarters, flipping an insurance company to the U.S. style. Do you still believe that's possible or likely? And if it doesn't happen over the next couple of quarters, what would be the reason that it wouldn't happen?

Well, I do think it's going to happen. The reason it will happen is associated with friction that you have today with the process. So you're requiring the insured to hold the vehicle and then have some buyer come to their house and pick the vehicle up. And there isn't an insurance company that I have met in Germany yet that isn't concerned with customer service, and Net Promoter Scores and giving the best possible brand experience that they can give. And obviously, when we send a uniformed driver in and pick the vehicle up, bring it to our location and then auction it, the end buyer's picking up at our location, you don't have that touch with the buyer and the insurer. And that can be a negative. There can be a number of scenarios that I could outline for you where the buyer ends up wanting to have a conversation with the insurer about the salvage. And clearly, there's not an insurance company that wants that. The second reason is return. We're able to buy cars, as Jeff has mentioned on previous calls, we're able to buy cars, basically, through the platforms and then bring them over to our yards, and auction them off and get a high return. And that is just very, very simple. You're allowing buyers from around the country and the world to bid on product that they know they're going to get. There is no contingency. There is no question. If they bid, they're going to own it. Whereas when they bid on the platforms, there's a less than 10% chance of knowing you're going to get the vehicle. Even if they're a high bidder, they may not get the vehicle because the insured may sell it somewhere else, through a body shop or through some other, through the dealership rather than to the buyer. So you're taking away that element of unknown and making it a sure, a guarantee that you're going to get the vehicle. And then you have the logistical component. And we've said this before, it should not be underestimated. When a buyer has to -- when a buyer bids on something, and has to pick it up, and go to 3 different locations, 3 different insurer's homes and pick that vehicle up within a certain amount of timeframe, that adds a degree of difficulty. When a buyer could bid on 3 different Copart locations and buy a half a dozen or a dozen vehicle so they can get a lot more product, and then they can take 3 weeks to pick it up as opposed to having to get it out within 4 days, that's just better experience for the buyer as well. So I guess, I don't have any doubt at this point that we're going to see continued traction in that market.

Great. That was really helpful color. Thank you. And then just one last quick one. Obviously, you just discussed very impressive growth of noninsurance from dealers, finance, wholesalers, rental cars, et cetera. Who are the primary buyers of these? Are these the new unique bidders coming online? Or just talk a little bit about the buyer base there, how it may be different than your core or if it is exactly the same?

No, Bob. I think the profile of our buyer base changes constantly based on the product that we're offering. And you're seeing buyers that have an appetite for these different types of cars and even heavy equipment that we're offering. And that just demonstrates the efficiency of our auction platform. We easily get to those buyers, those buyers easily get to us and are able to view these cars.

Bob, I'd just add to that, that I think both are true. So we expand, as the nature of the cars that we offer evolves, our buyer base expands further as well. But it's also true that the existing buyer base is thirsty for the kind of cars that are offered by the dealer consignors. Otherwise, by the way, they wouldn't consign through us, right? As much as we'd like to say we're fantastic for independent dealers and so forth, they vote with their feet and they vote as a reflection of the auction prices they achieved at Copart auctions. So I think it's a testament to the power of that buyer network that they're doing quite well and bringing more cars to Copart over time. I think that also has a virtuous cyclical effect as well because then those newer or less damaged or non-damaged cars then bring further buyers into the network as well. So I think it's a marketplace that continues to expand on both sides, buyers and sellers.

We'll take our next question from Craig Kennison of Baird.

Will, I wanted to start with you on the noninsurance business, continue to see great growth there. Can you just provide examples of systems integration tools that you're using to drive your noninsurance volumes with dealers, fleet operators or rental companies?

