Popular Inc (BPOP) 2003 Q2 法說會逐字稿

Welcome to E-LOAN second quarter 2003 earnings conference call. (CALLER INSTRUCTIONS) This conference is being recorded today. If you have any objections you may disconnect at this time. I would like to introduce your first speaker for today's call, Mr. Chris Larsen, Chairman and CEO.

Thank you for joining us for E-LOAN's second quarter 2003 earnings call. With me today are Joe Kennedy,our President and Chief Operating Officer; and Matt Roberts, our Chief Financial Officer.

Let me begin by giving you a very brief overview, followed by Matt who will review our financial results and forward guidance, followed by Joe who will provide some further color on our quarterly results and operational initiatives.

For those of you online, we have also made available a PowerPoint presentation, which can be found at eloan.com on the investor relations link.

The second quarter of 2002 was simply an outstanding quarter for E-LOAN. In the quarter which achieved an effective profit of $8.1 million, or 12 cents per share, on record revenues of $45.9 million. Overall our revenue increased 27 percent sequentially, and 119 percent from the same period last year, as we delivered our seventh consecutive profitable quarter. This represents a one cent beat (ph) to the high end of our previously increased guidance.

Most importantly, the fastest-growing parts of our revenue were in our diversified products businesses. In the second quarter, diversified revenue totaled $21.2 million, growing 41 percent in the quarter, and 104 percent year-over-year. In fact, our diversified product revenue in Q2 exceeded our total revenue for the entire company in the same period last year.

Purchased mortgage volume growth was simply spectacular, surging 67 percent in the quarter and 117 percent year-over-year, and growing much faster than the industry overall, meaning we gained substantial market share of purchased mortgage business, a difficult accomplishment during a continuing refinance boom. Purchased mortgage business is particularly important because it is expected to remain strong into the foreseeable future as the refinance boom winds down. Slide 4 of the presentation provides a graphic of the purchased mortgage market.

We also saw good growth in our home equity business, reaching a new revenue record, which increased 35 percent from the prior quarter.

We believe these performances are particularly strong indicators that consumers are increasingly adopting online lending, and that the category is on a powerful long-term growth trend, independent of short-term interest rate cycles. For example, a recent study we commissioned indicated an increasingly favorable view of online lending by REALTORS, a key influence group in-home financing. The independent study showed that realtors nationwide believed that their consumers are better served by online lending, had more options and saved more money. Further, the view buyers who use online lenders as better informed and easier to work with. As a result, a sizable a number of realtors nationwide are likely to recommend that their customers apply for loan online. That's a powerful change from just a few years ago.

We believe that the advantages of online lending are clear -- lower costs through technological driven efficiency; more transparency in terms, fees and process; and less conflicts of hassle, with pricing based on consumer's credit, not their negotiating skills. While we believe these benefits are obvious when compared to traditional lenders, we also recognize that lending is a big, scary consumer decision that requires consumer confidence, as well as consumer awareness. We believe that confidence is growing and will continue to fuel a migration of market share to our category and to E-LOAN as the embodiment of online lending.

We also believe that the E-LOAN model is far ahead of other lenders in delivering the features and benefits that consumers want. In fact, during the quarter a leading consumer finance education group created a new designation called the upfront mortgage lender, which was designed to certify lenders according to strict measures on pricing, disclosure, transparency and lack of conflicts. We were pleased that E-LOAN was the only mortgage company to make the cut.

Q2 was a great quarter for E-LOAN, but more importantly we believe we have the right model, the message and the right people to execute on execute on the tremendous opportunities that lie ahead. And with that, let me now turn the call over to Matt, who will discuss our Q2 financial and detailed guidance.

Before I begin the financial review, I'd like to remind you that during this call we may make forward-looking statements based on current expectations that involve risks and uncertainties. Actual results may differ from the results described in the forward-looking statements.

Factors that could cause actual results to differ include, but are not limited to -- general conditions in the lending industry; interest rate fluctuations; and the impact of competitive products. These and other risk factors are detailed in the E-LOAN's filings with the Securities and Exchange Commission's.

Now I would like to review a summary of our financial results for Q2 2003. Our earnings press release includes a detailed presentation for your review.

During the second quarter we closed 1.6 billion and sold $1.6 billion in loans. The mix of our sold loans among our three primary products was 77 percent mortgage, 11 percent auto and 12 percent home equity. Closed loans volume from diversified products was 47 percent of total closed loan volume, compared to 45 percent in the prior quarter, and 45 percent in the second quarter of 2002.

Slide five shows the highlights of our second quarter of 2003. Revenues totaled a record 45.9 million, an increase of 119 percent compared to the 21 million in the same period last year, and up 27 percent compared to the 36 million in the prior quarter. Of the total revenue, 80 percent was mortgage including interest income, 6 percent was auto and 13 percent was home equity including interest income. Diversified product revenue was 46 percent of total revenue, compared to 42 percent in the prior quarter and 49 percent for the second quarter of 2002.

Net income was 8.1 million, or 12 cents per share, on 66.7 million diluted shares. This compares to net income of 0.9 million, or 2 cents per share, on 59.1 million diluted shares for the same period last year.

