車美仕 (KMX) 2006 Q3 法說會逐字稿

At this time I would like to welcome everyone to the CarMax 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 session. (OPERATOR INSTRUCTIONS) I would now like to turn the call over to Dandy Barrett, Assistant Vice President of Investor Relations.

Thank you for joining us this morning.

On our call today 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, prospects, and financial performance are forward-looking statements that we make relying on the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995.

These statements are based on management's current knowledge and assumptions about future events.

They involve risks and uncertainties that could cause actual results to differ materially from our expectations.

For additional information on important factors that could affect these expectations, please see the Company's annual report on form 10-K for the fiscal year ended February 28, 2005 and our quarterly and current reports on file with the SEC.

Now I will turn the call over to Austin.

Thanks, Dandy.

Good morning.

Thanks for joining us.

I know some of you have had some traffic challenges this morning, so we doubly appreciate your being here wherever you are.

Today we reported third-quarter sales and revenues, increased 17% including a 13% increase in total used units and a 3% increase in comp used units.

I will point out that on a two-year basis the used unit comps were up 5%, which was higher than the 3% two-year comp of the first half.

Traffic was up and execution continued to be strong.

We sustained positive momentum even as the cross shopping benefit from the new car employee pricing promotions ended.

We saw a fair amount of sales volatility during the quarter including a significant slowdown in the last ten days of November that dropped our comps for the quarter from 4% to 3%.

The first three weeks of December have moved back to a fairly solid trend.

The new car unit sales and comps were down, reflecting the general trend in the overall new vehicle market.

Wholesale sales increased 31% driven by both higher average prices and unit growth.

We did not see much of a deceleration in overall wholesale pricing as normally occurs in the fall despite a significant drop in SUV pricing.

We believe contributing factors included the strength in compact and midsize car prices resulting from higher gasoline costs and the fear of those costs and the limited availability of 2005 model year closeout vehicles following the success of the new car employee pricing programs.

Wholesale unit growth was consistent with our total used unit growth and we achieved solid improvement in our appraisal purchase rate, on a trend basis from the summer it was actually down slightly on a new year-over-year basis.

Our overall gross profit per unit increased approximately $200.

This level of increase was unusual for what is normally our toughest, most volatile quarter.

Our used car gross profit per unit was similar to last year, while our new car gross profit per unit declined modestly.

Our wholesale gross profit per unit was up sharply however, as we normally do in the third quarter we adjusted appraisal offers to incorporate the anticipated typical drop of wholesale prices, however, our wholesale auctions reflected the more modest actual price declines in the marketplace, giving us somewhat higher profit on wholesale.

Gross profit on other sales and revenues also improved primarily from higher ESPs which carry 100% gross margin the way that we report them and higher service margins, which benefited from greater overhead absorption.

As far as CarMax Auto Finance goes, income was up 37% benefiting from a growth in unit sales and managed receivables, a $0.02 per share benefit from favorable valuation adjustment due to loss rate improvements.

We had a similar benefit of $0.01 per share in last year's quarter.

And a $0.01 per share benefit from the new public securitization issued in September compared to expectations.

For the quarter the gain on loans originated and sold was 3.6%, slightly higher than our original expectation of 3.5.

The gain spread last year was 3.5%.

SG&A ratio was up 11.4% this quarter, up 10 basis points from last year.

The modest growth in unit comps was not sufficient to provide SG&A leverage.

In addition as we have noted before, the high ratio of stores in our base that are not yet at basic maturity, which we define as four years, was 49% this quarter compared to 40% last year.

This continues to adversely affect SG&A leverage.

This factor will begin to flatten out in the third quarter of next year.

As far as fourth quarter expectations, we now expect used unit comp growth in the range of -4 to plus 2% assuming we don't experience abnormal winter weather events.

This would give us a range of positive 4 to positive 6% for the full year.

We are up against a challenging 12% comp in last year's unusually strong fourth quarter.

You may recall that last year's comps included the 5 percentage point gain related to the rollout of our DRIVE sub prime financing source.

We do not anticipate any additional comp pickup from DRIVE this year.

Given the volatile state of the domestic new car industry, we continue to be cautious and concerned about the potential for large short-term disruptive actions from the domestic manufacturers.

We expect fourth quarter EPS to be in the range of $0.25 to $0.31, compared to $0.28 last year.

This implies for a full year EPS in the range of $1.27 to $1.33 representing annual earnings growth of 19% to 24%.

We anticipate fourth quarter cap gain spreads at roughly 3.5%, similar to quarter three, and still at the low end of our normalized range.

The gain spread in last year's fourth quarter was 3.7%.

As we noted in the release we have decided to stop providing quarterly guidance at the beginning of this coming year and will provide only annual unit comp and earnings expectations.

We have received a lot of feedback in recent months from shareholders and other parties on our comp earnings and guidance practice, most of which we solicited.

We recognize that over the last two years the market has been more volatile than historically, making it more difficult for us to accurately forecast our short-term performance.

