車美仕 (KMX) 2004 Q3 法說會逐字稿

Good morning. My name is Andrea and I will be your conference facilitator. At this time, I would like to welcome everyone to CarMax's third quarter earnings conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question and answer period. (OPERATOR INSTRUCTIONS) At this time, I would like to turn the call over to Dandy Barrett, Vice President of Investor Relations. You may begin your conference.

Thank you, Andrea, and good morning. Happy holidays everybody. With us this morning on the call are Austin Ligon, our President and Chief Executive Officer; and Keith Browning, our Executive Vice President and Chief Financial Officer. Before we begin, please let me remind you that our statements today about the Company's future business plans and prospects are forward-looking statements according to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995. These statements are based on our current knowledge and assumptions about future events. Our actual results may differ materially from those indicated in our forward-looking statements as a result of various important factors. These factors include those discussed in the Company's annual report on Form 10-K for the fiscal year ended February 28, 2004, and our quarterly and current reports as filed are furnished with/to the SEC. Now I will turn the call over to Austin.

Good morning. Thanks for joining us. As you read in our release, third quarter sales represented a 13 percent increase in total sales, 15 percent increase in total used units, and a 2 percent increase in comp used units. Our used car sales trends as we said were stronger than in the late spring and summer. We saw wholesale prices fall in line with the normal autumn decline and we recovered some of the sales that we lost to weather, to hurricanes, in the second quarter and early first quarter. Early third quarter. As expected the the addition of subprime finance provider DRIVE contributed roughly 3 (ph)percent to used unit comps. We saw 7 percent decrease in total new car sales. This was primarily driven by the fact that we divested or gave notice to divest for franchises during the quarter, one Ford franchise and three Mitsubishi. We actually saw solid unit growth at the remaining new car franchises which now, what we consider our reduced set that we will stay with, constitutes two Toyota franchises, two Chrysler Jeep Dodge franchises, one Chrysler Jeep franchise, one Nissan franchise and one Chevy franchise. In seven different locations. We had an 8 percent increase in other sales and revenue, other sales in revenue grew more slowly than used vehicle sales primarily because of the addition of DRIVE. Loans that are sold to DRIVE are sold at a discount, which is the customary industry practice. This discount is reflected as an offset to third party finance revenues, partially offsetting the growth in extended service plan sales, customer pay service sales and other third party finance revenues. As far as gross margins go, gross margins increased on both used and wholesale vehicles. Margins benefited from the continuing refinements that we made on our appraisal cost recovery method and gross margins on other sales and revenue declined, due to both the impact of the DRIVE discount and to somewhat higher service cost during the quarter. As far as CarMax Auto Finance goes, quarter three cap income benefited from the adjustment and evaluation of retained interest. This is based on reduced default assumptions on more recent public securitizations, which reflect a lower net loss rate being experienced by these pools. These valuation adjustments are a normal course of events and benefited us by 1 ccent a share this quarter. The cap gain spread as reported was 4.1 percent versus 3.8 percent last year. If you exclude the valuation adjustment, the cap gain spread was 3.6 percent in the quarter. For the balance of the year we continue to expect cap gain spreads to be around the lower end of the 3.5 to 4.5 percent normalized range that we have discussed in the past. We believe a very slow pace of interest rate increases has made it difficult to raise consumer rates to allow us to operate in the middle of that range. As far as SG&A ratio goes, SG&A ratio at 11.3 was up 60 basis points from last year. I will point out that our newer stores, which are not yet mature, have higher SG&A rates and immature stores -- that is, stores that are less than 48 months old -- now constitute 23 of our 57 used car stores whereas in same quarter last year, they were 13 of 47. So they are now 40 percent of the mix versus 28 percent last year. The increased proportion of immature stores had obviously adversely impacts SG&A rates. It is also exacerbated because of the seasonally lower volumes that we always have in the third quarter. And the third quarter SG&A ratios was also adversely impacted by higher storage unit buses bonuses, resulting from the somewhat better than expected sales in the third quarter. We set our store unit bonuses on a rolling basis to target what we think current run rate is and, obviously, as sales improved in the third quarter more than we expected, this storage overachieved to our original expectation. Fourth quarter expectations that used unit comps will be in the 2 -- plus 2 and plus 7 percent range and the EPS should come in in the 19 to 23 cent range. These expectations take into account that continued improvement in sales trends we have seen since September, it assumes no abnormal winter weather and it takes into account the adverse calendar shifts which include losing a Saturday in December to Christmas. That Christmas was on Thursday last year. And losing an extra selling day because of not having Leap Year this year. So with that we will be glad to take some questions.+++ q-and-a

(OPERATOR INSTRUCTIONS). Stacey Widlitz with Fulcrum Global Partners.

Good morning, congratulations. Can you just comment on what you're seeing out in the marketplaces, pricing and option? Are there more all fleet cars available and if you could just expand a little bit on the comments you made about the service operations? Does this have any implications for systems issues? If you could just elaborate a little bit on that.

As far as what we're seeing in the marketplace we have seen the fall evolve into a pretty normal fall and when I say normal, if you look back over the last seven or eight years and track a basket of cars through the year and what its wholesale price is, this ball is pretty consistent with what typical falls have been, other than the 2001 post 9/11 year and 2002. So those are unusual years. This has been a more normal year. As far as more off lease (ph) cars, I don't know that I can knowledgeably comment on that I know we have heard there are somewhat more off rental cars. But I think in general what the buyers have been telling me is that the market pretty much looks like it typically does in December.As far as service, the reason we mentioned service is that the impact was a penny and we think that is material. It deserves mentioning. The key on service is to recognize that for us, service is a combination of reconditioning and retail service. Reconditioning is the bigger portion of it. In this particular quarter, we had several things occur. One thing, the third quarter is always a period where because of lower volumes and relatively fixed overhead in the service business, our expectations for making a profit are the lowest of the year. As we entered into the quarter, as you know we talked before about having somewhat higher inventories during the summer than the sales rate ever got to the point to justify. You also know we bring inventories in line with where we -- where the sales rate is and below when we get into September. That adjustment process occurs at the end of August beginning of September because Labor Day went over into the third quarter this year. Part of that adjustment process was in this quarter. What that means is we have a lot of techs who aren't turning wrenches during that period of time. We send them to training, some of them go on vacation. So you are paying those sellers and not producing any cars, so they are not getting absorbed. That was an impact. We also saw a gasoline impact. We also saw the same bonus impact that we saw in the rest -- we bonus our stores on total store performance, not department by department performance. Overall, those added up to a small loss as opposed to a small gain. Once again, recognizing that this is not what drives our business. The key to our business is what we sell on the selling side. We certainly would like to have done a bit better in service but we don't see any long-term over there. It just came together this quarter.