Direct margin -- which is defined as revenue minus variable and fixed operations expense -- totaled 25.4 million, an increase of 163 percent compared to the 9.7 million in the same period last year and an increase of 34 percent compared to the 19 million in the prior quarter.

Slide six outlines our financial model targets and the significant progress that we have made towards those targets over time.

As for our balance sheet, total assets at the end of the quarter were 387.3 million, which includes cash and cash equivalents of 43.9 million, of which 2.5 million is restricted, and loans held for sale of 301.2 million. Total liabilities at the end of the quarter were 315.4 million, and included 293 million in borrowings related to mortgage, home equity and auto loans held for sale. Total stockholders' equity at the end of the quarter was $71.9 million.

Now I would like to update you on our financial guidance for 2003. Our earnings press release again includes a detailed presentation for your review.

On slide seven you'll see the assumptions that support our 2003 financial guidance. According to the most recent Mortgage Bankers Association forecast issued July 14, 2003, refinance volume was expected to decline 3 percent of the third quarter, with a further decline of 25 percent in the fourth quarter 2003. However, that forecast did not anticipate the recent significant increase in rates that occurred in the first quarter of -- and it didn't expect that to occur until the first quarter of 2004, at which point the MBA was forecasting that refinance volume would be down 52 percent from last quarter's record volume. As a result, we anticipate that the actual overall market decline in refinance volume may exceed that of the July 14th MBA forecast.

We expect our primary finance volumes will be similar to the second quarter's record level and to decline approximately 25 percent in the fourth quarter. We expect continued strong growth in our purchase mortgage volume of approximately 15 percent in Q3 and a further increase of 10 percent in the fourth quarter of 2003.

The second quarter of 2003 was our all-time record for mortgage revenue per loan. As we mentioned in prior guidance, we expect our revenue per mortgage loan and interest spread to trend down throughout 2003 to historical levels. This trend is a normal component of a declining refinance market, as the level of capacity increases to meet demand and overall competitive pricing pressure increases.

Home equity loan volume is also expected to continue its strong growth, as consumers opt for that one type versus a cash out refinanced to meet their needs.

With the transition of auto operations from Florida to California complete, and the support of an expanded auto facility in place, we expect auto loan volume to grow 20 percent in both the third and fourth quarters of 2003.

We expect to continue an aggressive marketing spend to build on the demand momentum in our diversified product revenues and capture the remaining refinance mortgage opportunity.

Given the above assumptions, we expect 2003 total revenue to grow to approximately 165 million, representing a 60 percent improvement over 2002 results on a substantially different revenue mix. Prime mortgage refinance revenue is expected to account for 48 percent of our total revenue in 2003, down from 58 percent in 2002. We expect revenue from diversified products of approximately 85.8 million in 2003, representing a 98 percent improvement over 2002 diversified product revenue. This also represents a 17 percent improvement over our prior guidance that we issued in April 2003 for total diversified revenue, and a growth level well ahead of our targeted 50 percent annual growth rate for diversified revenue. Our goal is to grow our diversified products revenues by 50 percent per year throughout -- through 2005, while increasing our share of the prime mortgage refinance market each year. A compounded 50 percent annual growth rate on our diversified product revenues over expected 2003 results will produce over 190 million in revenue on just these diversified products by 2005.

Moving to slide eight, we expect 2003 total revenue to grow to approximately 165 million, representing a 60 percent improvement over 2002 results, and we also expect net income of approximately 25.1 million, or 37 cents earnings per share, on approximately 68.5 million diluted shares in 2003. That represents 135 percent improvement over 2002 net income.

As shown on slide nine, the anticipated revenue in 2003 continues to trend at significant revenue growth for the Company versus our base of 1998 results.

Slide 10 highlights our anticipated progress on diversified revenue, as prime mortgage refinance is expected to account for 48 percent of our total revenue in 2003, down from 58 percent in 2002.

As shown on slide 11, revenue from diversified products is expected to total approximately 85.3 million in 2003.

Finally, slide 12 provides an overview of the improvement that we have made in two key areas of our business model -- operations and marketing efficiency. From 2000 to anticipated 2003 results we will have increased our direct margin from 12 percent to 53 percent of revenue, while decreasing our cost to acquire customers from 79 percent to 25 percent of revenue.

Now I would like to turn the call over to Joe for his operational review.

I would like to focus my remarks on two topics -- first, additional insight into our diversified product growth; and second, a brief technology update. Please forward to slide 13.

As Chris and Matt noted, our purchase mortgage business, with volume in Q2 up 67 percent versus Q1, was the cornerstone of a great diversified revenue quarter. While our online real estate partnerships and corporate relocation business both grew to record levels, most of our purchase business and the greatest growth came from our business direct into e-loan.com.

As we discussed on the call back in January, we began review this year specifically targeting purchase mortgage business with some of our TV advertising and we are definitely pleased with the early results.