We believe this has contributed to some volatility in our stock.

We would prefer to place the emphasis externally as we do internally on the longer-term growth opportunities and the store base, sales, earnings, and returns.

Our growth plan continues as expected.

We opened Virginia Beach and Wichita as standard superstores in the third quarter and added satellite superstores in Miami and Nashville.

We opened a total of nine stores for the fiscal year representing a 16% increase in our store base.

We have no openings scheduled for the fourth quarter.

In the fiscal year ending 2007, we currently plan to add 11 superstores, another 16% increase in the store base.

This will include two more stores in our L.A. market in Burbank and Torrance, our first entry into the Northeast United States, first in Hartford and then in New Haven.

They are both in the same TV market, so those of you in New York and Boston will actually have easier access to a CarMax store whether you want to buy or sell us a car.

And includes our first test of a CarMax model in a smaller market.

We will be opening a store in Charlottesville, Virginia, which has a TV viewing audience or DMA of approximately 185,000 people.

This will be a smaller physical format with different staffing and different inventory levels.

We will be testing this to try to understand how the store, both stand-alone small markets, as well as smaller trade areas in our larger markets, we expect to learn quite a bit from this and this will help us continue our growth plan and get into some areas that we currently can't get into.

So with that I will be glad to take some questions.

(OPERATOR INSTRUCTIONS) Matthew Fassler, Goldman Sachs.

It's Matt Fassler and Mark (indiscernible) here at Goldman Sachs.

We have a couple questions for you.

Let me start out by asking you about the wholesale business because that seems to be where there is a lot of variability if you will in the P&L.

It sounds like you had a sort of unique arbitrage in the third quarter because your volume was particularly disciplined based on your models.

I guess the sellers bid.

What is your expectation in terms of the typical seasonal trend in wholesale grosses that you would see I guess, typically your wholesale margins typically your wholesale margins would go up sequentially from Q3 to Q4.

Given that they were high all year and particularly high in Q3 would you continue to expect that to occur?

My guess is -- why don't you answer what we're expecting.

A portion of that we will expect because one of the things we did is adjust the level of offer to our consumers, so some of the sustained benefit you saw early in the year in the first half certainly rolled into the third quarter and we anticipate gaining in the fourth quarter.

In the fourth quarter we do expect to be positive.

It's just that you're right, we did have a convergence of events that may make the over performance relative on a year-over-year basis less stellar, let's out it that way.

When you say (indiscernible) you expect to be positive, you mean versus the third quarter or versus the prior year?

Versus the prior year.

I think that just the adjustments we've made and how we put the offer together for the consumers is an ongoing benefit that we will sustain.

That is just basically getting smarter in a structural way than you have in the past?

Yes, let me comment quickly, Matt, on a couple of the wrinkles in the third quarter.

As you know, SUV prices were literally in collapse through the latter part of the summer and the early part of the fall and while we did quit buying SUVs in off-site auctions, we continued buying SUVs at our stores.

And for some period of time we think we were about the only people out there willing to make appraisal offers, particularly on large SUVs.

And as always happens, as you have heard us talk about a lot of times before, at some point price gets to the point gee, this is a pretty good deal.

And we saw SUV demand turn around during the quarter.

So the fact that we were in the market and demand turned around there, whereas demand for compact and mid-sized cars continued significantly higher during this time of year than we've ever seen it and therefore pricing, both of those things were kind of drivers behind that wholesale performance.

So it was not only our ability to put the right money on the car, but also to some degree our willingness to stay in the market making wholesale offers when nobody else was doing it on SUVs.

Got you.

I guess the second question that I would like to ask relates to your sales forecast.

You sort of opened the Pandora's box by talking about some of the shorter term trends, so I will try to wedge it open a little more.

You said that clearly it sounds like the last ten days for them to drive the quarter down by a point must have been negative.

You are talking to a rebound in the first three weeks of December, yet your guidance seems kind of rooted in negative territory in the aggregate.

Does this relate -- is that forecast just a conservative forecast or does it relate to the fact that DRIVE really kicked in after the new year?

First and foremost it is driven by how good the fourth quarter was last year.

I mean the fourth quarter was really good; it's a really tough comp.

I always tell you you can't judge everything by one year's comps, but even when you take DRIVE out it was a significantly better quarter than we expected.

And when we do the run rate trends, trying to level everything out, that is the primary driver of what we are forecasting.

We are taking into account the first three weeks in December, but we are also taking into account that any three-week period in the third quarter could've been in a whole variety places all over the map, so we don't want to put too much weight on that.

So I would say that it is modestly conservative but pretty much on trend.

On average I would say historically we do expect some level of weather to affect sales.

The last year we did not have a material impact on sales was actually favorable, so just our run rate normally puts us at a conservative relative to last year in that it does anticipate some snow affecting some stores over the weekends and negatively impacting business.

A final quick question.

If you look at SG&A, the SG&A ratio relative to sales and (indiscernible) but if we choose that one you had slight negative leverage, expense ratio slightly higher relative to sales, backing out CAF on a 3% used unit comp.

