Countrywide Financial Q2 2007 Earnings Call Transcript

TRANSCRIPT SPONSOR

Countrywide Financial Corporation (CFC)
Q2 2007 Earnings Call
July, 24 2007 12:00 pm ET

Executives

Angelo Mozilo - Chairman & CEO
John McMurray - Senior Managing Director & Chief Risk Officer
Kevin Bartlett - Executive Managing Director & Chief Investment Officer
David Sambol - President & COO
Eric Sieracki - Executive Managing Director & CFO
Carlos Garcia - Executive Managing Director Banking & Insurance

Analysts

Paul Miller - Friedman, Billings, Ramsey
Ken Posner - Morgan Stanley
Eric Wasserstrom - UBS
Chris Brendler - Stifel Nicolaus
Moshe Orenbuch - Credit Suisse
Fred Cannon - Keefe, Bruyette & Woods
Brad Ball - Citigroup
Bob Napoli - Piper Jaffray
Ken Bruce - Merrill Lynch
Samuel Crawford - Citigroup
Gary Gordon - PaineWebber
Howard Shapiro - Keefe, Bruyette & Woods
Jim Fowler - JMP Securities
Michael Harkins - Analyst
Jim Delice - Analyst
Larry Rittel - Analyst
Satish Pull - Analyst
Ronald Redfield - Redfield, Blonsky & Co
Christina Amon - Analyst
Tom Atteberry - Analyst

Presentation

Operator

Ladies and gentlemen, thank you for standing by. Welcome to the Countrywide Financial Corporation's second-quarter 2007 earnings conference call. During this teleconference, Countrywide's management may make forward-looking statements within the meaning of the federal securities laws regarding their beliefs, estimates, projections, and assumptions with respect to, among other projections and assumptions, among other things the Company's future operations, business plans, and strategies, as well as industry and market conditions, all of which are subject to change. Actual results and operations for any further period may vary materially from any past results discussed during this teleconference.

Factors which could cause actual results to differ materially from historical results or those anticipated include, but not are limited to, those items described in the second-quarter press release and detailed in documents filed by the Company with Securities and Exchange Commission from time to time. The Company undertakes no obligation to publicly update or revise any forward-looking statements.

At this time, all phone lines are muted or in a listen-only mode. However, after today's presentation, we will be taking questions; and we certainly encourage your participation at that time (Operator Instructions).

As a reminder, today's call is being recorded for replay purposes. We ask that you stay on the line at the conclusion of today's meeting to receive the replay information.

With that being said, here now is your host, Countrywide's Chairman and Chief Executive Officer, Mr. Angelo Mozilo. Please go ahead, sir.

Angelo Mozilo

Thank you. Good morning and welcome to Countrywide's earnings teleconference call for the second quarter of 2007. Joining me on the call today are David Sambol, our President and Chief Operating Officer; Eric Sieracki, our Chief Financial Officer, Kevin Bartlett, our Chief Investment Officer; Carlos Garcia, our Chief of Banking and Insurance; Ron Kripalani, President of Countrywide Capital Markets; Bob James, President of Balboa Life & Casualty; John McMurray, our Chief Risk Officer; and Ann McCallion, our Chief Financial Operations and Planning.

If you have not done so already, please download the second-quarter supplemental presentation from our website. John McMurray has prepared a detailed slide deck to walk you through the credit trends and Countrywide's credit costs for this quarter.

Since most of you already have seen this morning's press release, I will not spend time going through all of the information contained within it. Instead, for those who may be traveling or hadn't had a chance to read it, I will briefly summarize the earnings results, so that we allow plenty of time for Q&A at the end.

So to quickly recapped, diluted earnings per share were $0.81 for the second quarter of 2007, which compares to $0.72 for the first quarter of 2007 and $1.15 for the second quarter of 2006. Countrywide's results for the second quarter of 2007 reflect the strength in our core Loan Production business, but were adversely impacted by continued weakness in the housing market.

During the quarter, softening home prices continued to affect many areas of the country and delinquencies and defaults continued to rise across all mortgage product categories as a result. Due to these adverse conditions, we incurred increased credit-related costs in the quarter, primarily related to our investment in prime home equity loans.

Credit-related costs in the second quarter included impairment charges of $417 million taken during the quarter on the Company's investment in credit-sensitive retained interests. This included $388 million of impairment on residual securities collateralized by prime home equity loans, which equates to approximately $0.40 in earnings per diluted share based on a normalized tax rate.

In addition, the provision for losses on HFI, the loans that occurred in the second quarter, was $293 million, driven primarily by a loan-loss provision of $181 million on prime home equity HFI loans in the banking segment. Those are loads held-for-investment.

Partly offsetting increased credit costs, our residential Loan Production sector delivered strong results during the quarter. Consolidated quarterly funding volume was the third-highest in our history. Prime production margins were relatively stable, and subprime production margins substantially improved. As a result, Loan Production sector pretax profit for the quarter was $439 million, its highest level since the first quarter of 2005.

Looking to the second half of 2007, we expect difficult housing and mortgage market conditions to persist. Nonetheless, we remain optimistic that the long-term future growth prospects and profitability of the Company as the industry continues to consolidate.

We updated our guidance for 2007 and expect diluted EPS to range from $2.70 to $3.30. This compares to our previous guidance $3.50 to $4.30. The reason for the reduced earnings per share estimate relates to our anticipation that the second half of 2007 will be increasingly challenging for the industry and for Countrywide. Absent a reduction in mortgage interest rates, production volumes are expected to fall and competitive pricing pressures are expected to increase. In addition, volatility in the secondary market has increased significantly early in the third quarter; and liquidity for mortgage securities has been reduced as a result. These conditions are expected to adversely impact secondary market execution and further pressure gain on sale margins.

Furthermore, additional deterioration in the housing market, specifically in home prices, may further impact credit costs. We have taken or are continuing to take a number of actions in response to changing market conditions.

These include tightening of credit guidelines, particularly related to subprime and prime home equity loans; further curtailment of subprime product offerings, including the recent elimination of certain adjustable-rate products; risk-based pricing adjustments; use of mortgage insurance for credit enhancement; and expense reduction initiatives.

Notwithstanding current environment factors and their near-term impact on earnings, we believe that the Company is well positioned to capitalize on opportunities during this transitional period in the mortgage business, which we believe will enhance the Company's long-term earnings growth prospects. We expect to leverage the strength of Countrywide's capital liquidity positions, superior business model, and best in class workforce to emerge in a superior competitive position coming out of the current housing downcycle.

Furthermore, the ongoing integration of our bank and mortgage company is expected to enhance our business model through operational efficiencies, reduce funding costs, and enhance liquidity, among other competitive advantages.

That concludes my prepared remarks. I will now turn the call over to John McMurray, our Chief Risk Officer, who will go through the presentation with you.

John McMurray

Thanks, Angelo. With the supplement this morning, we are going to cover briefly several topics including a look at home prices, a look at how home prices affect loan performance, a look at delinquency trends, a look at some of our key credit exposures, and then we will conclude with a summary of actions we have either taken or have underway to address the transitioning environment.