Hello?

Andrea?

Sharon Zackfia with William Blair.

On the appraisal cost recovery initiative, can you keep us up-to-date on what the refinements are now that you are implementing?

Sure. I think that one of the things I have explained that we did as we shifted from having a specific fee that we charged to the consumer, when the consumer sold us a car, what became clear to us is that that didn't really align with what the costs were for us to handle different kinds of cars nor with what the impact of those fees on the likelihood we were going to buy a car was. What we have done is essentially do away with that fee as far as the consumer sees and just determine on each car what is it -- or each category of car, if you will -- what is the appropriate charge to put on that to fully recover our cost? And where can we most effectively recover that? So what we continue to do as we go along is to tweak an experiment with that methodology to understand where we do that best and we had just gotten a little more efficient with it.

Is this a meaningful impact on your gross margin dollars per used car?

Let me ask Keith B. Would you call that meaningful?

In a couple tenths. I mean it's (inaudible).

A couple of tenths so if you call a couple of tenths meaningful, yes. If not, no.

I guess I was just going to ask apples to apples because you have the ACR program in the year ago quarter.

We did and (MULTIPLE SPEAKERS)

We just happened to change that structure which has improved the performance, both in wholesale and in used cars. The greater impact was on wholesale.

So that we had the initial version of it in place a year ago and we refined it quite a bit over the year as we were trying to figure out how do we do this most effectively? Sharon Zackfia: I guess secondarily on these store level bonuses, can you give us any indication as to what kind of penalty you saw in this quarter? Is it 20-30 basis points? Have you recalibrated now going into the fourth quarter?

We recalibrate every month. So we set store sales targets. We set our target at corporate at the beginning of the year and we stick to that. We get paid a bonus on that. If we don't hit we don't get paid. That doesn't work effectively for stores, because if we made a wrong forecast or the market moved against us, you won't have happy productive store people who don't turnover if you don't adjust to that. Not their fault -- it is our fault. So we adjust those every month. Really, obviously, as we have said we essentially saw a turnaround beginning in September. We tended to lag for where that turnaround was because for six months we hadn't seen that. So it was fairly obvious. As we get more experienced, a little more comfortable forecasting. Believe me the stores will tell you that we are very good at adjusting back to where the trend is and giving them a bonus target that is usually in their mind a little higher than realistic and in our mind right on it. We have adjusted there. You have a comment on how big?

It was between 10 and 20 basis points, somewhere in that range.

Then can we get the specific revenue number for third party financing income? Or revenue, sorry.

Don't have that available.

Keith doesn't have it available right now. It's something we will be putting up.

Yes.

Matthew Fassler, Goldman Sachs.

Couple questions if I could? Just take another stab at DRIVE subfront (ph) financing business. In the aggregate, can you give us a sense as to how the economics of a car sold with DRIVE financing compared to the economics of a car that you would other fund yourself or finance yourself or work through one of your other third party providers?

The one thing that we are not going to do is specifically tell you exactly what we pay DRIVE. But what I will tell you is if you go back to our economics that are in our investor note, which says we make somewhere around $2800 all in margins from all sources on a car. Somewhere around 350 of that typically comes from cap and somewhere around 75 I think comes from third party. Actually 335 and 80. So a little over $400. We make none of that so we lose that portion of it. So that takes you down from about 2800 to about 2400 or 2350 and then we pay a fee and it is a non trivial fee. It is a very competitive fee. I think everybody else would be glad to have it because of how our portfolios have always performed but it is a non trivial fee.What we have generally said is we make 40 to 50 percent as much overall. That's as close as I can get you.

40 or 50 percent as much EBIT per car.

40 to 50 percent as much total all in margins.

All in before the selling expense or after the selling expense?

Before the selling expense.

Okay. But is it still a profitable unit transaction for you?

Sure, yes, I mean we are eager to have market share but we are not stupid. Or at least we try not to be. And believe me, one of the reasons we tested DRIVE for a long time before we rolled it out is we -- given these economics, we wanted to set a high bar to prove to ourselves it was incremental. Because it is absolutely profitable but it is incrementally not dramatically profitable and we wanted to make sure that we weren't substituting any sales that we could made otherwise. Okay?

Good. Second question. Combination of new stores and expansion plan. Can you give us an updated sense as to how your new stores, your noncomp stores are performing against plan? And also, a couple of weeks ago when you released your sales, you discussed plans to open 9 stores next year. I realized that your growth rates that that implies is consistent with your big picture statements. It is a couple fewer than I and perhaps some others had anticipated. So what are you seeing in you stores, what might that have to do if anything with the rate of expansion that you set on the table for next year?

The general point of view on new storage hasn't changed which is, as a group, we are happy with the way the program is going and see nothing that would change our view of how we want to go about doing it. The comments I have made in the past I think still apply which is in general, our ability to forecast is in a tight range. It's better on satellites, which you would expect because they are going into markets with established sales patterns. and less good on stand-alone or new mid-sized market stores simply because every market turns out to be somewhat different. In aggregate, both of those categories are categories that are in line with our expectations and that we continue to be happy with.In terms of how me stores we are opening this year, it's from year-to-year we ended up opening one more last year than we expected. We could have opened anywhere from 9 to 11 this year. It is a combination of how many stores do we have out there? Where are they? A little bit how the real estate goes a little bit in terms of how we want to phase the markets. In this particular case, we have got a couple going up in New York and -- not in New York. I am sorry -- in Los Angeles. Sorry. Keith just had a heart attack year. (Laughing.)