What do you think next year you need to do to hold the SG&A ratio flat?

Given the store maturation point you made but also I would think given the (indiscernible) some of the separation costs.

I think the biggest thing would be the store maturation points.

If you are going apples-to-apples where the percentage of immature stores is about the same, I don't know but I would hope that around 3% if you didn't get leverage you would at least be flat.

You seem to intimate that for three quarters of the year, the better part of three quarters, the average store could be a bit younger.

Is that fair to say?

For the first two quarters.

Third quarter it will be flat.

So it averages out basically as a (indiscernible) full year.

So on average you're saying for full year fiscal '07 store rate should be a nonfactor.

Certainly for the second half, it will be a nonfactor.

First half, (multiple speakers) minor negative factor in the first half.

Thank you very much.

Sharon Zackfia, William Blair.

I wanted to follow-on Matt's question on wholesale margins, so is this -- should we think of this now as a mid-teens growth margin business on an annual basis?

I know there is seasonality there.

Yes, that sounds a little strong.

This was an unusually strong quarter.

Not only did we have the SUV benefit from a higher mix of SUVs because they bottomed out and then rebounded significantly during the quarter.

We actually saw price appreciation on the high fuel efficient cars during the quarter, which is also very unusual for this time of the year.

And then the benefit of really not there being a low supply of closeout models I think also created a unique environment.

So I think double digits appropriate.

I don't know about mid teens.

You have done I think 14, 15 the last three quarters before this?

And I guess that's the question historically has been more like 11, 12 and now it's kind of taking this inflection point up over the past year really.

I understand November is a little bit extreme.

And the way I think the way to think about it is, there were a lot of things that happened this year where to some degree we were trading off wholesale margin versus retail margin.

And part of how that sorts out next year will depend on what the marketplace does and what the best way to respond to that is.

I mean, we saw that that turned out to be a very valid strategy for addressing some of the curveballs the marketplace threw at us this year.

The third quarter was a little different, but I think you would not want to focus on just that wholesale margin in isolation from used car margins.

Okay, that makes sense.

Then geographically I guess any difference in regions other than what you mentioned with hurricanes and particularly can you update us on how Los Angeles is doing?

L.A. is doing fine.

We are quite happy with L.A. and the performance of L.A. is consistent with what we would want to do to continue opening stores there, continue with the full advertising program.

And as I said we've got two stores coming there.

Chicago and the midwest have been tough markets all year, and all of the data that we have say that they have been tough markets for everybody, particularly Chicago.

So that continues to be a challenging area; not from a marketshare point of view but just from a sales point of view.

I suspect that that has something to do with the stronger performance of domestic cars in the upper midwest and the fact there has been so much promotion in that area, but I haven't been able to get the numbers to tie that down.

Then on the opening plans for next year, are those going to be pretty equally weighted throughout the year?

Is there any lumpiness?

Was that not in our release in terms of dates?

I don't think so.

(multiple speakers)

Would it be fair to just spread them out?

I think it is not as smooth as last year.

Except we do have fourth quarter openings next year which we didn't have this year, so there are some openings in every quarter I believe, but the other think I will tell you is even where we sit between now and when they actually open we will still have some more things move around.

But it is reasonably spread through the year.

When are you opening the Charlottesville store?

Charlottesville is toward the end of the year, in the fall, November I believe.

That is our best guess.

Thank you.

Scot Ciccarelli, RBC Capital Markets.

A couple quick questions.

First have you seen any change in the cannibalization trends as you have multistore markets?

I don't think one that we could talk about as a uniform trend.

I think the adjustment we made in our thinking previously is part of our model now, but beyond that every market is unique, the things that drive cannibalization between two stores really have to do with shopping patterns, geography, where the rivers are, where the highways are, etc.

So there's nothing I would point to as a trend that I think is meaningfully different.

Okay, that's fair enough.

Comfort with inventory levels?

Obviously the market has been changing a lot.

You've mention your ability to buy SUVs, the buy of last resort if you will.

What is the comfort level with inventory levels at this stage?

I think we are very comfortable with where we are on inventory right now.

We literally technically at the end of November we had a little more inventory than we expected because the last ten days were softer than the trends had been, but then December picked nicely and we were right where we want to be and this is pretty much typically the bottom of the market.

So it's a good time for buying inventory and we bought as you know because there were fewer auctions, and because of holidays, we buy and bills a little early to be ready for the 13th week, the Christmas to January 1st period which is a big sales week.

And we are right where we want to be for that.

The last point is and I guess it is a little bit more on the philosophical side, but is there a point where you think your business becomes less volatile?

I guess I have been under the assumption as it continues to grow and mature and more stores etc. we would wind up seeing less volatility.

But it doesn't seem to be playing out that way.

Is there a model in your minds that at some point you become less volatile and more consistent?

Look, there's no question that scale helps and the bigger you get -- I mean just looking at anybody in the retail industry, the bigger you get, the more spread you are across regions, the more balances out.