Let's start on page 1 with an update to a chart we have used previously. This chart was put together using the MRAC repeat sales home price indices. The red line running horizontally through the middle of the chart represents what is happening to home prices nationally. It is measured on each month by looking back exactly one year to calculate a rate of appreciation. For May, which is on the far right-hand side of this chart, national home prices actually declined by 0.4% year-over-year. This is the first national decline for the entire period shown on this chart.

The light blue bars each month represent a range of year-over-year appreciation by MSA. MSA is a metropolitan statistical area, which the Census Bureau defines as a city or town with a population of 50,000 or higher. There are currently 381 MSAs in the US. The tip of each blue represents an average of the five fastest appreciating MSAs. The bottom of each blue bar represents an average of the five slowest appreciating MSAs.

For most months shown on this chart the slowest appreciating MSAs are actually experiencing home price depreciation. In other words, home prices are going down. The darker blue portions of each bar represent plus or minus one standard deviation from the red line. So approximately two-thirds of the MSAs will be represented by the darker portion of each blue bar.

Most of the MSAs that make up the very bottom of the blue bars in recent months are from Central California around Stockton and Modesto. According to Moody's economy.com, these MSAs also have the highest mortgage delinquencies as well. Ruth Simon of the Wall Street Journal did an article last Thursday that you may want to look at. It addresses delinquencies and highlights this area; it is on the top right of page D3.

Another useful source of home price information is S&P/Case-Schiller. 14 of the major 20 MSAs that they track are showing negatives as of April. The orange area at the bottom of the chart shows the percentage of MSAs with negative year-over-year appreciation. It uses the right-hand vertical scale.

Pay close attention to the time period beginning around 1998 and 1999 and continuing into 2006. There were essentially no MSAs experiencing price declines. This recent period of appreciation was very broad geographically, and it lasted a long time. As of May 2007, the percentage of MSAs with year-over-year price declines is up to 50%. The transition from practically no MSAs with declines to 50% of the MSAs with declines has been rather abrupt.

Let's turn to page 2. On page 2, you see an uncluttered close-up of the red national home price line from page 1. This time series is a housing unit weighted average of all 381 MSAs. You can see the steepness of the current decline more prominently in this chart than the previous chart. So far, the home price appreciation rates or the decline in home price appreciation rates now underway basically matches the size of the drop we saw in the late '70s and early '80s. So that decline ended before the appreciation rate hit zero. And the current decline isn't over.

Let's turn to page 3. The graphs on page 3 show delinquency over a range of cumulative home price appreciation rates. For each graph, the x or horizontal axis represents the amount of cumulative home price appreciation for the first 24 months of the loan's life. The y or vertical axis for each of these charts represents the serious delinquency rate for each bucket or range of cumulative home price appreciation on the x-axis. The serious delinquency measure used here is an over-90-day delinquency using the MBA standard and expressed as a percentage of the starting count of loans.

We show three first lien types along with home equity. Home equity consists of HELOCs and fixed-rate closed-end seconds. The top two graphs are first lien conventional loans and include both conforming and nonconforming conventionals. The chart at the lower left shows home equity; and the chart at the lower right shows subprime first lien. Across these four loan types we observe a similar pattern, where high serious delinquency rates are negatively correlated with home price appreciation. The higher the appreciation rate, the lower the serious delinquency.

Another important pattern is CLTV, which is the combined loan-to-value ratio at origination, giving effect to both the first and any second lien that may exist. Notice that higher CLTV loans, the ones with more leverage, tend to have high serious delinquency rates across the spectrum of home price appreciation rates. You can see this pattern for all loan types shown here, though the effect is somewhat less pronounced in subprime.

Let's turn now to page 4. Here on page 4 we provide a refreshed look at some of the odds ratios shared with you in our previous serious delinquency presentations. We are showing these charts again for several reasons. First, we wanted to provide an update. Second, we now have data that reflects recent home price declines. Third, the type of variables used to put together these studies are among the types we use in our automated underwriting scorecards as well as our pricing models.

These graphs were generated using empirical models and Countrywide data. The models control for numerous loan and borrower attributes, as well as environmental variables such as seasonality and home prices. The purpose of the study is to understand how each attribute affects loan performance on an isolated basis, which in these cases is measured is an over-90-day delinquency. In all charts, the y or vertical axis represents an odds ratio, which can be thought of as a risk multiplier.

Let's use the FICO chart on the top left of page 4 as an example. We use a FICO of 800 as a base; so we will set that FICO to a value of 1. If we look at the blue line, which represents prime first liens, we can see how the odds ratio increases as the FICO declines. A prime loan with FICOs in the low 500s is going to be over 30 times more likely to be seriously delinquent than a prime loan with an 800 FICO, holding all other variables constant.

On the right side of page 4, we show odds ratios for origination CLTV. For first liens in the upper right, an origination CLTV of less than or equal to 50% is set to the baseline of 1. For second liens in the lower right, an origination CLTV of 51 to 80% is set for the baseline of 1. The pattern here is consistent with the charts on the prior page. Leverage at origination matters. More leverage means more serious delinquencies.

On the bottom of page 4, we show odds ratios for documentation types with full doc being the baseline of 1. Let me explain the various doc types that we are showing here. A streamline is a streamlined refinance. Preferred is a low documentation approach offered by a number of institutions; Countrywide's is called Fast & Easy, where borrowers who meet certain criteria are allowed documentation waivers. A SIVA is a stated income verified asset program; it's sometimes called reduced doc. A NIVA is a no income verified asset program. The primary difference between SIVA and NIVA is that the borrower states but does not document their income for a SIVA loan, but neither states nor documents their income for a NIVA loan. Assets are verified for both SIVA and NIVA.

A SISA loan is a stated income stated asset; the borrower is stating but not documenting their income and assets. A NINA is a no income no asset loan; the borrower is neither stating nor documenting their income or assets. The takeaway from this chart is that documentation matters. The less documentation, the higher the serious delinquency, all else equal.

Let's move to page 5. On page 5, you see four more adds ratio graphs for occupancy, property type, and loan size. I won't spend a lot of time on this page, but let me summarize some observations. Remember that these observations are also on an all else equal basis using the same serious delinquency measure I described previously.

Owner occupied loans are generally less risky than investor properties. Loan or line size matters, with higher loan or line amounts generally being more risky, with the exception of very small first liens.

For property types, manufactured housing exhibits elevated risk. While the other property types seem to be very similar, I think two to four units and condos may be worth watching in the months and years ahead. In past downturns, say the late '80s or early '90s in the Northeast, two to four units properties tended to have worse performance than single-family.

There are a couple reasons I think condos deserve extra attention. First, they appear to be oversupplied in certain markets. Miami and San Diego are two MSAs that come to mind. Second, there have been more conversions of apartments to condos.

Let's turn now to page 6. Page 6 shows the composition of three important portfolios. The first column of numbers on the left is the home equity HFI portfolio in the bank. As a quick reminder, HFI stands for held-for-investment. The second column of numbers shows the portfolio of loans underlying home equity residuals. The column of numbers on the right shows the HFI pay option portfolio in the bank.

