Encore Capital Group Inc (ECPG) 2003 Q4 法說會逐字稿

Good afternoon. My name is Deborah, and I will be your conference facilitator. At this time, I would like to welcome everyone to the Encore Capital Group fourth quarter conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer period. If you would like to ask a question during this time, simply press star then the number 1 on your telephone keypad. If you would like to withdraw your question, press star 2 on your telephone keypad. Thank you. Ms. Conlin, you may begin your conference.

Thank you operator and welcome everyone to Encore Capital Group's fourth quarter conference call. Today's webcast is being accompanied by a slide presentation which is available at www.Encorecapitalgroup.com on the investors page. With us today from management are Carl Gregory, President and Chief Executive Officer; and Barry Barkley, Chief Financial Officer. Management will discuss the fourth quarter and outlook for the company and then open the call up to your questions.

Before we begin today, I would like to note that certain statements in this conference call constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. Such statements involve risks, uncertainties, and other factors which may cause actual results, performance, or achievements of the company and its subsidiaries to be materially different from any future results, performance, or achievements expressed or implied by such forward-looking statements. For a discussion of these factors, we refer you to the company's annual report on Form 10-K as of/and for the year ended December 31st, 2003. Forward-looking statements speak only as of the date the statement was made. The company will not undertake and specifically declines any obligation to publicly release the result of any revisions to any forward-looking statement to reflect events or circumstances after the date of such statement, or to reflect the occurrence of anticipated or unanticipated events; whether as a result of new information, future events, or for any other reason.

In addition it is the company's policy generally not to make any specific projections as to future earnings and the company does not endorse any projections regarding future performance that may be made by third parties. With that I will now turn the call over to Carl Gregory. Carl.

Thank you, Mara. As Mara mentioned, today we're going to use slides and the first of these slides is our comment on forward-looking statements. By any standard, 2003 was an exceptional year for Encore. We had four profitable quarters, consistent growth, and a successful public offering. We ended the year with a very strong balance sheet and good momentum going into 2004. While there certainly are challenges ahead in our industry, I believe Encore is very well equipped to succeed in the year ahead. Earlier today we filed our 10-K which is full of important information. I encourage you to read it thoroughly. In a moment Barry will go through our financials in detail, but I'd like to mention some highlights. For the year, collections, revenues, net income, and cashflow from operations were up 28%, 30%, 34%, and 38% respectively over 2002. You can see this on slide 3. As we look at these numbers more closely we see that for the year we collected $190.5 million, a 28% increase over 2002's $148.8 million. This number does not include the $11.1 million we received from the settlement of litigation during the first quarter of 2003.

Revenues for the year were $117.5 million, a 30% increase over last year's $90.4 million. Net income for the year was $18.4 million, a 34% increase over $13.8 million in 2002. Finally, cashflow from operations for all of 2003 was $33.9 million, 37% greater than last year's $24.7 million. Looking at slide 4, we can see for the full year net income grew from $13.8 million in 2002 to $18.4 million in 2003, a 34% increase. As I mentioned, in the first quarter we settled a lawsuit. We received $11.1 million in cash and recorded a pretax gain of $7.2 million, which resulted in an after-tax gain of $4.4 million, or 21 cents per fully diluted share. If we adjust for this one-time occurrence and two fourth quarter occurrences, which I'll discuss in a moment, net income grew from $3.9 million in 2002 to $14.6 million in 2003; a 273% increase. Earnings per fully diluted share, excluding these one-time occurrences, grew from 24 cents in 2002 to 70 cents in 2003, an increase of 191%.

If we look at the slide number 5, we can see that during the fourth quarter we collected $47.7 million, an 18% increase over last year's exceptional $40.4 million. Revenues grew 14% to $31.5 million from last year's $27.7 million. Net income was $3.8 million in the quarter compared to $10.3 million in the fourth quarter of 2002. We look at slide 6. We can address the question of net income, because net income deserves some additional discussion because there were several unusual or one-time items in the full year of '03 and the fourth quarters of both '03 and '02. For example, in the fourth quarter of 2003, we recorded a one-time, pretax, non-cash expense of $900,000; or in after-tax $500,000, amounting to 3 cents per fully diluted share resulting from the early pay off of senior debt with the proceeds from our public offering. Compare this with the fourth quarter of 2002 when we received a tax benefit of $8.8 million, or 48 cents per fully diluted share from the reinstatement of our net deferred tax account. If we adjust for these one-time occurrences, fourth quarter net income grew from $1.5 million in 2002 to $4.4 million in 2003, a 189% increase.