There's just no question though, you read the newspapers.

There's a knockdown dragout to the death battle going on in the auto industry.

And it is clear that two of the largest auto guys in the world are really struggling to find the strategy.

And I think that just introduces some unanticipated volatility; how long that takes to play out before it stabilizes, I don't know.

But everything we see says that if you step back and look at it on a year-over-year basis it is really short-term volatility, not long-term volatility.

So it doesn't have any fundamental impact on the model.

It's one of the reasons believe me, that what is going on in domestic industry is one of the reasons we've decided, gee, trying to outguess the quarter for you just isn't productive and we want to focus on the year and get you to focus on the year as much as possible.

I know you have shareholders who want to see closer information, but we don't see anything yet that says it's going to get easier for us to forecast quarter to quarter.

But I think year over year, look, right now if we hit our range we will be at to slightly below the midpoint of the range for comps that we expect on the year.

Okay, great.

Thanks, guys.

Kind of top end of earnings, Keith points out.

Mike Heifler, Deutsche Bank.

A couple questions.

Austin, could you talk about any changes that you're seeing to the competitive landscape?

Specifically we have been hearing from the NADA and folks like Auto Nation that there's this renewed emphasis on the used business to boost overall margins and I am just wondering if you guys are seeing more activity from the franchise guys.

I think if you were to step back and study Automotive News for a ten-year period, you would see every time that new cars are bad, there is a statement by various auto guys that, well, we have renewed our focus on used cards.

It has never been clear to us that that actually means anything in terms of certainly not in terms of change in fundamental behavior or the way they address the consumer or anything else.

How much they inventory really is a function of how much risk do they think they are taking.

So it is a very dynamic and competitive market out there, but we have not seen anything that suggests systematic changes in the way people are actually selling cars.

Auto Nation has announced that they're going to be friendlier with customers, but as far as we can tell that doesn't mean they're going to quit negotiating or any of the other things that make it -- they're not going to change their pay plan for instance.

And when the salesperson's pay plan is a function of the margin on the car, it is pretty darn hard for the salesperson to be on the same page as the consumer.

So there's not anything that we see globally at this point that really represents a trend there.

And we certainly haven't seen anything that suggests people are either changing their sales methods or fundamentally altering their inventory levels.

Its good competitors and the best competitors are usually the private guys in each market, always focus on the used car market.

They don't ever take their focus off used car market and they are tough competitors every day.

Those guys are always out there.

They tend not to be public companies.

Okay, on the wholesale side of the business in terms of the buy, maybe you could just touch on what you're seeing at the auction level.

We've been noting that the auction inventories have been going down, and I'm wondering if you would attribute this to retail demand or do you think it is fewer vehicles still coming off of lease and perhaps fewer program vehicles coming back for remarketing?

Well, I think there are some fewer program vehicles.

Just because of the hurricanes and things like that, people are -- I think the rental car companies certainly kept their vehicles as well as long as they can keep them, and there's a healthy demand there.

And certainly we are still in the cycle where there are fewer off-lease cars as we are coming to the end of that cycle.

And there has been pretty healthy demand in the wholesale market.

That is why prices held up so well.

Having said that, this is the low point of the market every year, so relative to what we need to buy, we are finding plenty of vehicles out there, but it is not as soft a market as it typically would be in December.

Would you care to make a prediction for next year?

I think you know I used to be an economist and I quit doing that, and that's why.

As we have noted, we are not any better at predicting than anybody else.

Okay, thanks.

Bill Armstrong, C.L.

King and Associates.

Austin, could you comment on the spreads between new and used car prices, how that progressed during the quarter given the fact that the Manheim Index and other indexes did indicate rising rather than falling used car prices?

I think that the best indication we have is that actually new car prices in real terms also went up during the fall.

So to the degree that that was true, it allowed used car prices to rise.

As I said, the reality behind the index is by category of car, because when you disaggregate into the different types of cars, what you actually see is the compact and mid-sized were really strong.

Prices were higher this fall than they were last January in absolute terms.

So you were paying more for a car nine months later than you were back in January, whereas SUVs, large SUVs where there is no description other than a total collapse in pricing.

So while that averages up to an aggregate trend, what is really meaningful is the trend for each specific type of vehicle, and that was driven by all these macro factors.

The compacts and mid-size were driven by people's concern about gas prices and to some degree about by all the replacement volume, at every store going into Mississippi and Louisiana.

There were a lot of cars destroyed there and pretty much everybody there needed to get a vehicle.

Some people had to get vehicles that didn't have them before, because if you lived in New Orleans and didn't have a vehicle and you'd moved to Baton Rouge or Jackson or Atlanta, now you've got to get some kind of car.

And I think that particular type of car, particularly the compacts given the gas price issue, they were very strong.

So I do think that new cars gave some leeway there and the lower availability of closeout gave more leeway, but it was a whole variety of trends at the individual model level that was the real driver.

Have you had any difficulties in obtaining sufficient supplies of mid-sized and economy cars?