Let's take a look at the rows for this table starting from the top. You can see the size of each portfolio both in terms of UPB as well as loan counts. There is a row labeled piggyback that shows the percentage of loans where a combination of a first and second were used in the transaction. In many cases, piggyback transactions were done as a substitute for mortgage insurance. I point this out because it is an important difference between our HELOC business compared to traditional HELOC business.

The next row down shows weighted average CLTV. You will also see a distribution of CLTVs towards the bottom of this page. We also show weighted average origination FICOs. Just below the CLTV distribution at the bottom of the page, we show a distribution of FICOs. Next, we show the percentage of low documentations in the middle of the page. Just below the low doc row are two rows that show margin for these loans that are floating rate. Floating rate loans will include HELOCs, which float over prime, and pay options, which have accrual rates that change monthly based on various indices such as COFI, MTA, and LIBOR.

In addition to showing the weighted average margin over prime for HELOC, we also show an estimate of the rate spread compare to LIBOR. So normally prime is 275 basis points or so above LIBOR; and so we wanted to provide that calculation for you as well. For pay options, the non LIBOR indices are lagging, but generally approximate LIBOR over time, so the margin can be thought of as a margin over LIBOR as well as over the index.

Let's move to page 7. Page 7 is a portfolio composition similar to what we saw on the prior page, except here we are looking at subprime. Starting on the left, the first column of numbers show the portfolio of loans underlying our subprime residuals for the 2006 vintage. We expect the 2006 vintage to be one of the worst performing vintages ever for both us and the industry at large.

The right-hand column of numbers shows the composition of our 2007 Q2 subprime originations. Most of the rows match what we saw on the previous page, with a couple important distinctions. There is only one basic version of low documentation in subprime, and that is called stated. We also show a row that shows the percentage of loans that are both stated doc and have origination CLTVs above 90%. Also notice that the FICO distribution range here has been adjusted for nonprime collateral.

One of the most important points on this page is the difference between 2006 and more recent originations. As a result of guideline adjustments and other actions, the mix of loans has generally improved. The one area where the mix has not improved is FICO. The high CLTV loans that were previously done required higher FICOs; so part of the trade-off in eliminating the higher CLTV programs was that the FICO mix deteriorated in that process.

Let's turn to page 8. Page 8 shows total delinquency for our servicing portfolio using both the MBA and OTS standards. We introduced in this topic on the last call and I wanted to spend a little more time this morning on it again. While we normally use the MBA standard for corporate purposes, many other institutions use the OTS standard. Because most mortgage loans are due on the first day of the month, there can be a large difference between the two delinquency measures; and you can see that here on this page.

We show delinquencies on the left side of page 8, total delinquencies, I should say; and 90-plus delinquencies on the right side. 90-plus is our standard for nonperforming assets. Notice that we are using UPBS, or unpaid principal balances, as our denominator.

I have sometimes seen nonperforming ratios calculated as a percentage of assets. The percentage of total assets approach will obviously decrease the delinquency ratio. On top of that, the usefulness of the ratio is perhaps diminished by adding securities, furniture, and other non loan assets to the denominator. I hope you can see how important is to know the calculation standard when evaluating or comparing delinquencies.

In addition to delinquency trends for the servicing portfolio, we also show delinquency trends for the bank HFI portfolio and for loans underlying our residuals. These two sections show delinquencies using the MBA standard.

Before leaving this page, let me provide a comment or two on pay options. The pay option trend in delinquencies is being affected by at least three components. First, delinquencies are up significantly in their own right. The slope of this trend is exacerbated by two additional factors. One is seasoning. Loans in recent quarters are more seasoned than those in earlier quarters. For June, the weighted average life for pay option loans in our servicing portfolio was 17 months, up from 14 months in the first quarter of 2007, and up from seven months from the first quarter of 2006.

The second additional factor is the denominator effect. We are seeing more pay options prepaying than are being added to the portfolio. The total pay option servicing portfolio is down 3.7% in UPB and 4.2% in loan count for the servicing portfolio again, in just the last quarter.

The effect is even more pronounced in the bank HFI portfolio. As you will see on page 11 in just a moment, bank HFI payout option UPBs declined by almost 10% in just the last quarter.

Let's turn to page 9. Page 9 is a duplicate of a page shown in last quarter's supplement. We include it again as a reference to summarize the financial statement geography of how credit costs manifest themselves on our income statement and balance sheet. Rows represent major risks; columns represent the relevant financial statement treatment for the various credit costs.

Let's turn now to page 10. On page 10, we refresh a schedule presented to you in last quarter's supplement. The rows match what appear on the previous page. Let me describe the columns. For each of the major columns we show values for both the second and first quarter of 2007.

Balance sheet assets, which are the first two columns on the left, is simply the carrying value of these assets on our balance sheet. The underlying loan UPB is the unpaid principal balance for the loans underlying each asset or risk. The provision/earnings charge is the amount we recognized on our income statement. Some of these charges are to a loss provision and some are charged directly to gain on sale.

Balance sheet reserve/estimate for losses is the amount that we have on our balance sheet for future losses. As shown on the prior page, this could take the form of a reserve in some cases, or the reduction of the asset's value in other cases.

We are going to cover the bank HFI portfolio in residuals in more detail on subsequent pages, but before leaving this page let me provide some color on some of the other current period charges. You will see a $91 million charge in the first row of numbers for HFS, which stands for held-for-sale, compared to $266 million last quarter. Last quarter's charges were heavily influenced by subprime. Approximately $81 million of the $91 million this quarter pertains to approximately $484 million of loans that were moved from HFS to HFI during the quarter. The loans were marked to the lower of cost or market as they moved from HFS to HFI.

These loans consist of delinquent government, prime, subprime, HELOCs, and fixed-rate seconds. As with our loan sale decision, the decision to move assets into HFI is done on a best execution basis. We often realize better economics by retaining delinquent loans than by selling them.

On the next row down for HFI, there is a $297 million charge for provisions. This provision includes the liability we accrue for unfunded HELOC draws, which is why there is a difference between $297 million here and then the $293 million that you will see in the press release. $246 million of the $297 million is for the bank HFI portfolio; we will see more on that momentarily.

The remaining amount relates to the HFI held outside of the bank; and most of that was additional provisions we booked on subprime loans. Recall from last quarter's presentation that the nonbank HFI portfolio is very different from that inside the bank. The nonbank HFI portfolio consists of loans that are part of our government rewarehousing program; rep and warranty related repurchases; repurchases made in connections with our exercising cleanup calls on securities; and loans moved from HFS.

Another important distinction between these two portfolios is the balance sheet credit held for future losses. Both HFI portfolios rely on traditional reserves and credit enhancements. The nonbank portfolio will also include basis adjustments where the loan's balance sheet carrying value was reduced through a charge to earnings. The rep and warranty charges principally reflect normal deal activity this quarter as well as some increased incidence of early defaults.