Sure. I mean, each of these segments tend to migrate to different platform for their system operations. For example, the finance companies use a system called AutoIMS. The dealers use a Dealertrack or a DealerSocket, a number of those other systems. The buyers seem to come to us from AuctionACCESS. And it's extremely important for us to write the integrations that are needed to reduce any friction that is caused by operating the 2 different systems, our system and theirs. And we have a significant initiative along the lines of creating those integrations, and that's just part of it. It's not just the systems. It's the processes. For example, heavy equipment. Transportation piece of heavy equipment could cost $4,000 or $5,000 as opposed to well under $100 for an insurance company. The timing process for charities, where you've got to pick up the title when you pick up the cars, completely different from the timing processes for dealerships. Dealers are much more concerned and in need of after-auction services like counterbidding. And I could go on and on. Finance companies have rules that surround repossessions, even voluntary repossessions with which we have to provide special documentation in compliance. And so it's not just a matter of being integrated into their software systems. It's being able to change our processes to accommodate their specific needs.

That helps. And then to what extent is the growth fueled by new customers trying your services versus customers that you've already landed that are dramatically increasing the use of your service?

It's both. We're seeing organic growth. We're seeing people -- customers that we've had for years that we're contacting again. We're, like I said, eliminating the friction, whether it's systematic or process, encouraging them to send us cars. And the returns they're receiving have driven more and more volume. It's not one big thing. It's a combination of a lot of little things that we're doing internally.

Got it. And then I also wanted to ask about the total loss rate trend in the U.S. and in Europe. Where do you think that loss rate is headed in 2019 and beyond? And then when you look at Europe, do you have any data to frame where the total loss rate is today and where that may be headed?

I'll comment on the first half of your question, that's for total loss rate. I think this is, as you know, the one-way tailwind behind the business for the last 40 years, right? That cars, once they're in accidents are ever more prone to be in total than repaired. I think that's an assailable macro factor that's really hard to read on a micro basis. So trying to forecast that quarter-to-quarter, even year-to-year is tough. We do think the overwhelming forces at work here will drive it upwards over time. There's no particular ceiling that we have in mind. So I don't have a good point in time forecast for you for 2019. As for Europe, I think the data -- we haven't seen a comprehensive data source that's quite as exhaustive as CCC here in the U.S., so don't have a comparable number to share with you. The U.K., as you know, is very different from Germany, for example, even at how they practice or how they handle total losses. But no, we don't have any point estimates as precise as the major sources we have here in the U.S.

We'll take our next question from Bret Jordan of Jefferies.

On the purchased car trend in Germany, is it possible to get the agency volumes up without flipping the insurance companies, in the sense that if the individual is still selling the car, once you have enough auction traffic, will they send you the car on consignment as opposed to you having to buy it?

Right now, we're acquiring cars so that we can hold auctions. So we're doing that through the platform. We do have some noninsurance volume coming in now as well. So we're starting to process vehicles for companies that service the insurance industry as well as rental car companies. But currently, the strategy is to illustrate the benefits to the large insurers and then have them switch over to our model.

Okay. Great. And then on Will's comments around the real estate pipeline, could you put maybe some timeframe around those thousands of acres? Is that going to be a very large near-term acquisition of real estate? Or is that 46 projects over a period of years?

I think years is too long. I think within 24 months, the vast majority of all those 47 projects should be delivered.

We'll take our next question from Gary Prestopino of Barrington Research.

Will, when you talked about the noninsurance, did you give the percentage breakdown of what percentage of vehicles were noninsurance this quarter versus last year? Or can you give that? Hello?

Do have any other questions while we're gathering that information?

Well certainly, I do. In terms of the gross margin on the purchased vehicles, sequentially, it was down a couple hundred basis points. Is that just an impact of more growth in Germany? Or is that currency? Or what?

It is a reflection in part of growth in Germany. It is largely not currency, because currency would affect both sides of that ledger, Gary, right, because of the buy-ins. Also in Germany it's partially also that purchased car mix can affect this as well, so as the prices, for example, kind of rise for certain purchased cars, and we talked about this before, but the more a car is bought and sold for [dollar] of course, dollar spread grows as a percentage. You cannot expect on a $10,000 car to make double the profit that you would on a $5,000 purchased car, for what it's worth.