Well, the answer is how much do you want to pay?

The answer is we have not had a difficulty.

We had to pay more through August and September certainly and into October.

We actually reversed our policy at one point.

I said earlier at one point we were not buying any SUVs at auction.

During the fall we reversed our policy and we quit buying compact and mid-sized at auction because we saw the price value beginning to deteriorate.

So although the prices remain fairly strong, the sell through rates softened somewhat and we were able to supply our needs from direct purchases at the stores.

So we have not been unable to supply what we need.

The challenge is always price and the question is always the trade-off there versus something else you could have in that inventory slot.

But in general actually we are extremely happy with how we came through the fall on all of that.

It was a very dynamic fall and we are very pleased with how we were able to manage through that.

Okay, just two others.

You mention the last ten days of November you saw a significant slowdown.

What do you think was behind that?

I have no idea.

I mean at first I was running around getting everybody to do analysis on red-tag pricing just to see is there any possible way that that could be having this effect.

We certainly didn't see any customer flow from it.

But all our analysis indicated and I think GM's results indicated that that wasn't it, so truthfully I have no idea.

It's just traffic slowed down at the end of November and picked right back up at the beginning of December.

It's one of those mysteries you have now and then.

One final question.

I see you are opening a store, your first store in the Northeast.

I remember asking you a year or two ago about that and you said that like many Sunbelt-based retailers that are expanding, the Northeast would be the last place you would start opening stores, so I was just wondering what the rationale was for you to go up here at this relatively early stage of your Company's nationwide expansion.

You probably slightly misheard me.

I'd certainly say New York is the last place we will come in all likelihood, not because we don't like it but because you got 20 million consumers there and you have to buy the entire TV market.

And what we have said in general in the Northeast, in general the Northeast is slower growing.

The problem with Philly in Boston is not that they are slower in growing and they are smaller than New York, they're just tough markets to get into.

We will eventually get there.

New York will probably be last as we will wait till we are on national TV.

The two markets that are most attractive in the Northeast in terms of economic performance, market attractiveness and scale of the TV market, so that they are relatively easy to enter with a small number of stores are Providence and Hartford/New Haven.

So if I said what you just repeated, I misspoke because I was talking through shorthand.

We have said before that problems in New Haven were the ones in the Northeast that were on the relatively near list if we wanted to go there.

Now the bottom line in terms of why are we going, we went out there and we had a look and we found some real estate that was attractive.

And we found a good piece of real estate in Hartford and a nice piece in New Haven.

We are able to store both markets which will give us good media efficiency.

And we thought that was a good opportunity.

It's a good opportunity to sort of test out the Northeast, have a nice market up there and also I don't think it will hurt at all to make it easier for the investment community to get there, but that is not why we did it.

We did it because we think it's a pretty good little market.

Well good luck with that and I am glad now I can drive to one of your stores instead of flying there.

So are we.

John's Zolidis, Buckingham Research.

Just want one clarification.

The guidance you provided at the beginning of the year, that was for earnings of $1.20 to $1.30, is that correct?

Correct.

We're now looking at $1.27 to $1.33?

That is correct.

Okay, great.

On the small market you're entering, Charlottesville, can you talk about what you're going to do differently there and what you think the longer-term opportunity is in terms of the number of stores you can have in small markets, assuming that is successful?

Yes, one of the things we're doing with Charlottesville is we're kind of skipping over a whole range of intermediate markets and going to -- right now the smallest trade area we have ever gone into is Fayetteville, North Carolina, which is about 375,000 people.

And it is within the Raleigh television market, so we have already been advertising there for ten years.

We are skipping over a whole range of medium-sized markets to try something quite a bit different in Charlottesville, which is a store with a physical format about half the size of our standard satellite store.

We're not sure exactly what the inventory will be, but I am guessing in the 150 car range as opposed to 250.

And we will use a different staffing model for management, where we will size the management to the sales level and what that will probably mean is some of our managers will have to do dual tasks.

So they might work as business office manager one day and a sales manager the next day.

We are pretty sure we can make this work and the numbers look pretty good.

The reason we're going to Charlottesville is it is right down the road from Richmond where we have been for 13 years and right down the road from Washington, where we have been for ten years.

So it is a market that is pretty familiar with us and in some sense it is the lowest risk of the very small markets we could go to.

It is also a very self-contained market and it has nice demographics.

So we think it is a good place to do a relatively low risk but significantly different test.

In terms of how many markets will this open up, if we can do it well it will certainly open up -- if you grab yourself a book and look at the size of DMAs, Charlottesville is way down the list toward the bottom.

So it would open up everything between Charlottesville and 1.5 million people.

The other thing that is harder to estimate is how many small trade areas within large markets will it open up because then the question gets to how much inventory do you need to be in a really peripheral or small trade area of Chicago?

So we don't know that yet, but the answer is more.

And we think significantly more, so we think it is something that at some point we needed to do.

Now is the right time to test it, so that we can figure out how to make it work and get this into our portfolio of development.