Mortgage insurance, which is roughly three-quarters of the way down the page, reflects normal activity this quarter. Recall that last quarter included the release of approximately $74 million of reserves associated with the 2003 mortgage reinsurance book. Servicing advances are the last row of numbers on page 10. Charges here are up over last quarter as a result of increased delinquency and foreclosure activity.

Let's turn to page 11. Page 11 shows reserves and reserve flow for banking operations for the held-for-investment portfolio. Let's start at the top and work down. The first section shows the flow of reserves for the past five quarters. Each quarter, we start with a beginning reserve, add provisions, and then deduct charge-offs to get an ending reserve. Footnote 3 is important for understanding recent charge-off activity.

The current quarter's charge-offs are elevated as a result of moving to a more aggressive charge-off policy for the entire HFI portfolio. We estimate the timing effect to be in the neighborhood of $30 million for banking operations and $50 million on a consolidated basis. The second section of this page shows reserves by key loan type for the same five quarters. Note that the reserve includes ALLL as well as the liability for undrawn HELOCs.

The next section shows UPBs in total and for key loan types. The far right-hand column in this section shows the percentage of loans or the percentage of UPB, rather, covered by credit enhancement. In the second quarter, we secured approximately $5.7 billion in additional credit enhancement for pay option loans held in the bank's HFI portfolio. Of the 77% of pay option loans now covered by credit enhancement, approximately 48% of this coverage is first loss coverage, and approximately 29% is mezzanine coverage.

At the bottom section of this page, we show reserves as a ratio of UPB, nonperforming loans, and as a percentage of the nonperforming loans that are not covered by credit enhancements.

Let's turn to page 12. Page 12 is an update of a schedule we showed you in last quarter's supplement where we show information for residuals. Residuals are first loss securities and represent one of our most significant credit exposures. The creation of residuals comes about mostly from the securitization of subprime and home equity loans.

The top section of this page shows subprime residuals, and the bottom section shows home equity. As with prior pages, home equity includes both HELOCs and fixed-rate seconds. Let's work across the columns from left to right. The first pair of columns with numbers show the carrying value at the end of Q2 and Q1. Note that the home equity carrying values include both residuals and transferor's interest.

The next column is UPB for the loans underlying these residuals. Moving to the right, we see the pool factor, which is simply the percentage of the pool that is currently outstanding compared to the original pool. The final four columns on the right show losses in dollars and percents for this quarter and last quarter for each of these residual categories. Subprime valuations this quarter are similar to the end of the first quarter, except for the addition of new deals during this quarter. Home equity valuations are down quarter-over-quarter, primarily as a result of higher future loss expectations.

Let's turn to page 13. Page 13 summarizes the major actions we have taken or have underway in response to the transitioning environment. The guideline cutbacks in subprime include the elimination of all second liens, the eliminations of the 2/28 program, the elimination of stated loans with CLTVs greater than 90%, the curtailment of 100% financing, the imposition of additional restrictions on first-time homebuyers, and increased FICO requirements for interest-only loans. More restrictions are either underway or being contemplated.

In prime, guideline cutbacks include curtailment of 100% financing and adjustments to 95% financing. Exception boundaries have been tightened. As with subprime, more restrictions are underway. In addition to guideline adjustments, we have made and are making a number of other control enhancements. CLUES, our automated underwriting system, consists of a rule set and quantitative scorecards. As we note on page 13, the quantitative scorecards have been recalibrated to reflect recent experience. We have also increased escalation requirements for higher-risk loans.

Other controls are designed to detect and address speculative activity on the part of either the borrower or the property seller. We continue to use geography-based controls including filtering out certain geographies for the bank HFI portfolio.

With that, this concludes my comments, both on this page as well as the supplemental presentation. I will turn the call back over to Angelo.

Angelo Mozilo

Thank you, John. Before we go into the Q&A period, many of you are aware of the increased volatility in the secondary market. I thought it was appropriate that Kevin Bartlett, our Chief Investment Officer, provide a few brief comments on what has recently occurred and its implications to Countrywide. Kevin?

Kevin Bartlett

Thank you, Angelo. In the secondary market, three key events have occurred this year. First, we had the disruption in the subprime markets in the first quarter. When we discussed this event during the first-quarter earnings call, the market for subprime loans had worsened; and the opening questions at the time were the spillover impacts to Alt-A and other products like pay options.

The other unknown was the impact on the CDO participants, which had been the largest buyers of principally mezzanine credit risk for some time and had previously pushed credit spreads to historic lows. The more traditional buyers of credit risks, banks, insurance companies, money managers, and hedge funds were in part driven out of these markets due to the lower spreads.

Also during the first-quarter earnings call, we discussed changes in our subprime programs. We noted that we expected the mix to change rather dramatically and overall volumes to decrease. John McMurray covered this in one of his slides as part of the supplemental presentation.

During the second quarter, the market for subprime loans improved dramatically as CEO investors returned to buy the reduced supply of credit-sensitive securities, owing to the perceived improved quality of the newer production as underwriting guidelines were tightened. Credit spreads for key portions of the securitization structure rivaled the tight spreads the market saw before the disruption in the first quarter.

Our resulting gain on sale for subprime loans reflected this improved market. In addition, we also saw the Alt-A market improve, and subordinate bonds off those securitizations and whole loan sale exits were purchased at tighter spreads.

Towards the end of June, the second key event occurred, BSAM. As is well known, the BSAM funds came under pressure as a result of investments in credit-sensitive securities and resulting negative performance. As it became evident to the markets in late June, BSAM's sizable holdings added a significant uncertainty about the supply of credit-sensitive bonds to the market. As a result, spreads began to widen, and the last few second-quarter securitizations were impacted negatively.

Despite this late June development, we distributed most of the credit-sensitive portions of our securitizations during June. For home equity loans, the increased costs of those securitizations as a result of recent performance trends made portfolio investments in these loans economically more attractive. As a result, we retained a substantial portion of our second-quarter home equity production in our investment portfolio.

Finally, the third key event happened after quarter end on July 10, when S&P placed on credit watch for potential downgrade a large number of subprime-related securitizations and also enumerated changes to their rating approach. Changes to their approach are not well understood by market participants, as S&P did not fully disclose the exact impact of their newer methodologies. The market has been left with future downgrade uncertainty for many credit-sensitive products, Alt-A, second liens, and pay options, in addition to subprime loans.

At present, our view of the market is that, until the market can understand future rating actions, buyers don't know what to expect on their portfolios and don't desire to purchase additional credit-sensitive bonds until the impact on ratings are better identified.

In addition, shortly after S&P's actions, Moody's downgrade many subprime deals. As the rating agencies' intent becomes more apparent, my personal view is that the traditional investors in credit-sensitive bonds like banks, money managers, insurance companies, and hedge funds should reemerge as credit spreads widen and the market perceives the greater level of protection afforded by the new rating agency approaches compensates for the risks and reduces downgrade uncertainty.