I was very excited for a moment.

We will try to revive him. A couple going up in L.A. and we are going to start advertising in L.A. Now you might think wouldn't doing more be good. Actually because in the first year stores mainly add cost. That is one reason you don't necessarily want to operate at the high-end. But it is really all of those factors where, as long as we are somewhere in the 15 to 20 percent range, we are happy with it. You shouldn't expect that 15 to 20 means we will always be at 20 or always at 15. We will be somewhere in that range and on average, you probably ought to expect the average to be somewhere closer to the middle and over time, you would expect that to gradually drift down below the middle because, frankly, the bigger you get, the harder it is to keep up that percentage rate.

Got you. Final question, not to dwell on where there's no water under the bridge but when you preannounced a number of weeks ago you taught about comps that were two to three (MULTIPLE SPEAKERS)printing it to. Obviously, you feel good about the business and your guidance is suggestive of a nice trend. But just given that there's a little bit of (indiscernible)I think in terms of expectation on the week to week basis. Any kind of clarity you can give us on what you're seeing would be very helpful.

Yes. What you might guess turns out to be true. Which is, when we make that estimate it was based on the trends that we had seen and Thanksgiving weekend was a horrible weekend for us like it was for a lot of retail. That was the primary reason that we ended up at the lower rather than the higher end of the range but we've had 2 1/2 weeks since then and we're pretty convinced that that was a blip. We don't know why. We can go back and look at our indexing and everything else, but the best I can say right now it's a blip, but what we have seen in terms of the trend up through today says that that is what giving us what we're forecasting for next quarter. Obviously we give a range and we expect to be in that range; but it was really we got a bit of a curveball over Thanksgiving weekend and we were all a bit disappointed with that. ))Operator:Bill Armstrong. C. L. King and Associates.

Just another question on vehicle pricing. The Manheim index indicates that November wholesale prices were actually pretty high. I'm not sure how much that actually impacts your business but that's a widely followed index obviously. GM, I think, is having some kind of red tag sale. So I guess the question is what are you seeing, currently, in terms of spreads between late-model used car prices and new car prices? And what sort of assumptions do you have broadly in terms of that those market parameters in formulating your 2 to 7 percent comp expectations?

Okay. And high is not how I would describe the Manheim index. It was -- the Manheim index is the seasonalized. Prices fall in the fall every year and they have again. So prices in November were quite a bit lower than they were in October. But what Manheim is indicating is for their particular basket of goods, prices actually bumped up a little bit, compared to seasonal expectations. For our particular basket of goods, that is generally consistent not necessarily because we don't buy necessarily the same basket that the Manheim index reflects. In general what we have seen is, the price of what we buy has been pretty consistent with where we would expect prices to be on average in a year like this. As far as what is going on in the broader marketplace and what influence that is going to have. As you know, Chevy and Ford in particular and to some degree Daimler Chrysler but, particularly, Chevy and Ford, have struggled this yearto sell the inventory that they have decided to produce. They have gone off and on incentives. They have been off incentives not off incentives but the incentives have not been cranked up to their highest level recently. And during this time of year, you'd expect at some point they are going to raise incentives again to try to blow out some of the extra inventory they have. When I read the analysts who cover the manufacturers, there are some suggestion that the strongest element of that may come after the manufacturers' fiscal year. They haven't had that great a year and what they have had has almost all come from finance. It wouldn't surprise me if we saw a strong surge of incentives after the first of the year but you never know, because one of the things they are trying to do is outsmart each other all the time. So you never quite know, it wouldn't surprise me either, if we had a curveball at the end of year. But, during this season of year I would expect that that would translate through very quickly to wholesale market because you are not at a peak demand period of the year. And my expectation would be that used car, new car prices would stay pretty well in line because you are really at the soft end of the wholesale market. We have made our estimations using what I believe is sort of a prudent view based on what we have seen this last year and the best we can draw from the trends that are out there would certainly -- the possibility that we see a return to very high and very aggressive incentives. There are a lots of moving parts there and we try to do the best job we can of putting them altogether and that is what we've done again this quarter.

Are you seeing or anticipating any influx of supply into the used wholesale market from either rental car fleets or off lease vehicles?

That's continuous. There are always off rental and always off lease coming in. You probably meant are we seeing some dramatic change. And the answer is not a dramatic change. I think my understanding is rental flows have bumped up a bit; and we certainly know that leases are coming back. I don't know if off lease flows are actually increasing that much, but I don't think we are expecting any dramatic change there.

Finally could you just comment on how some of your smaller markets are doing? I know some of these stores are still fairly new, but any comment either in your small markets or also your midsize markets where you have satellite stores? How is that progressing so far, relative to your expectations in terms of market share? Do you think those markets --?

Yes. In general, all of those are doing well. The one comment that we've made, specifically, in terms of differentiation between how different types of markets are performing is that, among the newly opened markets, in general, it looks like markets that are close to existing markets where there is some well-established CarMax customer base before we come into a market. All other things being equal, we will tend to do somewhat better. And markets that don't have any established CarMaxcustomer base, all other things being equal, will do somewhat worse. But even that is not 100 percent consistent. We have had markets both directions that have defied that a little bit. In the grand scheme of things when you add them together we are quite pleased with how the whole group goes and to be quite honest, some day, I would love to get to the point that we can understand the many factors that go into determining how a market starts out, closely enough that we could be close on every store. But for right now, what we are happy with is if we can be in the ballpark on every store and in aggregate they can come out where we expect. And that is what we're achieving.

Michael Heifler, Deutsche Bank.