Our success in the purchase mortgage business is not just a story of being able to increase demand. In fact, I believe the most crucial aspect of our purchase mortgage performance is our service delivery. To be successful in the purchase mortgage business requires more than just great rates; the customer service delivered much engender the confidence of not only the borrower, but also the realtors and other parties involved. With so much on the line -- a major transaction, multiple interested parties, a family awaiting the move into their dream home -- any hiccup will quickly create major anxiety and a loss of confidence.

Our focus on both great rates and consistent high-quality service delivery, blending both technology and human touch, provides a confidence inspiring experience for all the parties involved in a purchase transaction, and that is what enables us to not just attract purchase mortgage shoppers, but to convert them into funded transactions. We expect at least 15 percent sequential growth in our purchase mortgage loan volume in Q3.

We are also very pleased to report that our home equity business continued its historically strong growth trajectory in Q2, and generated a 41 percent direct margin. As we projected in our last call, our focus on synchronizing our demand and capacity in this business has worked well for us and we expect at least 10 percent sequential growth in Q3.

Please forward to slide 14. In the auto business, we are pleased to report that we have successfully completed the transitions we had planned for Q2. Three major transitions took place over the past few months that provide what we believe is the right foundation for long-term growth in our auto business, even in an environment with extensive with extensive manufacturer financing incentives. First, we completed the relocation of our operation from Jacksonville, Florida to California. Moving the operation out to be with the rest of our business is already enabling us to better leverage the resources and knowledge of the entire company. Second, we brought in a new leader to take over the operation -- Jeff Halverson (ph) brings with him over 30 years of automotive finance and related experience, including time with Ford Motor Credit, Progressive Insurance and most recently, PeopleFirst.com. We are very excited about the experience that Jeff is bringing to our auto business. Third, as we announced on Monday of this week, we completed an expansion and extension of our prime auto funding facility with Merrill Lynch. In addition to extending the agreement out to July of 2005, and increasing the total amount to 800 million, provisions to the facility will enable a more streamlined operational process that will enhance our response time and the online customer experience. Leveraging the new foundation that these three changes have built, we expect auto funding to increase 20 percent sequential in Q3.

Please forward to slide 15. In all our businesses, technology is the foundation off what enables us to offer our customers a better experience at a lower cost. We continue to develop and implement a steady stream of technology-based improvements. Over the July 4th weekend, we completed a comprehensive upgrade of our mortgage and home equity loan origination system that provides us with numerous improvements in our workflow and process that will improve our efficiency and productivity.

Please forward to slide 16. Earlier this week we launched our first version of what we call the debt checkup tool. The debt checkup tool analyzes a customer's current debt and loans in order to provide customized recommendations to reduce their payments, lower their costs or get out of debt faster. Unlike the loan adviser, which recommends the best loan for a given purpose, the debt checkup tool recommends how to restructure a customer's existing debt to lower the overall cost of borrowing. The debt check up tool examines a ranged of money-saving opportunities, including transferring balances between accounts, consolidating debt and refinancing loans. This is the first tool to analyze a customer's debt situation holistically, as well as consider all key variables, like tax savings, credit scores and forecasted interest rates. The customer's information is stored, so that whenever they come back, the recommendations will automatically be updated, taking into account the latest rate and any updates they make to their debt situation. Integration with E-LOAN's credit score monitoring service also reduces the amount of information needed from a customer by pre-filling applicable credit information and automatically updating the customer's credit score every month. As with the loan adviser we have filed for patent protection for this latest addition to our suite of personal financing advisory tools. This supports one of our fundamental purposes of the company and that is to provide consumers with unbiased education and advice about how to best use debt to achieve their objectives.

Our steady stream of technology developments also includes an endless stream of improvements to our website design, the ultimate goal of which is to make it more usable to consumers. A key measure of our progress is the percentage of visitors to our website that move forward and complete an application for one of our products. Slide 17 shows the progress that we have made here since the start of 2001. Setting that period's visitor to application conversion percentage for each of our three main product lines -- mortgage, home equity and auto -- at an index value of 100, the graph shows the growth and conversion for each product line over time. Compared with the start of the base year of 2001, we are now seeing visitor to application conversion that is 42 percent higher in auto, 231 percent higher in mortgage and 436 percent higher in home equity.

There is no one silver bullet behind this progress; rather, it's a result of a continuous disciplined approach in which we understand how our customers are using the site, develop insights that suggest potential improvement, test those improvements and keep those that work. The net effect of all this progress is that it improves the customers' experience while ultimately reducing the amount of money we have to spend to acquire a customer.

As we started the third quarter, we accrued demand across aboard for all of our products, as our heightened marketing aggressiveness continues to work well for us. At the same time we're continued to scale up our operations, with a strong focus on keeping demand growth and capacity growth in sync. As a result, we're very excited about our prospects for Q3.

We would now like to open up the call to any questions that you may have.

(CALLER INSTRUCTIONS)

Rich Repetto, Putnam Lovell.

Nice quarter. First question, on the mix of just the mortgage volume, refi to purchase, I didn't hear that. I might have missed it if you gave that number.

What repurchase is as a percentage of --

Mortgage.

Just of mortgage. Actually, a great quarter there. We went to 34 percent of our mortgage volume being purchased. That's up from 26 percent in the prior quarter.