Earnings per fully diluted shares excluding these one-time occurrences grew from 9 cents in the fourth quarter of 2002 to 19 cents in the fourth quarter of 2003; an increase of 119%. Our balance sheet has never been stronger. As you can see on slide 7, at the end of the year we had stockholders equity of $71.4 million, a 266% increase over last year's $19.5 million. In addition we had $38.6 million of unrestricted cash on hand compared with $800,000 last year. This will be particularly helpful as we explore new financing and business alternatives in the months ahead. Accordingly the following key ratios have improved significantly. As of December 31st, 2003 our debt-to-equity ratio is 0.94. Our equity to total asset ratio is 0.52, and our debt to EBITDA ratio is 0.93. Regarding purchases, we spent $25.9 million to purchase $882.5 million of face during the fourth quarter of 2003, at an average of 2.9 cents. Of this, $1.9 million, or 8%, was spent purchasing alternative paper. To refresh your memory, we spent $19.4 million on purchases during the third quarter, buying $640 million in face amount at an average price of 3.02 cents.

During all of 2003 we purchased $3.3 billion of face for $89.9 million, a 2.73 cents purchase price. Out of the total purchases for the year, 7% were in alternative classes of paper. And 2002 we invested $61.5 million at an average price of 2.2 cents. Now I'd like to ask Barry to go through our financials in detail.

Thank you, Carl. Included in our 2003 investment and portfolio were $6 million in alternative paper purchases that are not subject to the high interest rate requirements attached to our credit card portfolios. This represents growth of $4.1 million or 216% over the prior year's investment of $1.9 million. In addition, as you can see, we purchased $23.2 million, or 38% more credit card portfolios in 2003 than 2002. On slide 10, as we previously mentioned, gross collections grew to $190.5 million, 28% growth or $41.7 million. That was comprised of collections and credit card portfolios of $172.9 million, an increase of 42.9%; or $51.9 million. It was also comprised of collections of alternative paper portfolios of $6.1 million, increase of 77%; or $2.6 million. Offsetting these increases were anticipated decreases of $12.8 million, or approximately 50% in the retained interest portfolio as well as in the servicing portfolio.

On slide number 11, if we look now at the growth by channel, during 2003 we experienced growth in the collections sites of $23.4 million, or 24.7%, over the prior year's collections; which resulted in $118.4 million in gross collections through the new -- excuse me, through the collection sites. Gross collections through the legal channel grew by $12.4 million, or 44.7%, over the prior year, resulting in $40 million in total collections during 2003. Collections through the sales channel amounted to $28.1 million, or growth of $9.5 million, or 51.4% increase over the prior year. The source of 2003 gross collections were 62% through the collection sites, 21% through the legal channel, 15% through the sales channel, and the balance of the 2% from all other sources. On the next slide, you can see that balancing the growth and purchases against the total amount collected for portfolios during 2003, the turn-over in the portfolio remained strong at 1.2 turns during 2003. We believe that this is an important metric and we watch this carefully.

Slide 13, turning our attention now to revenue, you can see that revenues amounted to 61.7% of gross collections during 2003, up 100 basis points from the 60.7 earned in the prior year. Total revenues grew from 2002 to 2003, as previously mentioned, by 30% or $27.1 million to a total of $117.5 million in 2003. This $27.1 million growth was comprised of $34.6 million or 42.7% growth in revenues from investment in portfolios, offset by anticipated declines of 94.7% in the revenues from the retained interest and 56.8% in the servicing portfolio revenues. This means that the investment in portfolios comprised 98.4% of total revenues in 2003, up from 89.6% in 2002. Slide 14 shows that in 2003 expense growth over the prior year amounted to 17.3% or $11.1 million. As a percent of gross collections, total operating expense was 39.4% in 2003, as compared to 43% in 2002. As this slide shows expense growth was appropriate in the volume-related areas and up about the same amount as the rate of inflation in general & administrative expenses.