But a large question mark remains as to whether the CDO investment vehicles will recover to become buyers again. As for the impact on Countrywide, our current views are that we will be able to sell at least the AAA-rated portions of our nonagency products; but that we are prepared to invest in the subordinate bonds until market conditions improve. We believe that we have the capital and liquidity to make these investments until the market adjusts. Our expectations are that the current market forces are likely more likely to force the weaker mortgage companies to either reduce their activities or align with stronger players.

We have also made many changes to our product offerings, pricing, underwriting guidelines, and processes, in order to improve the quality and secondary market execution of our production. John McMurray spent a few minutes on these changes during his presentation. While the exact impact on volumes and gain on sale margins is unknown, we expect to see lesser volumes and reduced gain on sale margins until the current market adjusts. Angelo, I will turn it back to you for the start of Q&A.

Angelo Mozilo

Thank you, Kevin. Okay, now the floor is open for Q&A. So we will start taking some questions.

Question-and-Answer Session

Operator

(Operator Instructions) Your first question comes from the line of Paul Miller from Friedman, Billings, Ramsey. Please go ahead.

Paul Miller - Friedman, Billings, Ramsey

Yeah. Thank you very much. Angelo, you have been through a lot of these different cycles out there, and this seems to be becoming a lot stronger and a lot harder than a lot of people thought. When do you think this ends? I know you have been quoted out there you think this, maybe it is an '09 event, but what's some of the things we have to see?

Does the CDO market have to come back? Or does liquidity have to come back? Or do the inventories in the housing market have to go away? What are some of the things that we need to look for?

Angelo Mozilo

I think the first thing is that the inventory in the house supply has to reverse itself. Because as I view it, and I have been through a lot of these things in 54 years, although the market is a lot bigger now, so the problems are a lot bigger. But as I try to walk through what happened here, and could a lot of this have been foreseen, if these tend to try to reflect on your own activities, and should we have known, could we have seen it?

But as I do reflect on it, and I do a lot, that nobody saw this coming. S&P and Moody's didn't see it coming, but they simply just downgrade bonds, they don't take hits. Bear Stearns certainly didn't see it coming. Merrill Lynch didn't see it coming. Nobody saw this coming.

So it was the deterioration in real estate values that was the base cause of all of this. We had none of these problems as real estate values were going up. So it is an oversimplification, admittedly, but clearly the deterioration in real estate values as a result of the affordability issue and an oversupply.

So I think one turning point is the supply of housing. Because I think once that turns around, the psychology of the country changes. Because now the borrowers, potential buyers simply say, look, I will just wait, because I will be able to buy the house cheaper tomorrow than I can today. That psychology has to change; and the only thing that is going to change it is supply.

Just so I can reflect on this as you people think of your questions. The other is that the Fed knowing that well over 50, 60, 70% of the loans made in 2003, '04, '05, and '06 were indexed variable-rate loans, indexed one way or another to the Fed funds rate, increased the Fed funds rate 17 times. 17 consecutive times, with most of the product out there being variable-rate product.

You know, and you never knew when they were going to stop increasing. But the fact they did that had a material impact on affordability. As people went to refinance or people went to buy, major, major impact.

So for a Fed Governor to say that the lending industry had this coming is unbelievable when the Fed, to a great extent, was unknowing contributing factor to this. Plus they came in with the Fed, with the joint AC guidelines, which restricted the type of loans we could make in an environment where we had limited liquidity to start with.

The other was the rapid increase in real estate values that we have faced over the last few years caused people to stretch themselves and their financial resources in order to get into a home. There is a high desire for people to own a home. The mortgage product then was developed to try to help them get into the home, which was known as exotic products of various kinds. But there was a secondary market for it. There was plenty of liquidity for that. So the primary market responded to that.

I say also the traditional underwriting standards, which John McMurray alluded to, were altered to use more technology and more credit scoring as a judgmental factor in whether a loan could qualify or not, versus the traditional documentation. And you had on top of that, which again John related to, was speculation. So I just want to share those thoughts with you.

Now getting back to your base question, as I say 2009 because my experience is that it just takes a long time to change, to turn a battleship around. This is a huge battleship and it is headed in the wrong direction. So first, we have to get it to slow down, then stop, and then turn. I think this is just a gut feeling, based upon what I have seen in the past and the size of this market -- is that it's going to take 2007, the balance of this year, to get this thing to look like it is slowing down; 2008 to get it to slow down and stop; and 2009 to head in the other direction.

My feeling is that by that time, you will have reduced competition, very substantial pent-up demand, because of all of the people who have been closed out of housing, probably lower interest rates. I do think that this ultimately has to have an effect on the economy. I just can't believe that this economy is totally insulated from housing, and will be 2009, 2010, 2011, I look to as sort of like 2003, 2004, 2005, great years for the industry.

Paul Miller - Friedman, Billings, Ramsey

Okay. Thank you very much, Angelo.

Angelo Mozilo

Very welcome.

Operator

And your next question comes from the line of Ken Posner. Please go ahead.

Ken Posner - Morgan Stanley

Hi. Good morning, Angelo, and John. I wanted to ask, if I could, about the performance in the prime portfolio. I have got a lot of questions from investors wondering does this mean that all prime mortgages are in trouble? Or is there something specific about your portfolio that could explain the increase in charge-offs from 4 basis points last year to 19 basis points in the first quarter to around 47 basis points in the second quarter?

What specifically is driving that, because these are prime people, right; they haven't lost their jobs and the housing markets have only just started to soften?

Angelo Mozilo

Since you mentioned that, before I turn it over to John on this issue, so far what we have seen in delinquencies to a great extent are not resets at all but people losing their jobs, loss of marriage, loss of health. And the problem is they can't either refinance because the value of their homes have gone down, so they are under water, or the program that they used to get into the home is no longer available to them.

So right now, the delinquencies are being driven by more traditional issues than they are by the concern about resets. John?

John McMurray

Sure. So the way I think about prime is that it covers a very vast spectrum, so I don't thank you can paint it with a single brush. Part of the reason that we shared the odds ratios charts with you is to show you some of the key things that are driving both prime and subprime performance.

So in the prime sector, recall how I showed you the line that covered FICO. There is a belief by many that prime FICOs stop at 620. That is not the case. There are affordability programs and Fannie Mae, expanded approval, as an example, that go far below 620, yet those are still considered prime.

Documentation and leverage are also important factors. So if you have the existence of one of these high risk factors or the combination of several of these factors in a prime loan, it is going to exhibit higher delinquencies.

Ken Posner - Morgan Stanley

I understand. John, would it be safe to say then that in Countrywide's prime portfolio, it would not be the FICO scores, obviously; they're very high. It would not be the LTVs; they're very moderate.

But we should assume then that there are, in fact, multiple risk factors that are driving the spike to 47 basis points in charge-offs, because that is just not a charge-off ratio one would expect for at least for an old-fashioned prime portfolio.

John McMurray

Well, a couple things. Again, remember even in prime, there is a very broad distribution. So even though on average, the FICOs are very favorable and the CLTVs are very favorable, you're still going to have tails at both ends, both very good loans and loans with a lot of high risk factors.