I'm trying to calculate that out. That equals -- that's about 428 was purchased. Is that correct? Around 430 was purchased mortgage, Matt?

Right. It is about 400 million, right.

What I'm trying to identify is out of that -- there is -- when you classified things as diversified, you said it was also non-prime. This is the way I back -- if you take -- go from 47 percent of the total loan volume, that equals 762, and you back out auto and home equity you come up with about 400. We can go through this off-line. That's no problem.

Great.

The next question is when you go into '04, the MBA, even beyond the fourth quarter, is forecasting some pessimistic, or a slowdown from the great environment you had the last couple of years. So any color -- will it become more difficult, do you believe, to growth purchase product since other lenders will focus more there, given the easy refi volume? Will it dry up? What's the dynamics that you expect next year?

I think what we have seen in this quarter -- something that is very encouraging -- is really a couple of things. I think broadly we have a very constructive purchase environment. That's a very healthy market. But moreover, you are starting to see now the realtors out there which heavily influenced their consumers, not only feeling comfortable about online lending, but in fact now we're seeing sizable numbers of realtors actually encouraging their customers to use online, as we've shown in the study that we commissioned. That's a pretty radical change. I think that is partially what we're seeing is the spectacular growth that we had in purchase. That's always the key thing you need -- not just your purchase customers going online to get information -- which I think is what they started out years ago -- but actually now clearly seeing the advantages -- lower costs, much more open, getting rates that are not based on haggle and negotiation. And on top of that, the Realtors feeling not only comfortable, but actually recommending it. That, we think, is one of the most powerful trends we're seeing and we think it is very constructive.

And that should continue on, even if refi slow dramatically next year?

Yes.

Another question for Matt. On the margins on mortgage loans, whether refi versus purchase, are they about equal? Do you make more right now in the environment for refi versus purchase? Or what's the relative profitability of the two products?

Just on a direct margin basis, refi is definitely -- has more margin at this point because of the fact and that there's more price competitiveness, there's less pricing pressure on the refi, given the demand and the capacity imbalance that we're currently in. But that should equal out over time as the environment shifts away from a refi market.

I guess, the last question to wrap up with Joe, on the technology side, any -- you talked about a lot of the different things you're doing and what you're seeing. Any more progress in regards to a paperless mortgage or anything to help you have the model be more scalable on the fulfillment side?

The changes to the loan origination system on the mortgage and home equity side enable a number of improvements that will make the process more efficient, enable us to be more productive. Included in that is setting up a foundation to reduce the paper and the process as well, but it certainly -- the improvements that the changes set up are far beyond that.

When you say the diversified product revenue is comprised of purchase and non-prime mortgage, in non-prime mortgage is there refinance non-prime mortgage included in diversified?

In diversified revenue, non-prime mortgage is totally included in diversified because the characteristics of that market is such that the sensitivity to interest rate movements does not produce the volatility that you would have in a typical refinance transaction. So we believe that it's more of of a stable book of business, similar to purchase, and therefore have all non-prime in our diversified calculation.

Just a little more color on that -- sub-prime mortgages are almost entirely cash out refinances and really have much more of the character of a home equity loan in terms of the borrower's situation and desire to take cash out of the equity of their home, as opposed to primary refinances which are very heavily refinancing to improve the interest rate.

Where is your cut off in regards to the prime versus sub-prime? What are you using as a FICO score?

Typically, that cut would be around 620. Obviously there are other factors involved, but the credit score cut off would be 620, typically.

Jordan Hinawitz (ph), Global Capital.

First of all, it seems like you guys only have a 10 percent tax rate in the quarter -- 11 percent. Is that what's assumed in the estimates for the 5 cents in the fourth quarter as well?

That's correct. The reason why is we have -- basically should be in almost just an A&T (ph) tax situation, but California suspended the use of any NOL tax carryforwards in the current year. So that's really what's driving any tax rate at this point.

Second, it seems like had a big switch in your auto portfolio from an increasingly loss to an increasing gain in the second half of the year. How comfortable are you with that? And if it would just be breakeven in the second half of the year, and you would assume a full the tax rate, would the earnings be closer to 2 to 3 cents in the fourth quarter, or is that unreasonable?

Yes. A couple things to note about the first two quarters of the year, one of which is that we had incurred some transition expenses that we talked about on the call to move the facility from Jacksonville, Florida and consolidate it here and Dublin. So that was about 1 million 3 in expenses that were occurred that are not recurring expenses. So you would back that out of the first couple quarters to get more of a normalized rate. The other is, as Joe mentioned, we have done a number of pretty foundational things to the auto business to set us up for success over the next couple of quarters, and based on the confidence in those we're comfortable with giving the guidance that we're showing here.

So you think your auto business, ex the onetime costs, is close to breakeven now?

We actually believe that it -- correct, and we actually believe that the growth is going to produce the direct margin that we're showing here.

So just I can get the numbers right, I'm trying to think of the core earnings, which are increasingly good from your core mortgage business, that on a fully taxed rate, if auto is breakeven where it is now in the fourth quarter it would be probably 2 or 3 cents?