Okay, great.

Happy holidays and good luck with the fourth quarter.

David Campbell, Owl Creek.

Thank you.

My question has already been asked.

Gabriel Holmes, Philadelphia Financial. (ph)

I have a question on your financing with DRIVE financial.

I understand the contract came up for renewal on December 1, and you had a very favorable terms with them, I think it is only about a 4% discount.

Has that been renewed yet and has it been renewed on similar terms?

We have an ongoing arrangement with them and they have had to make some minor modifications to the program, but overall it will be very similar and we don't expect any material change either to the profit ratio that it brings on an incremental basis or the level of sales.

Okay, thank you.

Hardy Bowen, Arnhold & Bleichroeder.

It seems like every time we lower the prices that we're going to pay for cars for our wholesale operation that it is successful.

Are we beginning -- this is kind of a unique offer that we have.

I'm wondering if we're thinking differently about the business; that maybe the wholesale business ought to operate at a profit.

Well, obviously right now whether you want to call it an over recovery of cost or a profit we're clearly making money on wholesale cars beyond what our net cost is.

And part of that as I said has been driven by our analysis of the last couple of years of what has gone on in the new car market and how best to respond to it.

Some of it has been cross subsidization from wholesale to used car pricing when it was appropriate.

And as you would imagine knowing us, below the surface of what you can see in terms of this aggregate number are dozens if not hundreds of tests that we're running to try to understand the subtleties of how we should price wholesale cars.

Because the one thing we never want to do is we don't want to give people an offer that is unfair.

We don't want to give an offer that would hurt sales.

We don't want to give an offer that would reduce.

We're trying to increase the number of cars that we buy from consumers.

Having said that, we do believe that it looks like this is a business where in fact the model that we have evolved adds value.

That a wholesale car sold through one of our auctions may actually obtain the highest price it can in the marketplace.

So we want to behave intelligently to the degree that is true.

But we try to be very thoughtful and cautious and we're running a lot of tests there.

It does seem like there may be some value added there that we ought to capture, but this has been an unusual year.

So that's why we don't want you to rush off and think this is what we should forecast going forward, because a lot of what we were able to accomplish in wholesale was really driven by some of the circumstances created in the new car market.

And therefore the late-model and used car market.

Is there anything different in marketing that we're thinking about given the kind of environment we have with General Motors and Ford which probably isn't going to change for the next couple of years?

What do we feel about our message getting across and do we think anything about that?

Yes, you know we continue to run lots of tests in advertising too, both in terms of -- as I think you know, every TV ad that we make we go out and test with consumers.

And if it does not score well on -- is this more likely to make you visit CarMax -- then we don't run it even though we've already made the ad.

We also have been doing a lot of testing across media to understand what daypart should we be buying, how should we run our ads within programs when we know we have a good ad that communicates?

What is the trade-off between TV versus radio, radio versus newspaper, and particularly newspaper versus Internet.

I would say globally with us just like with everybody else, newspaper is losing and Internet is winning.

So there is clearly a shift in media there.

The Internet is evolving so much particularly Google search and everybody trying to compete with that, that new opportunities are being created all the time.

Google local presents a really interesting new opportunity.

We don't have how that is going to work for us.

So there's a lot going on there.

I would say that by and large it is not driven by anything that GM or Ford are doing and I think they would be very pleased to know that you think things will go on like this for two years.

I think.

I think they may go out of business.

I don't know.

I don't think they are going to go out of business, but as you know I tend to be on the negative end of the spectrum in terms of my suspicion that at least GM will probably have to go through some sort of restructuring.

But our advertising is -- there is not a lot we can do or anybody else can do when GM does something like employee pricing that would change behavior.

As you saw, that was really good for us though because it drove a lot of people in the market and our appraisal offer, we upped the advertising on the appraisal offer.

That worked pretty well and we got a lot of sales, so really we're doing a lot on advertising but it's mainly to try to figure out how to get our message across more clearly and to more folks and to those folks who are actively in the market.

On Austin, Texas and Charlotte, how close are these units going to be to existing units and how much cannibalization do we expect out of that?

The Austin store -- our existing Austin store is on the far north end of the market.

It is actually right on the boundary of Roundrock where Dell headquarters is.

And the second store will be 25 miles away at the other end of the market, southern end of the market, on I-35.

And if you have driven through Austin or San Antonio recently you know that it's almost impossible to get through that 35 miles.

It is currently I think tied with I-95 as the busiest highway the country.

So we think there's pretty good separation in Austin.

In Charlotte the store will actually be in Gastonia, North Carolina which is a different town altogether.

It is in the Charlotte television area but it's actually a midsized town west of Charlotte, and we think that is a pretty good gap and a pretty good separation there too.

So these two my best guess would be, are not hugely cannibalistic, but that is just my best guess.

It looks cannibalization is really going to go down some by the end of next year because they don't have much else.

I mean, Fredericksburg, I guess (multiple speakers) --

Fredericksburg will potentially cannibalize some from Richman and some from D.C.