Salaries and benefits grew by 12% over the prior year while average staff grew by 18.5% in 2003 over the 2003 level -- excuse me, 2002 levels. Other operating expense grew by 43%, or $3.4 million, to $11.3 million. This is comprised chiefly of volume-related expenses such as mail, credit bureau, and so forth. The cost of legal collections grew 43% in proportion to the 45% growth in gross collections through this channel. On slide 15, during 2003, we borrowed a total of $78.2 million and we repaid $85.5 million. Repayment of borrowings under our current portfolio financing arrangements continues at a rapid pace, with substantially all borrowings prior to the fourth quarter of 2002 having been repaid as of December 31st, 2003. The percentage of remaining borrowings to the original amounts drawn, range from 18.7% in the fourth quarter of 2002 to 80.2% in the fourth quarter of 2003. Slide 16, by and large the biggest percentage of our interest expense is the contingent interest. This is interest which we owe to the lender for their share of the residual collections after both the lender and the company have been repaid their investments in the portfolios. The total amount of contingent interest expense accrued in 2003 amounted to $16 million, and in 2002 $13 million.

For 2003 the effective rate of total interest expenses including fees and other loan costs as a percent of the average borrowings outstanding amounted to 49.1%. For 2002, this amounted to 33% of the average borrowings outstanding. On slide 17, as Carl discussed earlier, we had a number of one-time items to make it difficult to understand the total picture. An apples to apples comparison of income before taxes, excluding the one-time items previously mentioned show that 2003 amounted to $23.1 million, 185.5% increase over the comparable amount of $8.1 million for 2002. Net income excluding the one-time items previously mentioned, show that 2003 amounted to $14.6 million or a 273.4% increase over the comparable amount in 2002 of $3.9 million. On slide 18, if we look further at the fourth quarter results, we think that looking at the results as a percentage of gross collections is the most revealing. Revenues as a percent of gross collections were 65.9% in 2003, this compared to 68.4% in the prior year's quarter. Carl will talk to you next about the subject of revenue recognition.

Expense in 2003 amounted to 40.5% of gross collections as compared to 43.5% in the prior year. We believe that the current level of expense to gross collections is at its approximate appropriate level. Interest expense is 12.4% in 2003 versus 10.1% in 2002. As we have previously discussed we anticipate that this will decline after we refinance our portfolio lending arrangement. Thus pretax income excluding one-time items is 12.4% of gross collections in 2003 versus 10.1% in 2002. And net income excluding one-time items is 9.2% versus 3.7% in 2002. With that I'll turn it back to Carl.

Thanks, Barry. Some comments about pricing seem in order. I believe that there's been a general rise in the price of credit card paper. If we look at slide 19, it will help us talk about this. The price increases haven't been uniform across the age spectrum and have ranged from very small increases to over 25%, with the greatest increases generally being in the older paper. Remember also that there are several factors besides market conditions that affect the average purchase price. Two of the most important ones are the mix and age of the assets purchased. As you can see, in 2003, our average price per quarter varied quite a bit quarter to quarter. It was 3.2 cents in the first quarter, 2.2 cents, 3.0 cents, and 2.9 cents in the fourth quarter. The fourth quarter of -- was actually lower than the third quarter-- or the first quarter of 2003 and lower than the fourth quarter of 2002. I believe that at every price point we've been able to make attractive purchases that will yield the results that we anticipate.

We, like all of our competitors, would prefer that prices decline. Until that happens, I'm satisfied that Encore's unique pricing practices and collection tools are more valuable than ever in choosing and working the best portfolios available. According to the federal reserve, the amount of credit outstanding and the delinquencies are continuing to rise; though the supply seems to be growing. However, our visible success and that of some of our competitors have continued to attract more money into the business. The majority of this money is flowing into established funds with solid business models and, I hope, the discipline to stick with it. However, there are some aggressive bidders who may not fully appreciate how difficult this business is, or be motivated solely by growth objectives. If this is true, then in a year or so we may find some of these firms back on the sidelines trying to work their way out of a problem. Of course, they may have a different, better business model and the higher prices work just fine for them.