Auto alone in direct margin for the fourth quarter is about 800,000, is forecasted for direct margin.

If you were too assume, let's say an 8 to 10 percent run rate in the fourth quarter, what type of growth for next year you think you can get off on a fully taxed basis on just the mortgage business?

We haven't delivered any 2004 guidance at this point. We basically have done that the last two years in our October conference call, and will do so again this year.

A couple things to note, which we have talked to, is that our diversified revenue, which is over $85 million anticipated for the current year, we have said that we're targeting 50 percent growth in that business. This year we actually almost will double our diversified revenue, was about an 98 percent improvement. But if you're to just do to 50 percent improvement in 2004, you get to approximately 130 million in revenue, without considering any refinance volume. So that just gives you a sense of what the growth potential is just for diversified revenue. And of course, we anticipate to get our fair share, if not grow our share -- we actually are targeting to grow our share -- of refinance, which will be added to that.

Are you using the 2 billion MBA number for next year in your estimates?

We have not provided the 2004 estimates yet. Relative to the purchase mortgage and diversified in general, we are using both the -- MBA as a basis for comfort with our 50 percent target, but more importantly, just our own internal metrics that we're seeing around the business and the confidence we have in our marketing to drive the business. And those are the elements that Joe talked about earlier in the conference call.

Pierre Cider (ph), Mensa Capital (ph).

Very nice quarter. I'm curious what you're doing on the expense side as you go into the fourth quarter and then the first quarter of next year, specifically as it pertains to the refi business and how much flexibility you have in constraining those expenses versus other expenses you might have.

We will look to continue the expense side at or below the cost per loan levels that we've been running. That is our ongoing focus there. The bulk of what you see in terms of Q4 in terms of diminished margin on the mortgage side is more a margin compression revenue issue. We expect to continue our very strong focus on the cost side.

What is the cost per loan levels that your are targeting for the refi business?

We have an overall mortgage cost per loan that we ran last quarter of just under -- about 1785 is what it comes out to be last quarter, which is down from 1868 in the prior quarter. The guidance that we have over the remaining two quarters is actually to come down to around just a little over $1,700 cost of loan for the mortgage.

Last question, again on this refi, id -- do you ever do any sensitivity analysis to what half a point would mean, or a quarter of a point means, in the long-range to your business?

As every mortgage lender does, we run financial scenarios that look at what different rates will mean to the business, and what cost structures mean -- are needed to support that level of demand, and we will continue to do so.

You want to share any of that with us?

Yes. We will not go into sharing sensitivity analysis on the different forecasts, but basically we have variable cost structure in place that allows us to tune the business to the demand that presents itself to us.

Chris Palermo (ph), Vio Stone Group Management (ph).

I just have one simple question for you. Going forward over the next couple of quarters, do you believe your company as a whole is going to grow faster than you expected previously?

I don't know if I understand -- I don't know if I understand the question. The guidance that we just provided shows that the diversified products will grow anywhere from 10 percent to 20 percent, depending on the product category. The only part of the business that will not grow just in the fourth quarter is mortgage refinance, for the reasons that we talked about.

Larry Jacobs (ph), Jacobs Asset Management.

Can you add a little color to the discussion you made about margins going from current rate levels to more historical levels over the next couple of quarters? Can you put some numbers on that? What is historical -- the level that you saw in 2002 or 2001 -- or just quantify all that? Then I have a follow-up.

So basically the revenue per loan is what we're talking about when we talk about margin compression due to the refinance environment. We were -- last quarter, almost $5300 in terms of non-interest gain on sale per loan per mortgage, which was a historical high for us as the Company. We have been running in the high fours, low fives over the last three-quarters. To give you some context, what we are -- what we ran in 2002 in the second quarter was $3300 rev per loan. We have -- in our current guidance we have $4600 rev loan in the current quarter and a further decline to $3900 in the fourth quarter. So that gives you a sense for what we've at least assumed in the guidance and a little bit of the history of where we've been.

Is the revenue per loan, that includes origination fees and the gain on sale of loan (multiple speakers) brokered loans?

Right, we actually don't broker any loans anymore. We actually fund all the loans ourselves and sell the loans into the capital markets.

I would also ask you to give me your thoughts on whether it's reasonable to assume higher gain on sale of loan and higher revenue per loan in the second half of this year than, say, last year in what is looking to be, now that rates have gone up 110 basis points, a post-refi boom environment, which is in effect what you're calling for, when every other post-refi boom cycle we have seen in the last 20, 30 years you see not average margin, you see much worse than average margins, as price competition heats up after capacity has increased so much in what is essentially a commodity business. How are margins going to hold up so well?

We actually -- I actually think that the assumptions that we have in here are appropriate, given the fact we're coming off of a $5300 rev per loan in mortgage and taking that down to 4600 and taking a further cut on that down to 3900. So over a four or five-month period here I think that's a pretty significant decline in revenue per loan. And given the fact that we've been running at kind of the 5300, 4800, 5200 level, we're certainly making the attempt to put into the guidance what we believe is a potential for margin compression.