Okay. And then lastly, I wanted to ask about the vehicle pooling costs. Year-over-year, they were up pretty dramatically. What would account for that?

That's largely the effect of the accounting change 606, Gary. So to make a long story short, the -- anyway, check the first quarter transcript and I think you'll find a pretty robust discussion there of how the accounting now forces us to push for revenue. We used to pull more revenue forward, now we push more back, which hangs more of it on the balance sheet, which is -- and the corresponding costs, which is why you can see that up so dramatically.

Yes. Okay, that would explain it. And then lastly, Jeff, I got on the call later, I actually got booked into the wrong call. Could you give me some of those unit volume numbers that you generally address at the beginning of the call, or unit volume changes or whatever?

Yes, I'll give you the big ones. So unit sales, nominal unit sales declined 0.6%. U.S. unit declined 3.7%, international growth of 18.7%. Excluding Hurricane Harvey, global unit sales growth of 7.7%. U.S. unit growth of 5.7%, ex-Harvey and ex-charities, we said, in the U.S. was 7.5%.

Gary, I got that number for you on the percentage of noninsurance vehicle sold last quarter. In the U.S., it's 22.9%. In the same quarter last year it was 20.8%.

We'll take our next question from Chris Bottiglieri from Wolfe Research.

I was hoping you disaggregate the increase in average selling prices. I think some of it's mixed. Is there a way to maybe just look at what the ASPs have done just for the insurance segment?

You're right. Of course, some of it is mixed. And we don't -- we haven't separated the impact, the change in mix versus the increase in ASPs solely for insurance companies. So I'm not sure we can provide that to you on the call.

I think, Will, did provide the stats on Manheim on a 2-year basis and U.S. insurance only being up 35%.

Right. But not the impact it has on revenue.

Right, not the impact on revenue. But the point is that ASPs are up and up significantly year-over-year, including for the insurance segment in isolation. So it's not just mix shift. It is also significant increases in insurance ASPs in the United States.

Got you, that's what I'm trying to arrive at. So when you think about what's driving this, is there any metrics you can cite in terms of like the total loss rates? Are you saying that total loss rates are increasing more for younger vehicles then for older vehicles? Or anything you can demonstrate, use to demonstrate that, like there's a younger vehicle being totaled today than maybe historically?

;

I think the themes you've heard us talk about on the last few calls all still hold true, which is that we are seeing, on average, slightly newer cars being totaled. So if we look at the model year of the car we sell, we are selling more newer cars today than we were a year ago, and that's been true for a while. We are also selling less damaged cars, so the cars are totaling more easily. We have certain metrics to -- the insurance companies do write and they provide them to us regarding repair estimates, so how much is the repair estimate relative to the intact value of the car. And we are seeing, by that particular barometer, less damaged cars entering our system over time. On the flip side of this is all the bidding phenomenon that Will described. So I think we are seeing newer and less damaged cars on the supply side. And on the demand side, you are seeing more international buyers, more buyers, period, by the way, domestic and international, but also more diversified and global buyer base for our cars. So it's both of those things working in concert that has driven selling prices up.

Got you. Well, just lastly there, I mean, the total loss rates are pretty amazing, and it sounds like still looking for 7% to 9% volume growth. Do you have any data on accident frequency, like what you are seeing there? Are accidents still down year-over-year? Do you think some of this collision avoidance technology is yet impacting accident frequency or do you think it's still too far off?

We probably don't have any better data than you do. So we follow third-party sources like fast-track data and so forth. And what I'd say first is that, for the vast majority of this company's history, accident frequency has declined slightly -- steadily but very slightly over time, and it's been dwarfed by, of course, total loss frequency on the other side of the equation. So that has driven organic unit volumes up very meaningfully over time. I do think that the rapid increases in accident frequency we saw from 2011, 2012, 2016 have tapered, so accident frequency may be declining somewhat, maybe flat, but it's not rising at the rate that it had previously. Total loss frequency, as far as we can tell, continues its upward trend.

We'll take our next question from Daniel Imbro of Stephens Inc.