Looking at the size of our company on slide 20, we ended the year with 716 employees, one more, or essential flat with the end of the third quarter, and a 17% increase from the 611 employees we had at the end of last year. Our average number of employees for the quarter and year respectively were 712 and 679. As you know, we believe monthly collections per average employee is an important metric that we watch closely. As you can see on slide 20, our monthly collections per average employee were $22,327 for the quarter, and $23,385 for the year. For 2002 monthly collections were $21,656. You can see the progress of this important statistic since 2001. As Barry indicated, I'd like to spend a few minutes talking about revenue recognition.

As you know revenue recognition in our industry is based on a combination of actual collections to date plus estimates of future collections. Every time we buy a portfolio we forecast the amount we expect to collect over a specified period of time and use that forecast to determine the reportable revenues during any reporting period. As you can see on slide 22, if we spend $100 to buy a portfolio we think will yield $270 the revenue recognition percentage is 63%, ignoring the time value of money. So for every dollar we collect, 63 cents is reported as revenue and, in essence, 37 cents is used to reduce the portfolio's basis. As stated in our last call, we believe it is prudent to maintain a conservative revenue recognition policy to allow for the normal fluctuations in portfolio performance. As I said, we generally expect to collect about 2.7 times our purchase price on average over 54 months. As you would expect, over a portfolio's life it can do better, worse, or about the same as we originally expected. In the ordinary course of business these variations from the original expectation can result in changes of revenue recognition.

If there's no fluctuation, the original projections are maintained. Up to this point, when a portfolio has performed better or worse than our forecast, we have taken two different approaches reflecting our conservative nature. If we look at slide 23, refresh your memory, when a portfolio's collections fell below our expectations we lowered our forecast for the total forecasted collections; which, in turn, lowered the expected IRR which lowered the percentage of that portfolio's collections we reported as revenue. Using our example if we lowered expectations from $270 to $225, the revenue recognition would become about 56%. This was one option and the result is shown on slide 23. In this case future collections were reduced. We show you the range of possible outcomes if you achieve a 75% ratio of collections to revenue, the multiple of total collections to purchase price would be 4 to 1. And to achieve an 80% revenue to collections ratio, the multiple total collections to purchase price would be 5 to 1. As I said, our business model is built on a conservative 2.7 times our purchase price.

Looking at slide 24 you see that this generates an internal rate of return of 77% given the rate at which we expect to collect it. Our model assumes that we will achieve the collections over a 54-month period, both assumptions are conservative. Looking at slide 25, this slide shows the impact of accelerating collections on the IRR. One could ask which portfolio would you rather have, and I'm sure we would all rather have the portfolio shown in example 2, as it has a higher IRR. Both portfolios have the same collection to cost multiple of 2.7 to 1, but the earlier collections on the second portfolio caused the IRR to rise to 91%. The second thing you should notice is the shorter life on the second portfolio. As a reminder, until the fourth quarter, even though we knew that most of our portfolios were outperforming our forecasts; we were unsure whether this was a result of collecting the anticipated amount faster, or whether we were actually collecting-- increasing the penetration of that portfolio. Accordingly we subtracted the excess collections from the tail of the portfolio, thereby shortening its expected life.

While we do know that we were increasing our collections significantly above the individual forecast of portfolios, we were unsure as to the cause. As you know, the oldest portfolios which this management team has purchased and to which we have subjected our disciplined collection strategy are just three years old this last December. Looking at slide 27, to recap, up to now we've taken the position that we didn't have adequate statistical support to answer the question, are we collecting the expected amount faster or will we ultimately collect more than expected? We knew what we had collected, of course, but we couldn't adequately answer the what's left question, it's more complicated than it sounds. Well, because we lacked a sound basis for changing our expectations, we maintained the original expectations by lowering future collections by the excess amount we had collected to date. This effectively shortened the life of the portfolio below 54 months. In other words, to be conservative, we have to date lowered our total expectations when our collections trailed our forecast; but have not raised our total expectation when our collections exceeded our forecast.