The other factor that is important to understand in terms of our rev per loan performance over our history, while the supply demand situation and its impact on industry margins is important to us, it's also important to understand that as we have grown in volume and grown in sophistication, we make certain changes that enhance our rev per loan execution.

For example, a year ago all of our ARM production was sold on a best efforts flow basis one loan at a time. We now sell all our ARM production using a bulk auction approach and have substantially higher volumes as well. That combination of bulk auction approach and higher volumes translates into a good jump in execution for us. So unlike a very large mortgage company like Countrywide or something where they've already kind of optimized rev per loan and execution as best as they possibly can, we continue to improve the fundamentals of our capital market strategy and take vantage of increased loan volume and superior strategies that give us some offset to the downward pressure that might otherwise exist.

The final part of that question would be, do you hedge out -- sell forward or hedge in such a way when you take an app or make a loans before you sell it in such a way that you did not benefit on your gain on sale loan when rates happened to fall in the warehouse period? Or did you actually -- part of the reason why the revenue per loan went up to 5300 is in effect not by design, but you just had the wind at your back in the interest rate market?

Yes. We do manage our interest rate risk by doing activity, which involve selling forward in essence 30 and 15 year Freddie and Fannie securities. And in a declining -- rapidly declining interest rate environment, you are going to get a positive to your gain on sale execution, as you noted. And part of that $5300 gain on sale is a benefit of that. Our objective when we're doing the hedging is to lock in the economics in essence that we price to make, and we have models in place that with that as their objective, no hedge is exactly perfect, so you're going to have a plus or minus pickup or a loss of execution by virtue of just doing of just doing any kind of hedging.

So your margin would get hurt in a rising rate environment, and that's what you have factored into the drop from 5300 to 4600 to 3900?

When we actually give guidance on rev per loan, we take an approach that tries to take a neutral view of where the hedge will come out, acknowledging that the hedge by definition has some cost to it because it is a transaction that is in essence in insurance, that you're trying to lock in your execution. So this is certainly not counting on any benefit from the hedge activity.

Last question, what was the fixed ARM mix in the second quarter? I don't know if I miss that. And what would you expect that to do now that rates have gone up? And are you as competitive in ARMs as in fixed-rates?

We're very excited about our ARM division. We have seen a significant improvement in the ARM mix over the course of the year from an ARM mix in the low 20s to an ARM mix now that is approaching 40 percent across the business. And that's consumers just doing what's smart, given the steepness of the yield curve. As I mentioned, we made changes last year in how we sell ARMs in the capital market that enables us to be extremely competitive there, as well as to get better execution than we have historically gotten. So we're very comfortable with the shift to ARM, and in fact we're quite proud that we've been able to flex from a fixed to an ARM world remarkably quickly by the standards of typical mortgage company.

We would expect the situation we are in now, with the yield curve as steep -- for example, if you go onto the E-LOAN web site to use loan adviser, more and more ARM and hybrids will be the conclusion -- the recommended conclusion for consumers. In fact, today we will be on several nationwide consumer and financial shows, and that will be some of the advice that we will be giving to our consumers -- it's time to be moving down the yield curve into hybrids and adjustables. That makes sense for consumer, so we would expect those volumes to relatively increase.

Peter Wade (ph), SAC Capital.

A couple of questions. One, can you tell us what percentage of your costs you would consider variable, either temp workers or fees tied to volumes? The second question is can you describe, or give more detail, on exactly what the structure is for the auto loan vehicle, that off-balance sheet thing? I'm not exactly sure how it works. What is the gain on sale accounting treatment that you use when you sell loans into that structure?

The first question, relative to fix and variable cost, most of our operations expense is headcount related; it is roughly two-thirds of our total operations -- non-interest total operations expense. And that includes both some cost that are fixed and variable in terms of temp costs. But we don't have a significant amount of temp costs relative to our total headcount costs. So the answer would be two-thirds of our costs to operations non-interest is headcount related. It varies by product line, but that's the answer to that. And of course, marketing is mostly variable. I would say, 80 to 85 percent of our total marketing spend is direct marketing, non-overhead or personnel related; therefore, that amount of money can be adjusted appropriately to scale up or scale down.

Before you walk to the other thing (ph), does your guidance include a specific scaling down of the marketing budget over the next few quarters, or no?

Yes. In absolute dollars, it's a slight decline but still in line with our percentage of revenue targets. And basically it will be about 26 percent. In the next two quarters our guidance is 25 percent -- or model rather, is more short-term, about 25 percent. So it's meant to acknowledge that it is probably going to be a little tougher to acquire customers in a rapidly declining refinance environment, but the absolute dollars are still at a pretty healthy level in terms of being able to attract and build on the demand for diversified products.

The auto structure -- that is set up to look like -- basically like a normal securitization structure. It has a -- the gain on sale that we recognize there is a discounted cash flow gain on sale. The assumptions are all detailed out in our quarterly filings. We also show what the sensitivity to the revenue would be for changes in those assumptions. But the basic assumptions are a prepaid speed rate of 1.2 ABS prepaid speed rate. The cumulative loss rate that we expect, or that we assume, is about 72 basis points. The discount rate that is used is 12 percent. And basically we just model out the expected cash flows from each month production and book it accordingly.