And you never know with L.A. L.A. has still got a lot of room and I doubt that we will see a lot of cannibalization from these two stores, but we do our best job that we can to predict.

I will tell you for instance we just built two satellites in Miami.

One of them cannibalized more than we thought and one has cannibalized quit a bit less, so it's always a little bit tricky to figure it out.

I would not try to change my estimate based on any expectation of change in cannibalization.

I guess we have a lot of midsized markets in the plan because this is what we are really finding is the best place to put stores.

No, the best place to put stores ultimately is large markets.

Our strongest market is unquestionably Washington/Baltimore.

Midsized markets are really good places as we've described when we started growth plan.

The great thing about midsized markets is they are on average lower risk because they are easier to enter.

You only have to add one or two stores.

And we've added L.A. as our large market in the portfolio right now and we are starting to work on some other large markets that would require 2, 3, or 4 stores to enter.

We have not announced them yet but I mean you can just look at the map and look at our prior behavior and you can pretty well guess what they would be.

And those will be coming over the next several years.

So we will continue to try to have a mix out there where a portion of our portfolio, 20, 25% or so will be larger markets in the future.

And putting two stores into Columbus, Ohio I guess in pre or prior years we probably would have put one in and then another would have followed a few years later.

We are you doing the same thing in Hartford?

Yes, in Columbus it's true that Columbus is a market that four or five years ago we would have tried to start with one store.

The way Columbus lays out, it works a lot better with two stores to begin with.

And so that is what we did.

And we got two great pieces of real estate we believe, so we did that.

Hartford and New Haven are two separate markets and its two markets within a single DMA.

And everything we have learned says all other things being equal that would be better to go ahead and start with both of them if you can.

Get more impact that way.

Yes, Winston-Salem, Greensburg, you get better advertising leverage from day one.

But in general we're finding if you can clearly identify two distinct trade areas like in Las Vegas or like in Austin, like in Columbus, all other things being equal we think it would be better to go ahead and do both of them to begin with.

Okay, sounds good.

John Murphy, Merrill Lynch.

You have touched on this quite a few times and I might ask it a little bit more directly.

In the event of a GM bankruptcy or a major restructuring here were they may go through a reduction of brand, just wondering how you might navigate that environment and maybe specifically sort of your experience with the Oldsmobile cancellation in the past, really how you dealt with that.

It had virtually no impact on us.

The problem that GM has and we have told them this, we are agnostic as to what cars we sell.

Whatever the consumer wants to buy, we will sell.

GM has always underperformed as a percent of our mix relative to the number of cars they are producing, which tells us that they are producing a bunch of cars nobody wants to buy.

And I have believed for a long time that they need to have at least one more and maybe two brands go away.

So as far as we can tell, Oldsmobile had no impact on us.

The excess cars that GM is making are the cars of last resort in the market, and nobody wants to buy them and they are not very attractive.

They have a hard time getting shelf space at our lots, so their disappearance if they weren't, if somebody else didn't come in and produce cars to supplement that, could reduce the overall level of new car sales and make a little bit less inventory available.

Each year's sales is only a small portion.

On the other hand, it certainly looks like the Koreans are doing everything they can to fill in the gap at the lower end of the price spectrum.

And Toyota just announced I think today that they decided to go full speed ahead with a 10% production increase this year, so I think the Japanese and the Koreans and to some degree the Europeans are more than willing to step in and try to fill that gap in volume.

And net-net I think that would be good for the used car marketplace because it would result in a mix of somewhat higher demand cars out there.

But I don't want to underestimate the potential volatility that that sort of restructuring might create.

But ultimately I think it would be positive.

Then on the wholesale property you've been posting, a lot of dealers have been using data provided by Manheim as far as the pricing their wholesale vehicles and their purchases.

Have you been increasing your use of that data or are you clearly just using your own data?

We have always used Manheim data.

We use more of Manheim data on a more timely basis than anybody.

Believe me.

And same thing from ADESA.

We use the Manheim data, the ADESA data.

We use the black book data from Hearst and remember we have an enormous source of weekly wholesale data from our own wholesale auctions, so we use all of that plus we use -- we capture every time make an offer on a car, did we buy it or not?

And we have that available to our buyer so that they can see, look, what appears to be the correct offer price on the market right now.

So we certainly incorporate Manheim data in depth and granularity and very quickly, but it's only one of the sources.

Then just real quickly on the Charlottesville store, it seems to us that there are plenty of markets large dense DMAs that you can work into.

I know you talked about this a little bit but what do you feel like you'll gain in the small market?

It does seem like you have plenty of room for growth in these larger markets still.

Just step back and look at -- take Boston or Philadelphia either one.

You probably need for either of those stores to open the market, you probably need a minimum of five stores, and how hard is it to get five great sites in Boston?

Ask Home Depot, ask Target, ask any of the guys that have done it.

We will eventually be able to do it, but its very hard to assemble those sites, have them ready, and open and be advertising efficient all at the same time.