I have a couple of questions. One big picture, one a little bit more micro and I realize that this is somewhat unconventional thinking but we are noticing that inflation is creeping back into the new car market. Just want to get your thoughts on the applications of a more neutral new car pricing environmenton used car volumes, margins, and inventory management? Then on the micro side given the adjustments that you have made to the retained interest this quarter, are there any changes contemplated to the gain on sale assumptions going forward?

I will answer your questions about volume margins and management relative to new car prices, and then I will let Keith answer about the assumptions.You, correctly, as you know and I know your group tracks the new car manufacturers, the domestic new car manufacturers have been trying to raise prices because they need to raise prices in order to make anything approaching a tolerable return. That, in effect, implies inflation. Whether they will be consistently successful at that, I don't know. Part of it will depend on do they make the production cuts that will allow them to do that. And you certainly heard a lot about that from analysts and even from some of our public sector, new car retail competitors. Most notably AutoNation who have been jawboning like Lyndon Johnson recently, to try to persuade the domestic guys to really ramp down their production some and let the new car retailers make some money and let themselves make some money. So if they are successful in doing that our observations from the past is, if new car prices go up, what it does is, it makes it a little easier for everybody to manage in the used car environment. In general, historically, we don't see that that has a particular impact one way or another on used car volumes. Used car volumes are more driven over a year by the overall turn in the stock of cars and less by the factors that drive new cars and as far as margins, same thing. Margins will tend to be in line. So it would make it a little bit easier inventory management environment. As we had talked about before that is kind of a two-edged sword because, at least for us, because to some degree when the environment is more difficult we are at more of an advantage. But more than an advantage, if everybody is at a disadvantage isn't necessarily good. So in sum, I guess what I would say is if new car prices somehow manage to go up we don't think there's anything bad about that. We'd be quite happy to operate in that environment. And we don't think it would be any more difficult and maybe a bit easier than the environment we have been in. I'll let Keith answer the question on the assumptions.

On the gain in sales assumptions, we had actually put in a new scorecard a couple of years ago and believe that it is doing a much better job of predicting the good versus the bad and wanted to see this season's losses because this is the peak season before we made adjustments. And, yes, the answer going forward is that given our ongoing confidence in the predictability the losses and future pools of securitizations will be reduced to a lower level from what we were assuming before reflecting the confidence and the performance of the scorecard for now over two years.

Certainly it would be our view that that is all good news.

Right.

Does that flow into the gain on sales?

Yes.

What kind of positive do you think that would translate into on a margin basis?

The answer is, I mean, we have to specifically fix the gain on sale based on the actual loans we are we are underwriting. So average cycle scores will move depending on the customer flow that we get. So the answer is that it will be positive somewhere in the range of 10 to 15 basis points going forward and then will continue to monitor the performance of our pools (ph) and make sure that we are more comfortable with that. I can't give you a precise answer because the impact of who buys for most changes quarter to quarter month-to-month and our loss assumptions has to reflect a characteristic of those people that are purchasing from us. ))Operator:Aram Rubinson.

I can't recall last year in the third quarter whether you gave us a full year outlook or not but -- .

Your outlook for next year?

For next year.

For next year -- yes I can recall. No. No. We never do that. That's way too early.

When you look towards next year, just curious in terms of the shape of the curve, whether you think it ought to be a smooth and more predictable year or whether you expect additional choppiness? I guess I'd just be curious on a broad brush and then have a follow-up.

Yes and no. The honest answer is, your guess is as good as mine. 90 days from now, my guess will be better and yours probably will be too. And as I said, in general, the used car business has a smoother curve. I think the competitive consolidation and price war that has been going on new car business particularly with the domestic guys has injected some volatility that we haven't seen historically. Part of how that affects the business will be what did dealers learn this year and how do they react if more of that happens next year? To be real honest I can't predict that because I haven't, we haven't seen and other dealers haven't seen as far as I know a year exactly like this one. It is not the way the two businesses used to interact. I'm hopeful that dealers are keeping a little more careful eye on what manufacturers are doing; and being a little more skeptical about how strong any recovery might be in terms of their willingness to commit to strong additions of used car volumes. And certainly what I hear from the publicly traded new car retailers is consistent with that. I am not willing to forecast next year on that yet. We will look another 90 days and take our best guess then, because when we do end of year we will give a forecast. And then I think particularly as we get into the spring and summer when we both see how the economy is and how dealers and manufacturers are reacting to it, we will know. You have to know a lot of unknowable things to be able to actually forecast that curve for next year.

So it just for a starting point you use let's say the square footage or the unit count growth (MULTIPLE SPEAKERS) 15 to 20 percent. To the positive side of that would be this potential change to the gain on sale and, then to the detriment, I would assume would be drive margins and some of the advertising. Is that a fair --?

Remember, on the finance side while on the positive, you have this improvement in loss rates, you still have a rising interest rate environment that so far we haven't seen gives us the leeway to move back up to the midpoint of our operating spread. So right now, you sort of expect that to go on. Yes, I think those are some of the key factors. Look, as I have always said that the most important key factor at the margin will be (indiscernible) comps. How well do the total set of stores that are comp do in terms of outperforming the prior year? A couple of points plus or minus makes a big difference there.

If you look throughout much of next year your comparisons are extraordinarily easy. Is that fair or do you prefer not to look on a comparison basis and the run rate is -- in the math but would still work that out?

The answer is the run rate is what drives you but there is no question -- extraordinarily is a strong word but certainly the comparisons next year are easier and you're absolutely right. The second quarter comparison is the easiest second quarter comparison we've had in the summer, maybe ever. Certainly in a long time. That works in our favor; but the real key will be how robust is the run rate and what comps drop out of that.

Just one or two other little things. The drive strategy, I think you were kind of suggesting that you started with it a little bit carefully and then would be willing perhaps to roll it out a little bit more even. The 3 percent lift. Is that what you're implying or assuming say for the fourth quarter as well?

Yes -- I think you misunderstood a little bit. We started, carefully, long before we told anybody about this.