So if the model -- if actual cash flows differ from those, then what the upside goes to -- the upside and/or downside goes to who?

The upside or downside to that would go through our P&L in our gain on sale over time.

Josh Goldberg (ph), Arksley Partners (ph).

First, on the revenue side, just so I understand, basically your diversified revenue, or your non-refi revenue, grew 41 percent sequentially and when you're guiding for second and third -- third or fourth quarter, you are guiding almost to an up of 9, 10 percent third quarter and a flat fourth quarter? Just walk me through the dynamics there, what you're seeing, why that would be so much slower then what it's been over the last couple of quarters.

The main driver there is the third quarter in particular is a revenue per loan compression in the mortgage side. If you look at going from a revenue per loan, $5300, down to $4600 per loan. That's really the majority that's going there. On the downside -- the upside there is continued growth above Q2 levels from our purchase home equity and auto businesses.

Going into the fourth quarter, again, it's a similar story, which is a rev per loan compression again, as I articulated earlier, down to about $39 a loan, an anticipated 25 percent decline in refi, on the downside. On the upside, another 10 percent growth in our purchase business, which is -- that's actually a seasonally down quarter. In fact, the fourth quarter from the MBA perspective is supposed to be down -- purchase business is supposed to be down 17 percent and we're saying we're going to grow our purchase business 10 percent in that quarter. Home equity is another 10 percent growth in the fourth quarter and auto is yet another 20 percent growth in the fourth quarter. So those are the dynamics that produce the revenue guidance.

Just in terms of the refi activity, a lot of your competitors over the last 10 to 15 days reported their quarter and a lot of them said they do not see a very big drop yet in mortgages, specifically on the refi side. Can you give us any commentary what you're seeing so far 15, 20 days in, in terms of what your sense as for the refi activity?

We continue to see good refinance demand. It's down from what it was a month ago. But our experience isn't markedly different from what you've heard. And for perspective, it's important to understand, the MBA came out yesterday with the last weekly index and the refinance index was still over 6000, which indicates strong refinance conditions. The data we've seen out of Citicorp (ph) analysts says 60 percent of mortgages are still refinanceable in the US. So there is still a very healthy refinance activity, and we continue to see that at this point in the quarter.

So your thought about the fourth quarter being significantly worse on refi would only happen if rates would continue to climb higher 5, 6 percent (indiscernible)?

Our fourth quarter is based on what -- the MBA has an estimate out for Q1 of next year that's associated with interest rates about where they are today, and we're using that estimate of refinance volume for our Q4 assumptions at this point.

If rates would stabilize around these levels you would think the refi activity should be consistent with Q3 levels, then?

Eventually, any time there's constant rates, the refinance business at constant rates will eventually diminish. In some sense, everybody who can refinance will eventually refinance. So it's not safe to say that constant rates equals constant refinance business indefinitely. That said, there certainly is a good amount of refinance business still there and a substantial opportunity and we're aggressively going after every bit of it that we can.

One last thing. Your diluted share count was up about 3 million in Q2. Your guidance is that it should be up another 2.5 million in Q3. What's going on that would contribute to a higher share count in Q3?

It's predominantly a function of the number of options that are included in the diluted share count calculation, and with a higher stock price more options become part of that pool.

Do you have any idea what that number was exiting Q2? In other words, why would it be up another 2.5 million into Q3?

It's a weighted average share calc, so Q2 would have had a lower weighted average share calc -- weighted average share price than what we're anticipating in our forecast of weighted average shares for Q3 and Q4.

Sounds like you're bullish on your stock price.

Basically we're just trying to make sure that we have captured the number of shares that are potentially dilutive so that we don't surprise anybody relative to that number.

What was your weighted average price in Q2?

I don't have that in front of me. It's just a simple average of the --

90 days trading?

Right.

Paul Mattison (ph), Castine Partners (ph).

I have a question. I heard the earlier caller's question about the residual for the auto loans that you have on your balance sheet and how it's marked to market each quarter. It's only about 12.5 percent of equity now, but it sounds from the growth that you guys are predicting over the next year in the diversified areas of your business that auto will become a much bigger part of your business. How comfortable are you in terms of growing that residual? What percentage of equity could that get to for next year?

It's difficult to give that percentage without kind of going through an '04 scenario, but we definitely keep an eye on that in terms of its size relative to the overall equity. The flip side of it is, is it is the vehicle that we have to grow the auto business in the way that we're articulating here. We have a lot of confidence it's the best capital market structure to accomplish our objectives there.

Why is that?

Frankly, we tried the other way. We had a secondary marketing structure that was a best efforts flow type of arrangement, correspondent lending arrangement on the prime side. And what we found is every time that there was a positive movement in the two-year, which is really the basis for a lot of the pricing on the auto side, the partner would take that economics and pocket it for themselves and we basically would be left with a choice of offering less than competitive rates to consumers, or making substantially less revenue per loan.