So what you want to do is you don't want to go take on several of those markets at the same time.

And right now we're working on L.A.

At the opposite end of the spectrum there are hundreds of smaller markets and trade areas.

Some of those smaller trade areas are in large markets, so for instance Frederick, Maryland, which is part of the Washington Baltimore DMA so it is part of a large market, but it's 35 miles north of Rockville, so it's really operates as an isolated market.

To make a Frederick work we probably need a smaller store like we are going to test in Charlottesville.

So this is not a let's go to Hooterville strategy instead of Philadelphia.

It's part of having a portfolio of stores that will let us address every trade area, whether it is a separate small market or a small market within a large trade area, within a large market.

Does that make sense?

Yes, thanks a lot.

Scott Nesson, Lehman Brothers.

I had a question on the finance side.

I think twice now this year, once in the first quarter and now again in the third where you've been able to lower the loss assumptions on your previously securitized receivables.

Just trying to understand and get a better sense of the market conditions that have allowed for this and what you perceived on that front heading into next year.

I will let Keith address both our performance and the market conditions.

Part of the challenge is that as we evaluate our performance one of the things that introduces a little challenge for us to really assess what ultimate losses are going to be is the fact that we put in a new scorecard in 2003.

So part of what we had to do is discern how much is market and how much is our own new scorecard.

And as we gain more experience with that as portfolios mature, we can get more confident about our level of adjustments.

But generally if you look at the asset-backed securitization market you will hear that overall there has been about a 40% improvement over the last couple of years.

And in loss rates.

And we think that a lot of that is due to just the general economic conditions, mortgage refinancings where people are flush with cash for a variety of reasons, and lower overall interest rates.

So that is really the best we can assess from the overall market and it's really looking, drilling down on our own portfolio and trying to assess which is which and trying to make sure that we're making the appropriate adjustments.

Okay.

Do you think that the increase in short-term rates could lead to a moderation in these lower loss rates going forward?

If its moderation, I don't think it will be material.

I really think that it's more the overall economic impact.

The fact that mortgage refinancings really have fallen significantly and the cash coming from that and credit card balances were dropping, whether that goes back the other direction, all of those things put it more at risk in the future, so it's not really one single factor as far as just interest rates.

Okay, thanks very much.

A follow-up from Matthew Fassler, Goldman Sachs.

This will be the last question.

Matt, go ahead.

First of all, Austin, I know the Board is really running the show on this but in terms of the search for your successor, any update that you can give us?

It is underway.

We have a search firm that we're working with and we have all met with them and talked with them.

I think they will do a good job.

We knew some common candidates that I thought would fit and the headhunter thought would fit, which was a nice start that we had some commonality there, so it is underway.

And beyond that we will sort of see how long it takes.

Typically doesn't happen overnight, but we are all eager to start seeing some candidates.

Got you.

Second question is actually a follow-up from the last one on credit.

Just kind of looking at the history of your adjustments, is it the default rates that you had in fact it took down again in the third quarter?

Yes.

And I believe that the range just to try to get the magnitude of that, the range that you had out last corner was 1.6 to 2.16.

Do you know what the magnitude of the reduction was here?

I believe we're going down to 1.45 at the low end.

And does this reflect sort of your real-time experience?

In other words to what degree do you try to make this sort of a sustainable average as opposed to an individual quarter's experience?

Part of the reason we don't do this every quarter is that the timing of this particular adjustment is because we're going through peak season.

There is a seasonality to losses and so we look at it very carefully.

The good news is that our more mature portfolios, the early 2003s where we have the more experience on our data are getting so mature that they help us ground a better basis for how much of that is really attributable to our scorecard.

So the further those develop, the better we will get at being able to isolate the scorecard versus the market conditions.

And that is the piece that has really been challenging for us.

So there has always been a little caution until you see another years' worth of data and really that means going through this time of the year and seeing peak seasonal losses, that is when you can get another benchmark where you can gain more confidence and say yes, that is an ultimate loss rate.

As we've said all along, over the longer-term we believe the purity and integrity of the credit data that we're able to collect given no negotiation on the car, or the trade, knowing what negative equity is if any, should give us over the longer-term a better scorecard and a better ability to evaluate people.

And I think part of what we're seeing here is some of that long-term benefit.

So if we look at just to have the numbers in front of me, six quarters ago in the August quarter of '04 you were still at 2%, 2.0%.

And I think you have taken it down three times, once last year and twice now this year.

You would say that that is -- would you say that is more a factor of your own understanding of your own performance or of market conditions that have been as you described pretty constructive for you?

It is really the combination of those two factors and the other opportunity there is that we don't necessarily on new originations want to just write down losses.

What it does is it allows us to find opportunities where we should be buying based on doing more analysis on our own data and not necessarily -- help support sales versus write down losses.

The losses really are our best estimate of what we've already originated based on that data.

Okay, thank you very much.

And thanks very much for joining us.

Ladies and gentlemen, this does conclude today's CarMax third-quarter earnings conference call.

You may now disconnect.