We actually started with a test in the fourth quarter last year. So while we might still see a 3 percent lift I think that is a reasonable expectation. We actually expanded it through the year as we gained more confidence and they gained more confidence. It would work for them as well.

So we were doing DRIVE for many months before we talked about it publicly on a smaller scale. And when I said roll it out, the larger rollout has all been done. So it is not that there are stores that it is not in, or that there's some effort to ramp up drive at this point and as we go into the year as Keith points out we will start to come up against sales that in fact, we already are in this quarter, have some DRIVE built into them.

Last thing, I promise, is if you could just help us quantify the impact of the calendar shift for the fourth quarter?

Well, losing a day is -- you can do that yourself. It's as far as leap year is what it is. The impact of having Christmas on a Saturday is an interesting question. We have calculated it as having a negative impact. If you look back at the one time in our history where we had enough stores that it mattered, actually having Christmas on Saturday was one of the better patterns we saw because the days between Christmas and New Year's typically -- well, in fact, always overperformed to those days any other time during the winter. It is one of the strongest weeks of the year. Having Christmas on Saturday, the one other time we had it like this, was very positive. We haven't built that into our forecast. So we are still assuming that we effectively lose probably a couple of points due to the loss of a Saturday in Christmas. As far as New Year's, we are open on New Year's. New Year's occurs on Saturday. It is less clear what impact that has. Depends on how many people watch Bowl Games. I think.

Thinking more of 1 to 2 percent total negative calendar shift rather than 2 to 3?

Keith is nodding so that (technical difficulty) (MULTIPLE SPEAKERS) about right.

Scot Ciccarelli. Scot Ciccarelli:Couple of questions. First of all, any commentary regarding the negative equity situation out in the marketplace? And maybe you are seeing more or less impact than you saw in the summer months? Just trying to get a feel for it what you're seeing on the negative equity side of the stage.

In terms of where people are as far as we can tell there is no big change there. In terms of what the manufacturers are doing some of the easing off on some of the extreme incentives hurt GM and Ford, because it put them in less of a position to do extreme dramatic negative equities. But I don't think there is any change in the customer base that we have seen. That's much more of a long-term phenomena. Now. In particular, if you happen to be the owner of a large SUV, you're hosed. Behind the scenes, if you disaggregate what is going on in the wholesale prices whereas the wholesale prices in aggregate may be normal, behind that, compact cars have gone up in price, very much against seasonal trends. And the large SUVs in particular have plummeted like a rock. So for that particular set of customers who happen to be driving a one- or two-year-old Suburban or Expedition their visit to try to trade that in would not be a very happy one. If on the other hand you happen to have a Civic you're be pleased as punch and you might be able to upgrade to a used Suburban at a great price.So we're not seeing dramatic aggregate differences but there are a lot of things going on behind the scenes there but they seem to net out.

Right but it seems to me that the manufacturers certainly the cap defiant (ph)companies may be easing up a little bit on what they are able to (indiscernible). I think that would actually favor you to a degree because you are not in as much of an adverse position. (MULTIPLE SPEAKERS) compared assumptions?

I absolutely think that in terms of how does it affect our competitive position and maybe I wasn't understanding correctly. I don't think the consumers changed but to the degree that the manufacturers decide there is a limit to how much negative equity, particularly how much extreme negative equity they want to finance, particularly on vehicles that continue to collapse in value from under their finance company. That is good for us. And we will just have to see whether they stick with it. There is enormous debate among the analyst and in the auto industry right now about where this -- particularlythe competition between Ford and General Motors is going and also how resurgent is Chrysler going to continue to the? They are a wild card, because those three guys are really fighting for the same share. They are not taking any share from Toyota or Honda and Nissan.

Right.

But to the degree that that continues that would be good to the degree that it goes back the other way, that would tend to probably work against us a bit.

Thanks. Next question in terms of I think you commented a little bit on terms of SUVs and compacts -- compact cars. Any change in the mix of vehicles? I've seen -- broadly speaking, we have seen some weakness in, say, the lower end consumer. Maybe some more relative strength in the higher end consumer. Have you really seen any kind of change in the mix of vehicles that you are selling?

It changes all the time and there's also a cycle to the year. Because this time of year sees relatively more rental cars on the market every year, there is opportunity to buy more current model year cars at relatively attractive prices, and so that will edge up a little bit. For some SUVs perversely enough the fact that the prices have collapsed has been good. That means that you're getting some bargains out there that some shoppers are saying, "Gee even at $2 a gallon, $1.85 a gallon, I need this vehicle and now is a good time to get it." So the mix continuously shifts. There is not a specific trend I point to. I do think that, in general, we have observed some continued strength in consumers who either come with their own financing or are able to get their own financing which tend to be stronger credit rating customers. And that may suggest that some of those are people that six months ago or five months ago were buying a new car. But as the finance as the GMACs and Ford Motor Credit have pulled back their horns a little bit, they are more active in the used car market.

Fair enough. And then the last question is you had mentioned about relative strength, vs. weakness in certain markets where you are more familiar. Can you outline for people different parts of the country? Are there certain areas doing better than others or have the trends you have seen been more uniform?

Well at least for our business, and I had a chance to talk to some other folks from Texas recently, and they said that for their business this wasn't true in new cars. But for our business, Houston has had a great year and if you know anything about the oil business whatever your view of Iraq is, oil business service companies got a lot of contracts this year and they are almost all based in Houston. And my sense is that is flowing back into the used car market in a pretty healthy way. We have seen a great year in Houston, generally good year in Texas. Other than that, I don't think there is a regional comment that we would make.

Brian Nagel, UBS.

My question relates to the guidance you laid out, the unit guidance you laid out for the fourth quarter. If you look on a two-year basis that's assumes things are going to pick up nicely against a little more difficult comparison. If you could just go through -- what are the factors that lead you to suggest things are going to pick up?