The charge-offs ratio that you're seeing in the bank, also, as I said, be sure to see footnote 3, a substantial portion of what you see this quarter relates to a change in our charge-off practices. So some of the charge-offs that would have been taken in future quarters were accelerated into this quarter. So.

Ken Posner - Morgan Stanley

John, I will confess, I missed footnote three. Could you explain what the magnitude of that effect is, and maybe just take a second if that is an important factor?

John McMurray

Sure. So for the HFI portfolio across the firm, we adopted a 180-day charge-off standard. Prior to that, for much of the HFI portfolio we used a standard that is similar to what is used in our securities, where we wait for the loan to go into foreclosure and through REO before doing the final charge-off.

The advantage of the approach to that we were using is that it is much more precise. The later we go in the process, the more certainty we are going to have, exactly, with respect to what is charged off.

The advantage of the approach that we moved to is that you charge the loans off earlier, though you do lose some of the precision. But in any case, by adopting those standards firm-wide, as I said, we moved some of the charge-offs that we would have had in future quarters into this quarter.

Ken Posner - Morgan Stanley

Did you quantify the impact from that change? Then I would like to go back in line.

John McMurray

I did, but just as a quick reminder -- and I wanted to make one other comment. So inside of the banking operations segment, the impact is going to be in the neighborhood of $30 million for the quarter. On a consolidated basis, it's going to be in the neighborhood of $50 million.

One final quick comment. As we have talked about in prior calls and during our presentations, Countrywide is a mortgage supermarket, so we generally offer what is offered by the market at large. So it is my belief that the portfolio that we have for the most part is going to be a good reference for what exists on a broader basis.

Then also, keep in mind the comments I had on the delinquencies. Many firms report out using the OTS standard. Eventually those delinquencies are going to catch up with the MBA standard, but you're not going to get as early of a read.

Ken Posner - Morgan Stanley

Thank you very much.

Angelo Mozilo

Ken, I would also comment that when you made the comment about the traditional prime and don't expect that kind of delinquency, I do think it is -- be important to observe what happens going forward, because we are experiencing home price depreciation almost like never before, with the exception of the Great Depression. So I think using standards or frames of reference on prime and the performance of prime in other environments may not be a fair comparison in light of what is happening to real estate values.

Ken Posner - Morgan Stanley

Point taken. Thank you.

David Sambol

Ken, this is David Sambol, let me also weigh in. I think you were attempting to identify what maybe subsegments of the portfolio have contributed to higher charge-offs and delinquency trends. If we had to identify any particular category of the home equity portfolio, it certainly would be the higher CLTV and the higher CLTV reduced documentation loans that we have in the portfolio, that are disproportionately contributing to charge-offs and delinquencies.

Many of those stem from the higher concentration of piggyback financing that we did and that we have in the port stemming from what was occurring in the market. I point out that when we originated those loans we certainly were cognizant of the fact that they would perform differently than traditional home equity loans; and we attempted to price for that.

So that one of the data points on John's presentation that you might have noted was the pricing on our average HELOC book, expressed as a margin over prime, and it is close to prime plus 190. You would compare that to, for example, typical stand-alone HELOCs historically done would be priced at prime flat, or even prime minus for many of the banks.

So that was a risk premium that we add to the pricing for those loans. But certainly, we didn't anticipate what would happen with HPI. Again, as John illustrated, the correlation between significantly up-sloping losses and the kind of environment that we are seeing.

We also enjoyed very robust earnings on that book with negligible charge-offs and losses historically as a result of the very benign environment we have come out of the last two years. Now we are really seeing some of that risk premium we charged or embedded in the loan being absorbed. So that is kind of the way I look at -- it's another way to look at what we are seeing.

Ken Posner - Morgan Stanley

That makes a lot of sense. Thank you, David.

Operator

Your next question comes from the line of Eric Wasserstrom. Please go ahead.

Eric Wasserstrom - UBS

Thanks. Just one point of clarification and then one question. I guess the point of clarification is I kind of summarize these comments that you have made on credit quality. Is it just too simple to say in areas, whether they were in subprime or prime, where you had excessive risk layering with an excessive dependence on home price appreciation that you're currently seeing the process of that unwinding in terms of the credit performance?

Angelo Mozilo

I think that is an oversimplification. We didn't rely upon home price appreciation. We relied upon an appraisal of current value as a lender. You can't forecast as a lender, which way values are going to be going, up or down. You at least traditionally and historically use appraisals, with certain protocols to establish what the value is; so you can determine what kind of loan-to-value you want to establish based upon the quality of the credit of the borrower.

So we did not make these loans based upon value is going to go up forever, because we know they are not going to go up forever. But you don't know what they are going to do, whether they are going to stabilize, go down; if they do go down, how far? It is just unpredictable.

So in our business, you're limited to what you believe the value to be based upon a professional appraiser valuing a piece of property at that time.

Eric Wasserstrom - UBS

Maybe I will follow up with a little bit more about that off line. The question I have is --just to get away from this topic for a moment -- you know, you guys have engaged in some capital optimization and some other elements of financial engineering. I think efforts to migrate some of your MSR assets into the Bank. Can you just give us update on where those various efforts are currently and what that suggests about capital optimization from here?

John McMurray

Yes, well, two things. In terms of our capital position, I would characterize where we are at as we are in good shape and strong capital position. We have talked in the past about having excess capital. We remain of the view that we have excess capital in the near-term, certainly over the next quarter or so. We intend to put that capital to use by investing in the balance sheet.

As Kevin mentioned, we expect to use our balance sheet to hold additional assets as a result of some of the volatility out there until the secondary market conditions normalize.

We also as a result of credit spread widening, we view this as an environment that will throw off opportunities for the Bank to accelerate growth of their loan portfolio, their HFI investment portfolio. That is evidenced by our decision during the quarter to move some home equity loans that we originated into the Bank.

So this environment will give us an opportunity to deploy capital back into the business as a result of spread widening. We also expect that other opportunities will present themselves over the quarter as the stressed environment results in issues for some of our competitors. In fact, we're already beginning to see -- this is a leading indicator I could share with you that over the last quarter our growth in the sales force was probably at a one- to two-year high. I am trying to determine the last time we grew it at that pace.

But all because things are beginning -- consolidation is beginning to accelerate and that means more opportunities, hopefully more production of volume, which will require balance sheet. So we don't expect -- we talked about cap optimization in the past; we don't expect, certainly in the near-term, any buyback.

In terms of your question about MSRs and the Bank, I think we previewed that it was our -- that we have a process that is occurring whereby we are seeking to migrate our mortgage banking operation from CHL into the Bank as part of our plan. We anticipated that beginning in 2008, we would start investing in MSRs within the Bank. We are still on track for that. Our plans are on track, but that is unlikely to occur before next year.

When it does occur, what we'll see happening is the Bank MSR portfolio beginning to grow, and the MSR portfolio in CHL beginning to shrink.

Eric Wasserstrom - UBS

Thanks very much.

Operator

And your next question comes from the line of Chris Brendler. Please go ahead.