This led to us embark on the development of the UCS model last spring. We started parallelling this model in July of 2003, and after six month of tests we implemented this model effective October 1st, 2003, with approximately 90 portfolios which comprise 92% of the book value as of October 1st, 2003. The result was that we extended the life by a simple average of 13 months. This resulted in an average life of these portfolios of 44 months. Based on the work of our decision science group over the past nine months, we now feel comfortable recognizing the possibility that some of our portfolios will outperform our original expectations. This means that from now on whenever our actual collections exceed expectations, we have the necessary forecasting model to answer the more or sooner question. In some cases the new model will tell us that we have simply collected the expected amount sooner and there's no basis to increase the amount we expect from that particular portfolio. In other cases, however, we will discover that the portfolio is more valuable than we expected, because we will ultimately collect more than we originally forecast.

This will cause to us increase the portfolio's IRR resulting in a higher revenue recognition percentage. It's also possible we could be collecting both more and faster. In other words, based on a portfolio's past performance, and what we know about it at that particular point in time, we can now estimate the amount of that portfolio's remaining value. Going forward, when we make our quarterly portfolio adjustments, we will be increasing some portfolios total expectation as well as leaving some the same and decreasing others. For instance, if after owning the portfolio for some period we determine that we will actually collect $400 rather than the $270, the revenue recognition percentage over the life of the portfolio becomes 75%. On slide 28, we can see that this has added aggregate collections to the tail of our existing forecast through September of 2008. As can also be seen, the forecast over the next 18 to 20 months, until about September 2005 looks relatively the same.

In the future when we conduct our quarterly reviews of portfolio performance using our new model, we will increase the future expected collections for those portfolios whose performance to date justifies it as well as lower the future expected collections for those portfolios who's experience dictates that approach. Based on our current results and expectations, I believe our revenue recognition will increase modestly and settle in around the mid-60% area. As we gain more experience with the UCS model, and the UCS model has more experiential data to work with, our revenue recognition percentage could move outside of this range. On slide 29, we show you that our results to date indicate that we are beating our model. It shows, for instance, for all the portfolios bought in year 2001 we have collected to date, through the end of the year 2.8 times their purchase price. Likewise for those bought in 2002, we've collected 1.9 times our purchase price. It's these results that drove to us develop the model to help us answer the question, is there more left in the portfolio or have we simply collected it faster.

The revenue recognition or accretion percentage for the fourth quarter was 65.9%. Every quarter we will remodel our portfolio to reflect the most recent data and collection history. Therefore, although the percentage may change every quarter we are comfortable, based on our current results, that it will be around the 65.9% for the year. If you were to look -- if you were to assume that the ratio of accretion to collections for all of 2004 looks like the fourth quarter of 2003 or 65.9%, the resulting increase in revenues would be approximately $8 million more than if we were not making an upward adjustment, as well as a downward adjustment. I hope that answers some of the questions about our revenue recognition policies.

As I look out at 2004 and beyond, I see three drivers of growth for Encore. First our core business will continue to grow as we innovate more and discover additional ways to profitably use our analytical skills. Second, we will replace our existing portfolio of purchase financing with less expensive financing. In the fourth quarter the contingent portion of our existing financing cost the company 13 cents per fully diluted share on an after-tax basis. For the full year it was 46 cents per share. Finally we have a large amount of cash, $38.6 million at year end that will allow us to purchase more non-credit card assets in the months ahead. Because these assets are not encumbered by our credit card financing, the results may be more profitable for the company. We're very pleased with 2003's results. We accomplished a lot. The year 2004 will have its own challenges and opportunities, some of which I've mentioned. We're confident that Encore has the people, resources, and commitment to succeed. At this time I'd be happy to entertain your questions.

At this time I would like to remind everyone in order to ask a question, please press star then the number 1 on your telephone keypad. We will pause for just a moment to compile the Q and A roster. Your first question comes from Audrey Snell with Breen Murray.

I can't hear her question, I'm sorry.

Your line is open.

Could you give us a little insight into the type of paper purchased in the quarter?

Yeah, we purchased-- 92% of the paper that we purchased was credit card paper, and 8% was non-credit card paper; and the credit card purchases were generally spread across the age spectrum with perhaps a bit more in the early segment than typical.

Carl, you mentioned that there is some upward pressure on pricing although it is not uniform. Going forward, do you see any change immediately in any of that pricing structure? Are you seeing any change in the market in the new year?