The key thing, the key way that we are looking at this is, we are really looking at the last 90 to 100 -- really the trends since Labor Day. Let's put it that way. So Labor Day is kind of a breakpoint in the seasonality for used cars. Up until Labor Day you are really in the full-fledged summer mode. Almost immediately after Labor Day, you start to go into the heavy discounting period of time. As we look back over that period of time, that is the key to what we're forecasting for the fourth quarter. What we're trying to do is make a forecast that says in general what that trend seems to tell us. Not necessarily where it would forecast going up but what that trend seems to tell us that we sort of stayed generally at the levels that that has gotten to will generally be where we are going to be. We are not trying to outguess the fourth quarter too much in terms of what additional developments will come so much as look at where the business has evolved to and take that forward. With a reasonably wide range, we used to give narrower ranges than this. One of the reasons we have given a wide range is if the trends continue to improve, obviously, we would be above the midpoint of that but there are lots of things that could put a curve ball on that in terms of not disproportionate but some big snowfalls. Not completely out of line with expectations, but maybe slightly more than average or a big crash and burn incentives right after the first of the year from Ford and GM, changes in how they decide to do financing. All of those things. So that is why it is a fairly wide range. It is really based on what we have seen since Labor Day and our best assessment of where the business is now.

So if you were to break it down then just (inaudible) you have -- is it more, is it improving demand or is it better supply that is helping the sales trends (inaudible)?

Keith is nodding yes. The answer is and we are not being facetious. The answer is, there appears to be improving demand. Now how much of that is demand coming from the new car side vs. how much is just improved demand in the economy, clearly, post the election things in general got better. Whatever that means. I don't know if it means people like the results or they are just glad it is over and glad there was a determined result, but since the election things seem to be better. There does seem to be some demand improvement we would guess from a little easing on the new car side. There -- certainly the rigorous of the fall market always bring the supply and demand on used cars back in line and this is no different than any other year we have ever seen. Our history, our historical experience is, it sometimes over adjusts but we have never seen it underadjust. It is pretty hard to go into the winter overconfident in inventory and holding on to too much inventory. Dealers get real rational as the weather gets cold and customers get harder to find. That, in our view, has helped bring the supply demand balance back in line. So I think all of those factors are played.

Just one lst question. Someone may have asked this before but I just want to be clear. The DRIVE component had a 3 percentage points in Q3. You guys are starting to anniversary that. You would expect a slightly less incremental contribution in Q4?

We are still expecting about 3 percent (indiscernible).

What we are anniversarying is a few test stores not the main rolled out DRIVE.

Bruce Babcock. Saybrook Capital.

I was just wondering as you enter various new markets where you are doing not the satellites but the major new stores, is there a similarity as you go from market to market that you can really learn from? Is each one different and you have to go through lots of mistakes errors or basically can you get better from the experience you have been having in these new markets?

No, unquestionably, you get better. I will tell you we now operate 59 total stores and we to the degree you know us we are about as analytical as you can be about this stuff. If there is statistical knowledge to be gotten out of what we have done, we take every angle that you can to squeeze it out. When we have gone to some of the outside consulting companies that do this for other retailers to say, look, in terms of our ability to forecast sales, are we doing something wrong? Is there something else we can do? What they all say is, my goodness with the sample size you have to work from, I don't believe you can get any more knowledge out of the stores that you have got than you're getting. So the answer is, we have absolutely gotten better over time. And we continue to get better, but the sheer scale of our stores and the fact that for a $5 billion retailer, we have very very few stores, just makes the statistics a little harder. We work on that, I will tell you we pour a ton of effort into that and I believe we continuously learn more every year and get better at it. One of the other things we try to do is really focus on sort of risk assessment. That is how good are we at this particular type of forecasting and, therefore, what range of errors should we use? And we get better at that too.

Great, I will ask one other question. Shouldn't there be a fairly large number of new stores now coming into comps? Since you started that new store program a couple years ago?

Yes we started the new store program 3, 3 1/2 years ago and, yes, we have got a fair number that are coming into comps now. One thing I'll say about this last year is when the business was bad, it was bad everywhere. And the new stores weren't exempt from that and it wasn't because they were new, it was bad everywhere. So we didn't see the benefit out of the new stores to the degree that we would normally expect, just like we didn't see it out of any of the other stores. (MULTIPLE SPEAKERS) Going forward in a normal environment gets some benefits from an increasing number of maturing store -- it is the inverse of what we talked about on SG&A. An increasing number of new stores in the mix hurts your SG&A but it should help your comps and we should, in a more normalized environment, see some benefit from that.

David Campbell, Thompson Davis.

Can you say anything about the midsize market satellite stores you have opened and what your experience has been with those?

In general, we have been pretty happy with those; and the only -- I think the only one that we thought about specifically is Charlotte because it is the only one that has been out there long enough. That has pretty well -- well Charlotte and Las Vegas. Charlotte and Las Vegas have been out there long enough. We can say -- what we can say is in aggregate, the total volume ended up about where we expected. It doesn't always end up at the store you expect. So with the sample of two, that is where we are. We now as you know have Richmond that we have added. We now have Orlando that's coming around. Orlando has, overall, been a tough market in the last couple of years so that has been a little hard to judge exactly how that is splitting up. But in general, we are pleased with what we have seen so far so we think that, roughly, we're able to correctly forecast the market. We are not sure that we know exactly how to forecast the split of sales. But if we get to aggregate right, forecasting a slow sale is not so important.

How many more of these do you have planned for next year? Are you going to just observe what you have got for the time being?

No. We have a store on the north side of the river in Nashville called Rivergate. That is our only midsize market satellite, Dandy?

Let me look.

I think we've got one more coming this year but we will continue to do them as we see opportunities based on kind of that risk reward estimate.

Kansas City and Nashville.