I have one on Germany. I think you mentioned, Jeff, you guys have 12 locations. As a footprint, is that a sufficient network to service the country today? And then how is the salvage industry in that market growing? Obviously, you guys are growing rapidly taking share, but is the market also growing high-single-digits similar to the U.S.?

So as for the footprint, I think this is sufficient for us to participate actively in Germany. There's no doubt in my mind that, over time, as we penetrate the market, grow our platform, we will invest dramatically more still in land and capacity there. So the 12 is a good spread, geographically, across the country. But in terms of sufficiency of capacity, we are still in the very early innings of our participation in Germany. Your second question again was there?

Just on industry growth in Germany, is it similar to the U.S. in kind of that high single-digit range?

That's a tricky question to answer. So even your point about our taking share is a very nuanced concept in the sense that we are taking share from what is a very different, traditional salvage model through the listing services, et cetera, which I'm sure you heard and can review again from the first quarter call. But in a nutshell, I don't think the underlying characteristics should be different from the U.S. and the U.K. in the sense that the cars are relatively old across the system. So Germany is a mature economy that has had cars for a long time, so the average fleet age, likewise, is old. In comparison, by the way, to developing markets or to economies that have grown tremendously in the last decade or 2, China and India and the like where the cars are relatively new by comparison. So the cars are old. The cars are expensive to repair, meaning labor costs are high, parts costs are high, regulatory burdens are also meaningful, meaning you have to restore airbags back to impact condition to drive cars. So all of those same underlying forces are similar in Germany, but for like-for-like salvage auction statistics in Germany, we don't have them because they don't exist, right? We are the first ones to attempt to deploy the Copart model, so to speak, in Germany.

Fair. Will, I think you mentioned that over 30% of U.S. vehicles are now going abroad. And I'm assuming that some of those vehicles are going over to Europe. So can you talk about the buyer base in Germany to the extent -- I'd just love to hear your thoughts around, over time, as you develop the German market and the further EU market, does that cannibalize any of the international demand that you're seeing at your U.S. auctions today?

No, it really doesn't. Most of our international activity is in the less developed countries. Most of our cars provide affordable transportation and our 3 top are Mexico, which is obvious because of its proximity to the United States. The next 2 are the UAE and to Nigeria. And we're also seeing a significant growth in the Caucasus countries. There is some cross-pollination in our buyers in Germany, but it's not significant in its scope. And we don't think it will have a cannibalization impact on our international activity as Germany develops.

Okay. Great. And then last one for me. Just on capital allocation, you guys, obviously, chose to deploy capital toward share repo in the quarter and you funded it with some short-term debt. Understanding you don't want to comment on any future activity, but has your appetite, culturally, to maybe carry more leverage on the business, changed today given some of the scale you've gained in recent years?

No, I don't think there's been a philosophical shift at Copart. If you go back just a few years even at the end of 2014, we had leverage on the balance sheet for share repurchases we consummated in the summer and summer of 2015 and December 2015. So we had a little bit more leverage then even than we do now. But no, there's no philosophical shift in how we think about leverage. We generally prefer to be -- to have meaningful financial flexibility, which gives us strategic flexibility when it comes to acquiring land, pursuing international growth and so forth. So we'll continue to be a relatively low leverage institution.

(Operator Instructions) We'll take our next question from Derek Glynn of Consumer Edge Research.

As you think about additional growth opportunities outside of North America or Europe, China and India stand out as potentially 2 large markets in the long run. Can you provide an update on how you view those opportunities and whether investments have been made to expand there?

I think you captured the thought well there. They are very promising markets long-term. But for a host of reasons, the markets haven't yet materialized to nearly the same extent that they have in Europe and the United States. So we'll be there when it emerges, but it's not in the next couple of years anyway.

This concludes our question-and-answer session. I'll turn it back to management for closing remarks.

Thank you. Thank you for coming on the call. And we look forward to reporting on the next call on Q3. Thanks again. Bye-bye.

Ladies and gentlemen, thank you for your participation. This concludes today's conference. Have a great rest of your day.