Chris Brendler - Stifel Nicolaus

Hi. Thanks. Good morning. I guess one thing I'm struggling with is the undiscounted losses embedded in your home equity and subprime, and I guess also the Bank portfolio. I'm surprised that you haven't seen I guess issues besides life events in terms of rising delinquencies. So what I'm trying to get a handle on is, given how it seems like we're still early in the housing price decline that could accelerate as the year progresses and into 2008, how conservative or how appropriate are these marks today? What is the housing outlook embedded in there?

Maybe you could give us any color. I do think we have seen some resets hit in the second quarter. How are the resets performing? Doesn't that add a wave of pressure on these valuations? Also, what is happening from a loan modification standpoint? Thank you.

Kevin Bartlett

All right, let me work backwards from your questions. On the loan modification front, it has been our long-standing practice to always pursue a modification or a workout with a borrower as a first line of attack for somebody that falls delinquent or gets into trouble.

Two reasons for that. One is it's the right thing to do for the consumer; and secondly, for Countrywide it usually results in better economics. So on the modification, in the area of modifications, we are going to continue with the philosophy that we have had, although there obviously will be more of that activity.

With respect to your second to the last questions, which had to do with resets, we haven't seen a lot of that activity yet, with the exception of subprime. So if you look at the 2/28 and 3/27, if you look at the 2/28 and 3/27 subprime loans, most of the loans tend to prepay prior to the reset. But those that are left subsequent to the reset tend to have much higher delinquency rates.

So, one of the concerns that we have is just the less availability of subprime financing as a result of the guideline cutbacks. So we are watching that carefully.

Now, let me go back to your first question. So did you want me to address that with respect to residuals?

Chris Brendler - Stifel Nicolaus

Residuals, I guess, and the bank portfolio. It seems like it is hard to figure out where we are. I thought last quarter --?

John McMurray

Let's do a couple things. Let's start with a key distinction between the Bank portfolio and residuals. In the Bank HFI portfolio, credit losses are provided for via a traditional reserve. So that reserve is going to consist of two parts.

The first part is a general reserve. So that is going to cover loans that we expect to become 90 days or more past due in the next 12 months. So it provides for 12 months of coverage there. On loans that are now 90 plus or higher delinquent, we provide for a lifetime loss expectation. So that is the Bank HFI portfolio.

In the residual valuation, there what we're doing is we are estimating losses over the life of those loans; and then that is being incorporated into the cash flow. So you're going to have a combination of interest-only cash flows or excess servicing in effect, plus the projection of losses. Those are netted together and then discounted back to the present. So on residuals, we are able to always contemplate our current view of lifetime loss expectations; that is not going to be the case in the bank portfolio.

Kind of the final comment here is that given the environment that we are in, I still think there is a fair amount of uncertainty with respect to anyone estimating what future losses will be. In fact, most of the increase in loss estimates that we made on the home equity residual portfolio this quarter was basically entirely on loans that are currently not delinquent. In other words, they are current. So they are loans that we expect to become delinquent and default in the future.

Chris Brendler - Stifel Nicolaus

Let me just ask it a different way, I guess. The sequential change in some prime loss rate was zero; so you are still looking at 6.3% losses. If I took the pool factor of 54%, so you're looking at cume losses I think which are basically half that, assuming that most of the early loans didn't have any losses. They prepaid.

So given Angelo's comments about the housing downturn extending into 2008, maybe 2009, I'm trying -- and given the fact that housing data over the last three months has been pretty much a train wreck -- how do we reconcile? How do I feel comfortable that there is not further residual write-downs coming?

John McMurray

Well, we believe we have a -- well, not we believe; we have a housing price forecast embedded in the valuation that is more severe than anything that we have seen in the past. I hesitate to give any comfort, because the future is uncertain. So it is possible that it will have to be revisited. We hope that is not the case.

David Sambol

This is Dave Sambol, let me see if I could give a little bit of extra color. With respect to the remaining losses we have provided for on an undiscounted basis in the subprime portfolio, it is the case that it is approximately 6.3% as John reflects. But it's a much higher number on more recent vintages like 2006, and a lower number on older vintages that have more accumulated equity appreciation.

So I would make that point. I would also make the point that the entire balance sheet exposure in subprime residuals -- while not immaterial on an absolute sense -- is only $382 million, representing -- excuse me, $442 million, representing an aggregate kind of cap as to what the exposure could be if we were wrong and additional impairment was necessary.

As it relates to the home equity write-downs that we took, and fairly significant increases in projected future losses that drove that write-down and is embedded in our current numbers that John showed, I would point out that we're providing not only -- a significant portion of that is on current loans, and not necessarily on delinquent loans.

We're providing for future losses at a level that is greater than anything that we have ever seen. Just the size, kind of the scope of it. Again, absent a crystal ball, we don't know. But it is probably. as it relates to the home equity portfolio, our view, more conservative than what -- from our reading of other press releases and so forth -- that we think others are providing for.

In terms of rate resets, that was mentioned and Angelo commented on that. Just to elaborate on that, as we -- and I think I discussed this last quarter. As we look at and do an autopsy or post mortem on loans that go through the foreclosure sale process, for the purpose of identifying the driving cause and what the consumer told us, it really is to date the case that a very, very small percentage of those customers that lost their home through a foreclosure lost the home as a result of payment shock or rate reset.

I think the numbers are close to 60% are attributable to some form of loss of income or material curtailment of income, something happened or -- and this is slightly under 60%. Another 25% is attributable to death or divorce or disability and illness. Then increasingly we're seeing investors; a significant percent now is explained by investors that are unable to get out of the property. A very small percent really, so far, is explained by rate resets.

In terms of the last topic you mentioned, modifications, as John mentioned we do, I think, as much if not more than most everybody in ensuring that we keep people in their homes. Not only because that is the right thing to do, but because it mitigates our losses as well, as I'm sure you know.

Just in the month of June, we effected approximately 2,000 modifications on loans that were otherwise in the foreclosure process that resulted in the loans leaving the foreclosure process. That compares to approximately 1,000, it is more than doubled or approximately doubled one year ago. So modifications are being used, wherever it makes economic sense and otherwise makes sense for the borrower.

Angelo Mozilo

Okay, I just wanted to make a clarification. I know the press is on the line and get you out of that statement that. I am not saying that house prices are going to decline from 2009. Going back to my battleship analogy, I think it is going to take all of 2007, all of 2008, to get this whole negative process, house prices, increased delinquencies and foreclosures to slow down; and the balance of 2008 to get that battleship stopped and 2009 to get it going in the right direction.

As to when house prices will decline, I don't know. What I was referring to was in terms of the performance of the industry, performance of Countrywide more specifically, that at least my view of it and just take it as that, this is a visceral view of it just based upon past experience. It does not mean it's going to happen that way.

But that we at least internally prepare ourselves to make sure that we are doing everything we can during the next year and a half to reverse the trend and then to take advantage of the opportunities we're going to have in 2009. That it is what I am referring to, not that -- I don't know when house -- the cessation of house pricing declines are going to take place.

Chris Brendler - Stifel Nicolaus

Thank you very much.