You know, Audrey, the market is so fragmented that it's really hard for to us generalize. I would expect that the market will continue to be disturbed here for the next six months as we deal with the same kind of issues of more money coming in and increasing the demand, but I'm pretty sanguine about the market long-term because I think the total unsecured consumer debt universe is very large; and as more and more issuers are -- or financiers come around to selling their paper, the supply will grow substantially.

Okay. Thank you.

Your next question comes from Rick Shane with Jefferies & Company.

Good afternoon, guys.

Hi, Rick.

A couple sort of high-level questions, and then a couple of details just to help us fill in our model. First of all can you talk a little bit about what you're seeing for the collections environment currently and what your expectations are related to tax refunds this year?

I would answer in a general way that our collections have been excellent so far this year and the tax refunds can't do anything but help.

Are you starting to see those come through already or is that something will start -- that will really kick in, in March and in April?

I think that we may have seen some of that already, and I'm -- but I'm speculating because our collections have been very strong so far this year; and our type of customer tends to get their tax refunds earlier rather than later.

Okay. When you look at the sort of collections expectations on the non-card portfolios, we've never talked in the past about how expectations could be different for that asset class.

Yes.

What are you seeing there, and can you start to break that out a little bit?

Well, let me start by saying that because we do all our analytics at the account level, we're sort of agnostic about the origination of the account. We're trying to ask the question and answer the question, can the debtor repay the money and if so when? And consequently, it's not of particular importance to us whether that's a Citibank original -- originated credit card or an X Y Z auto deficiency; because we're, again, looking through the portfolio to the individual accounts to make a determination and then scoring accordingly and then driving accounts through our processes based on the individual account scores. So we do not take all of the auto deficiencies and treat them one way and all of the credit cards that are primaries and treat them another way. We deal with the individual accounts based on the individual account score.

Got it. Okay. That's helpful. Then just a couple of detail questions for Barry. Barry, what was the total weighted average cost of debt for the quarter? It's a little bit hard to break out because of the redemption, and I just want to make sure we have that right.

You want the quarter or the year?

The quarter, if I could, please.

Why don't we go on, and I'll look at that.

Great. Then the other question I had is just related to the tax rate. I assume that that has to do with the write-off, but could you just help us work through the slightly lower tax rate for the quarter?

Yeah. The part of the history of the taxes are back in 2000 when we first impaired the portfolios, we wrote off our deferred tax inventory. At the end of 2002, we reinstated the same deferred tax inventory. And as we got down to the end of 2003, we did a very careful analysis because virtually all of our significant items are done on a different basis for tax than book. As a result of that, we found that we had some adjustments that we could take which lowered our effective tax rate from about 39.3% down to the present tax rate. So it is a one-time sort of occurrence that should not repeat itself. We did keep some reserves because of the California NOL and a couple of other issues. We kept what we believe are adequate reserves. But all in all I would look for a more normal tax rate, normal in our case being about 39.3% all in tax rate.

Guys, thank you. And I certainly appreciate all the detail on the slides. That's very helpful.

Your next question comes from Michael Chapman with Lee Munder Capital.

Hello, gentlemen. Couple questions relating to your comments on revenue recognition for your current portfolio, set of portfolios. You mentioned 30 portfolios make up 92% of the book, right?

Ninety portfolios.

Alright, 90. And do you expect any significant portfolios from the earlier vintages to fall into fully amortized status in the calendar year '04?

Yes, I do, because we've been collecting so rapidly that our amortization has been quite high.

How many roughly out of the 90?

I don't know that I would predict that, but I would point out that of our $117 million in revenue for 2003, $20 million of that revenue came from a 0 basis income, which is the source of what you're asking about.

And that's about a four-fold increase over the prior year.

Alright, and so would some of that '03 recognition be on 2001 vintage portfolios?

Yes.

Alright. Would you expect a similar type of scaler to the growth rate in that fully amortized collection stream in '04?

I think has we'll continue to achieve a high amortization rate, but a logical outcome of our now ability to increase the revenue recognition is that the amortization rate will go down correspondingly. As we do a better job, if you will, of predicting the future and matching revenues with our collections.

Right. But that applies only to portfolios today that have book value left, right, or unamortized capital left.

That's correct.

And how many portfolios are in that status?

We don't split that out by number of portfolios.