I'm sorry, Dandy points out, I forgot this. We also have a second one coming in Kansas City. That is a little different situation. Kansas City is not Richmond. It's a much bigger market. Richmond -- this is the lowest population per store that we have ever had with about 450,000 to 500,000 people per store trade area. Kansas City, we are adding a store in Independence, Missouri. There is a clear break within the Kansas City market because the Missouri people are less likely to come shopping in Kansas at the same distance than the Kansas people and, therefore, we need a store on the Missouri side. Everything indicates that that is a big enough market that if I were looking at that again today, we would have built it as a two store market to begin with.But we will continue to do these, based on our estimate of, what have we learned so far? And what is the risk reward probability?

Adam Kumora (ph) with Intrust Capital.

I just had a broader question about how we should think about margins sort of when we think we are going to reach the inflection point and start to see them stabilize and maybe expand? And if you could give us a longer-term outlook for them?

When we should reach the inflection point and start to see them expand. Margins are (indiscernible) car dollar margins go through a cycle every year. The third quarter is the lowest point. Typically the first and second spring and summer are the highest and winter is December, January, February -- somewhere in between. We expect it to go through the same cycle. We do not expect to see a big dollar margin expansion. We have been a little bit below margin on the used car side. A little bit above on the wholesale side. I don't think that is where primary upside is going to come. Primary up -- as far as gross margin, primary upside will come from sales and the net margin that comes from the overhead leverage.

I was talking about corporatewide operating margins -- when we should start to see them expand again?

Well when sales go up enough that we get that overhead leverage. Not to be facetious but, look, if we get in the middle range of the target, I would hope we will get some benefit out of that depending on how next year comps. If we are in the middle of that 4-8 percent comp range if we will get more if we are at the low end of it. Will probably be flat so maybe a little bit, would be my guess. In terms of overall operating margin.

Charles Grom with Morgan.

Could you just for modeling purposes just outline the timeline for your stores next year? Where are the quarters?

Well, Dandy, hold on just a second. Yes, hold on just a second and I will get that. (MULTIPLE SPEAKERS)

Yes sure and then also nice to see the upside in the used margin and wholesale. How sustainable is that as we head into the fourth quarter and into '05?

In terms of the dollar (MULTIPLE SPEAKERS)

In terms of year-over-year (MULTIPLE SPEAKERS)

Per car level or it that (MULTIPLE SPEAKERS)

That, too, and more of the year-to-year gains in the margin. (MULTIPLE SPEAKERS)

I would not expect continuing gains. I would expect that what we have achieved will hold and as we come around and anniversary on that, I would not expect continued gains but I think what we have achieved we believe will be a consistent performance.

Barrett:Just forget the fourth quarter benefit (MULTIPLE SPEAKERS) ACR change.

Fourth quarter and any and the first quarter?

A little in the first.

So some in the fourth, a little bit in the first but after that will have been anniversary donors is what Keith was saying. Going back to your store question it looks like four of them will be in the first quarter -- March, April, May. Two and the second quarter; and three in the third quarter.

When, exactly, will we fully anniversary DRIVE. Will that be in the third -- second quarter?

Fully anniversaried to roll out? August. Actually even during August we were still rolling out, right? So really September next year, third quarter next year.

Last question to follow-up Adam's question on the long-term profitability, at what point will we begin to see some SG&A leverage? I understand the maturity of your stores; but some of the stores that you rolled out in '01 I guess will be approaching that (technical difficulty) soon. It would be nice to start to see some SG&A leverage and really make the earnings power of the Company a lot stronger. What are your thoughts on that and --?

You can go through the cycle. We open two in February of 2002 and five in the following year, ten in the year after that, nine this year. That right?

Four, I mean -- sorry tow, five, nine and then nine (MULTIPLE SPEAKERS)

Okay two, five, nine and nine. You can work through the math of how they work their way up the cycle. And as they do that -- what we have said we are about a year to 18 months away from getting to what I would call the steady-state growth point. Where you now have a full set of stores, a full year's worth of stores if you will coming into the mature base. So each year that you add stores, you'll also be getting something out of it. So each of these last several years, we have been adding new immature stores but not getting fully mature stores out. We are about 18 months away from being at what I would call, what I would typically call mature growth if you will.

I would remind you that each of the last few years we have been adding incremental SG&A, as we have taken on the expenses of being a stand-alone Company, and even next year we still have the last piece of that coming from doing our own (indiscernible) center which will be a few million dollars. So that is a challenge in leveraging SG&A expense.

Those are exogenous challenges from the operating business if you will. Really, the key will be, how good are used car comps in terms of that -- if we had a nice 6 percent comp year, that makes all the difference. If we have four, it will be a fine year okay year but we won't get great SG&A leverage. If we were fortunate enough to have 8, we would hope to have quite substantial leverage. So it's really on how strongly the year comes out.

Matt Namer, Thomas Weisel Partners.

Quick question on the mix of what you're buying at auction. Just wondering with the decline off lease, if you could share with us what percentage of cars are rental or repo or other sources?

In general rental is typically 5 to 10 percent, but more than that, actually, I wouldn't share because that is something we consider proprietary.

On that same point, have you benefited much from maybe a change in mix towards smaller cars? I know that the compact wholesale prices have been really strong this year. Just wondering how much of that is in the numbers? Maybe what your mix is, in terms of passenger car vs. SUV?

Yes, well, and it may sound perverse to you but the increased demand for compact cars has driven compact car prices relative to new cars up to a somewhat disadvantaged level. And, perversely, the weakness in SUVs has in many cases particularly with very large SUVs sometimes driven the large SUVs down to an advantage level because to some degree you would say the market was oversold. Relative to what people were willing to pay for them in the used car market. So, I don't think we have seen saying anything that says our sales bump is due to increased volume of compact cars. There is more demand for them, but everybody sees that. And so the prices have been quite strong and there are no more of them than there were before. So it really hasn't changed our mix that much. Maybe at the margin but not a big factor. That's a wonderful thing about the used car market is price adjust a lot of these things. So that it stays fairly stable.

Just some housekeeping questions. Does the fourth quarter guidance include the impact of any of these franchised dispositions or is that all done in the third quarter?