So if you were to try to sell them for cash in the secondary market, you wouldn't get as much -- the economics would not be as good? Almost always that's the case because folks that will pay up cash today will have some assumptions that are never going to be reflective of what you believe your production is really valued at over the long-term, because they're not willing to take any of the assumptions in terms of performance and just take it at face value. They will haircut everything and then they will have their own discount rates that they will apply that will be much greater because it will represent their return on capital.

Like I said, we tried the correspondent flow basis. That was not allowing us to do -- either get the kind of execution that we need for the business, nor was it allowing us the flexibility to provide a really streamlined excellent process for the consumers. And that's why we went to the structure which we are very pleased with and is providing a good framework for us to grow the business. You guys obviously know the business a lot better than I do. I'll take it that you are booking them at the right assumptions, but it seems like historically whenever people have booked assets higher than the secondary market will pay for them, there's historically been a surprise, and unfortunately not a good one.

I understand --

Just an area of concern.

I understand the concern. A lot of that, if you look at the history on that, particularly the auto, much of that has happened in the sub-prime area and the production -- just to emphasize -- the production we're doing under the structure is all prime, if not what I would call super-prime production.

How do you get the best customers when all the captives (ph ) were offering the zero percent financing?

it's important to understand that despite all of the hue and cry about manufactured financing incentive, they still account for only about 45 percent of total new vehicle sales. You have still got 55 percent of the new vehicle market that's not some form of presented (ph) financing. And, there's all the used vehicles. Typically, the number of used vehicles sold through franchise car dealerships is equal to the number of new car vehicle sales. So you've got 55 percent of 16 or so million new car market available and 100 percent, essentially, of the 16 percent used car market available. So there's still plenty of addressable markets in the auto space.

Can I ask one last question on that -- on the auto? Do you guys service the auto loans yourself then, since you retain the equity pieces?

We have a sub-servicer. It's a big established player in auto servicing called SSG (ph). It's actually a subsidiary of J.P. Morgan Chase.

So if someone defaults on a loan, you don't have to actually go repo; you have someone who you have hired to do that for you?

Right. We work very closely with them and we monitor the performance extremely closely. But yes, they are our arms and legs (multiple speakers)

As that grows, that business, would you expect to ever take it in-house or would you always use a third party?

I think we will continue to evaluate that over time.

The only thing I would add on the auto side, there's a very clear precedent for what we're doing. If you look at the securitizations that have been done by PeopleFirst.com, which is now a subsidiary of Capital One, you will see an excellent performance in terms of the portfolios that they have taken into securities. You can look up the performance of those securities. Can also look at the discussion of their performance on the Capital One investor web site presentation. And I think it might give you substantial comfort that this is a very high-quality type of the type of portfolio that we're issuing here.

Unfortunately, they did a lot of their early deals in private trusts and we can't see that data, but my guess is that (multiple speakers)

(multiple speakers) People First auto portfolio is public information.

I will take a look at that. Thanks for the suggestion.

Jordan Haniwitz (ph), Global Capital.

A follow-up question. I didn't realize that the auto users deal and sale (ph). What in the quarter was the net non-cash gain? In other words, the non-cash gain net of the accumulated tax loss (ph) from the older pools in it?

The quarter revenue was 1.7 million from -- was recognized relative to the discounted cash flows for the production in the quarter.

That is net of the cash you actually received in?

Yes. That's new production, so it's basically the assumptions that we have for the production that is going to take place during the quarter. Some of that cash flow comes in; some of the cash flow does not come in, it amortizes out over time.

I'm wondering if your portfolio lender, instead of a whole loan sale lender (indiscernible) you're saying that the revenues for the quarter would be 1 million 7 less? Is that right? Less what you would actually earn in interest income in one quarter?

If we had the assets on the books, we would recognize the revenue as the cash flows come in.

So you would probably -- they are two year average loans, right?

The average duration is a little that, so about 2.5, 2.7.

So approximately then, 1/8th or 1/9th of the 1 million 7 would actually be recognized then if you were a portfolio lender (ph)?

I would have to get back and do the math on it. The cash flows start coming in right away. There's actually a delay between month of origination and it takes a month for two for a level of overcollateralization that we need to establish in the facility to build up, and then cash flow will start to come in.

Peter Wade (ph), SAC Capital.

A quick follow-up. Can you breakout the level of non-prime mortgages and then further the amount that's refied versus purchase in the quarter? Rough percentages would be fine.

I don't have the non-prime in front of me.

The overwhelming majority -- over 90 percent of the non-prime are going to be cash out refinances (multiple speakers) just the nature of sub-prime lending. In essence it's a home equity loan for sub-prime borrowers.

Okay. You don't have the number of what the percentage non-prime is?

I just don't have that number in front of me. It's a relatively small number, but I just don't have in front of me.

Larry Vitelli (ph), KBW.

Just following up on the auto book, are your auto securitizations wrapped with monoline (ph) insurance? And if so, by whom?

No, they're not.

There are no further questions.

We would like to thank everybody for taking the time with us this morning and we look forward to speaking with you all real soon. Take care now.

(CONFERENCE CALL CONCLUDED)