Alright. And then if you look at the 2001 vintage of portfolios you acquired, given that you're going to be recognizing forward recognition rates by portfolio, could you give us some sense as to the proportion of the 2001 vintage that is collecting ahead of plan today; you know, whether it's, you know, just a few stars, or 80% of the portfolios that are collecting ahead of plan?

I wouldn't want to be real specific, but I will say that we are generally collecting ahead of plan.

And that goes across both the 2001 and 2002 vintages?

Yes. And let me be clear there are obviously some portfolios that don't collect ago well as we expected, but in those instances, they're simply collecting at less than 2.7. They might be checking at 2.4 times our purchase price, and so it's just a lower yield; and I draw that distinction because of what Barry said a moment ago. When we took over the company in 2002, we took a big impairment charge for portfolios that had been purchased prior to that time. And that's a completely different can of worms. All we're saying is that in some cases our portfolios don't perform as well as we originally expected. So we reduce the IRR accordingly.

And last question, in your negotiations to find new credit facilities in '05, do you see any room for compromise with the current lender so that he can retain your business by granting you concessions earlier than January 1st, '05?

I think the best thing to do is not comment specifically about any discussions with potential lenders, and simply say that we are actively engaged in the business of discussing our future financing needs with potential lenders and that when we come to some conclusion we will make an announcement that describes what it is we have done. And obviously we are striving to make the best conclusion for the company, taking into effect all of the things that go into a financing conclusion, including the remainder of the term of the existing lender's agreement.

Yep. Appreciate it. Thank you.

Sure, Michael.

Again, I would like to remind everyone if you would like to ask a question, please press star then the number 1 on your telephone keypad. Your next question comes from Steve Delaney with Ryan Back & Company.

Good afternoon.

Hi, Steve.

Couple of questions about the legal channel, if we can. I know that both the cash collections and your expenses pretty much tracked, you know, 43, 44% in both cases. I want to be clear. The actual expenses that -- in the legal channel, you're recording all of the expenses as they're actually incurred, correct?

Yes.

As opposed to matching with revenue, nothing is -- you spend the money on the legal system, and the dollars come down the road.

They're actually sort of two classes of costs. The first is the cost of filing suit that we advance to the attorneys, and we capitalize those. Virtually all of those are fully reserved, so, in essence, they're for the most part booked up-front. And then as the collections come in the door, we match the costs against the collections, so that the fee paid the attorneys is matched on a dollar for dollar basis or percentage of dollar to dollar basis.

Okay.

As we're paid.

But, Barry that will generally track pretty well, then, on a year-to-year basis?

Yeah, we've had some success in continuing to leverage that. We think we're fairly close to the maximum level of leverage we're going to be able to get in that, but, yes, generally from here on out it will track fairly consistently.

We shouldn't expect that there's like a specific tail on that such as that, you know, you've incurred all the expense you're going to incur but then there's revenue in the next -- the following quarter, the following year?

I would say that there's a tail on it but the tail has cost associated with it.

I see. The costs are just going to match across the board.

Right.

Can you estimate what percentage of total accounts are currently in the legal channel? Can we have a handle on that?

I don't know the percentage of total accounts. I will say that we use this channel a lot, and we may put as many as 20,000 new accounts down this channel every month.

Twenty thousand accounts?

Yeah.

Okay. It strikes me that, you know, that arguably, you know, while it's complicated and you have to properly select the account; that this potentially is the most profitable channel that you have if you were to do a segment profitability there. Do you have a feel -- have enough history to look -- you use an average in your model of 2.7 times purchase price for, on average, over 54 months. Is there a multiple that you would use on the legal channel alone?

We actually have such information for every channel but wouldn't want to discuss it just for competitive reasons.

Okay, could you say whether I'm accurate in the assumption that those accounts that are suitable for legal, that they are -- the cash collected is greater there than they are on average?

Well, I'll say that for the accounts that we score, and based on the score, we put down the legal channel, we believe that's the most productive channel for those accounts.

Okay. That's -- thank you, Carl.

Sure.

Operator?

At this time there are no further questions. I will now turn the call back over to management for any closing remarks.

Thank you, and I appreciate everybody paying attention during our fourth quarter call and we'll talk to you in about two months.

Thank you for participating in today's Encore Capital Group fourth quarter conference call. You may now disconnect.