As far as earnings, it is all done in the third quarter. As far as sales, it is inconsequential. They are all new cars anyway. We hardly make any money on those. It is kind of a charitable business we run. Seriously, with Mitsubishis, we haven't made any money on Mitsubishi in a while. If anything, it'll help.

In terms of your transfer fees. Can you raise those? Do you need to raise those to offset higher gas prices or transport prices or any thought on that?

Well the thought is you can always raise a fee. If you lose a sales to raise a fee and recover $10 worth of gas it has not been a good decision. We have been, we believe that our ability to transfer cars between stores is one of our most fundamental competitive advantages. We have seen that continuously reinforcing the market. So we are -- rather than necessarily add that as a fee, I think what we are continuously looking at is can we still get our net margin? But one of the things we don't want to do is we don't want to introduce some penalty there that would hurt sales inappropriately. So we are very cautious and thoughtful about how we do that.

Lastly, your internal auctions have been really successful. I'm just wondering if that is something you might put on line at some point? Some of the auction companies are doing pretty well with online auctions. I was just wondering if it is something you could bring to CarMax auctions?

The CarMax -- there are two places where online auctions have been successful -- or successful may not be the word. Have achieved a certain degree of success in the wholesale market. One is at the very high-end with either very high-end off lease cars or off rental cars where there's very limited variation in the cars themselves; and I would call the success there modest at best. With the possible exception of Mercedes and BMW. But in general there is some modest success there and then there's a very low end with salvage where the condition doesn't matter anymore because the car's wrecked, it is nondrivable and you're buying it for parts.The average car that we sell in our auctions is $3,000. It is a drivable car and the dealer's evaluation of condition is absolutely critical to their decision on price; and what we found is that, among our particular customer base, there is very little if any interest in online auctions. We will probably try to test, making an online capability available so someone can join the physical auction from an online site. But based on all the feedback that we have gotten from the dealers that come to our auction we think it is very unlikely that that will become a significant source of demand because condition and exactly how dented or scratched or what the does engine sound like? What do the tires look like all. Of that is so critical to value determination on a $3000 car which typically is going to the buy here, pay here industry that there are just not many people who believe you can do that online.

Scott Nessen with Lehman Brothers. Scott Nesson:My question is on the mix of new stores in the comp base and, arguably, with a large number of stores coming into the comp base fairly soon. I just want to get a sense of how the stores are performing relative to your pro forma model?

In general, both on sales and on economics the stores are performing in line with where we would expect them to be as a group on both the sales and economics. Specifically, if you said, "Well, gee how about the period April to August?" Everything underperformed. And those underperformed just like everything else. Taking the broader look, both before and after, we would say that both on sales and on economics those stores are roughly where we would expect them to be.

And would that be assuming that even during the soft period in April and beyond that, those stores were even a little bit about the corporate average or in line?

Maybe a little above but it is hard to say stork by store and I don't know that we have even looked at that. The answer is they were absolutely affected; and I think the macro trumped the micro during that period of time.

Bradley Brainard (ph) with BML Capital Management.

Can you just comment on is there anything you on the competitive front and number two, your buy percentage of cars you appraise, is that going up or down?

Nothing consequential, really, on the competitive front. It continues to be true that there is no consequential format to format competitor out there. Most other people are out there running something that looks like a CarMax or the few kind of leftover Drivers Mart franchisees who when AutoNation shutdown the particular dealers that on that franchise decided to keep running it and they have been out there for a while and that hasn't really changed. I'm sorry, the second part of your question was?

The buy rate.

Our buy rate in general over the last couple of years has gone up a bit and we continue to try to find ways to increase our buy rate and for the last couple of years it has gone up nicely.

Do you give a percentage of, in the past I was checking my notes on the amount of cars you get from the appraisals the buy rates or auctions. And what is that mix again?

Roughly? Between 50 and 55 percent roughly. The cars that we sell at retail are cars that we are able to buy at the store. One of the things to recognize is, our ability to buy cars from consumers depends on awareness. Making consumers aware of the fact that we will make an offer on any car and we will buy a car whether they are interested in buying from us or not is such a novel idea that we see that there is a lag in a new market before that awareness really penetrates. So new markets start out at a much lower sell supply rate but ramp up fairly quickly over the first couple of years to sort of get in line. Net net, we have stayed in that range, actually, some of our older stores may be improving a bit above that, but the new stores bring the average down.

One last question. On the buy rate what percentage of it is -- and I don't even know if you guys give this out or break it down, of people who just come in there and say I am not looking for car I just want to sell you this vs. people who actually trade it in?

Roughly, I believe -- I think the rough number is 60 percent of the cars that we buy, a little more than half, let's say, are from people who don't buy a car from us. Now that doesn't mean that they are not buying a car. Our view is in a sense less focused on the percentage -- I want everybody out there who is bound and determined to buy from somebody else to still give us a chance to buy their car. So if you are sold on the the new Honda Odyssey minivan I still want to buy your used car. So we don't -- that's the observed percentage. We don't focus on what it is so much as focus on what we would like to get is 100 percent penetration on appraisals of all used cars put up for offer in the market. We are not nearly there yet.

Eric Genence (ph) with Rockefeller Company.

Okay and Eric I believe you are the last question we can take. Eric Genence:I had a question just about the first time you've really spoken at any great length to the maturation of a store base. Could you characterize for me how long it takes for a new store be it a satellite or a stand-alone to achieve maturity levels of productivity? And at what -- let's say that is four years, what percentage each year it achieves?

Yes let you refer you to an 8-K that we put out in July of 2003. Which is called the CarMax Investor Note because we went true that in great detail and rather than replicate it for you on the call. If you will have a look at that, I think that will answer the model that we use and that we have laid out there for folks. And if you have further questions than that, Dandy would be glad to elaborate if you'll give her a call.

Okay and there has been no material change since allocation in terms of how that read?

We think that that model is still roughly the best model to use for understanding overall the business. Thanks very much I appreciate your joining us. Bye.

This concludes today's CarMax third quarter earnings conference call. You may now disconnect at this time.