Operator

And your next question comes from the line of Moshe Orenbuch. Please go ahead.

Moshe Orenbuch - Credit Suisse

Thanks. I was wondering, in the slide presentation you mentioned that about three-quarters of your option ARMs were covered by mortgage insurance. Could you talk about, first of all, how that will play out in terms of when you will actually be able to draw on that; and how that is going to impact; and your future expectations for the provisioning within the Bank?

John McMurray

Sure. Well, to start with, the provisions that we make contemplate this credit enhancement. As I mentioned, roughly 62% or I think I said which equates to $47 million out of the $77 million or so -- is first loss. So what that means, as soon as a loan that is covered by that policy defaults, because the coverage is first loss, we will be able to submit a claim immediately.

There's other coverage that is mezzanine coverage. So the way that will work is the pool of loans that are covered by those policies are going to have to accumulate a certain level of losses before we are able to start submitting claims.

But again, on just under two-thirds of the coverage it's first loss, so we from the very first loss on loans that are covered we can submit claims.

Moshe Orenbuch - Credit Suisse

How does that impact your provisioning as you go forward? And what should we look for in terms of that provisioning?

John McMurray

It affects the -- if a loan is covered by a credit enhancement, particularly a first loss credit enhancement, that is going to affect the calculation of our reserve, because the expected loss is provided for via a credit enhancement as opposed to charging it off against a reserve that we're holding on the balance sheet.

Moshe Orenbuch - Credit Suisse

Do you have a sense as to how much that benefited the current quarter?

John McMurray

It basically has had very little to no benefit so far, so those benefits are going to come in future quarters. So it is not been used heavily yet; but it is there standing by in case pay option losses start to ramp up.

John McMurray

Next question.

Operator

Your next question comes from the line of Fred Cannon. Please go ahead.

Fred Cannon - Keefe, Bruyette & Woods

Thank you and good morning. A question. You guys floated about $4 billion in convertible securities in the second quarter, and they had a pretty low coupon on them because of the convert. I know there were some accounting issues. I was wondering if you could tell us what the interest charge that went through the income statement for the quarter on those was?

Eric Sieracki

Hi, Fred. You are correct. We did issue $4 billion of convertible debt. One of the key advantages that we had -- and there are were many. We were able to tap investor groups other than MTN investors. We were able to save a significant amount of coupon; probably about 325 basis points was the savings on the coupons. So $4 billion, 325 basis points for a quarter would have been theoretically the savings.

We didn't have those converts outstanding for the entire period. But it also should be noted that there are some ramifications of future accounting promulgations. We may see in line with fair value reporting the GAAP benefit go away from the difference in the coupon. But at least for now, until that new promulgation comes about, we do get the benefit of the lower coupon through P&L.

Again, it would be 325 basis points times the $4 billion; if it was outstanding for the full quarter, you would have to quarterize that.

Fred Cannon - Keefe, Bruyette & Woods

Okay. Then, I know they have a fairly short option on them, for the owners to put them back to you. If that occurred, would you -- what is your thinking on this security, Eric? Did you like it? Would you do it again, I guess, is the question?

Eric Sieracki

Obviously, we like the security. We were able to place almost 40% of our full-year financing away from MTN investors, away from term debt investors. Got the 325 basis point reduction in the coupon. We are able to fill the Delta short in connection with that trade as well. So it was a great trade.

There were two tranches, $2 billion each. One was 17 months; one was 24 months. So we would very happily consider issuing additional converts in the future.

One of the things we're trying to do was monetize very high VAL in CFC equity, which is a Company asset that is difficult to monetize, and through this trade, we were able to do that.

The bottom line is that we were at $48 million in interest expense on those $4 billion of converts during the quarter.

Fred Cannon - Keefe, Bruyette & Woods

Great. One other question. I know you're talking about making incremental investments given the market conditions. And I noted at the Bank I think your securities portfolio was -- most of the growth in the Bank assets was -- in fact, all of that lives in the securities portfolio, which would appear to be earning a relatively modest spread in this environment. Could you kind of expand on the strategy regarding securities investments at the Bank?

Angelo Mozilo

Let me just -- we will have, I guess, Carlos cover that, but let me. I would just point out, maybe modest, but it's safe, which is important particularly in this environment.

Secondly, it was an alternative course because we couldn't find the assets, the appropriate assets that met all our hurdles for the Bank. It was a good alternative for us at the time to put assets in the balance sheet.

The negative part of that is that it exacerbates delinquencies because you can't put that in your denominator in your delinquency calculation. It is not a loan. So the delinquencies in the Bank are exacerbated by that factor. But they're a good safe AAA investment. So Carlos, do you want to talk about it?

Carlos Garcia

Yes, I would mirror what Angelo just said and just add that in this period where credit risk is a key consideration, these securities offer the Bank a transitional investment strategy that provides us with, A, a low level of credit exposure, very high-quality assets, low risk base capital implications, and predictable returns on the security cash flow themselves.

But I should remind you that the securities are short-term in terms of their expected life, and they have a significant prepayment. Those prepayments will be reinvested; and the ultimate return will depend highly on what those cash flows that come back to us get reinvested in.

We don't see the housing market staying permanently this way. We don't know what to call the end of it. But certainly at some point in time it should improve.

Fred Cannon - Keefe, Bruyette & Woods

I guess, Carlos, a question I had also is kind of vis-à-vis Kevin's comment about willing to invest in certain parts of the straight loans. Given that growth in the securities portfolio, should we expect that to stabilize or decline, given that short duration? Start to see loans go back up? Or are you guys being pretty cautious regarding credit risk at the Bank at this point in time?

Carlos Garcia

First, let me say that we are being very cautious with regards to credit risk. Secondly, let me say that in this environment, I think Kevin and others alluded that credit spreads are widening and proposed investment is a good thing, because that means a better return. So we will be very opportunistic about what assets we select. But the opportunities, I think, will be greater.

I should also point out that some of the suboptimal returns that you may see in the current quarter represent a snapshot in time of the return of an asset. The lifetime returns of these assets are very good, but we have just come out of a period that suffered benign credit into a very difficult period. So we're transitioning something that required less reserves to something that requires a lot of reserves. And that is all getting concentrated in a few quarters.

But going forward, the returns -- you know, assuming our assumptions about the future are correct -- these reserves will start to improve. Certainly the new assets that we put on the books are all priced at very acceptable, in fact attractive, returns.

Fred Cannon - Keefe, Bruyette & Woods

Thank you.

David Sambol

Yes, I would also add -- this is Dave Sambol -- that maybe emphasize that one of the positive byproducts of the current volatility in the market is this widening of credit spreads. It is just providing the Bank an opportunity to grow at HFI portfolio, its residential loan portfolio, more than it has been able to over the last several quarters.

Because credit spreads have been so tight that we do not believe that we were getting sufficiently compensated for credit risk, and that explains why our investment strategy became weighted to more very liquid high-quality securities.

That is changing. In the near term, we do expect -- as is the case in the second quarter -- to maybe divert more production that otherwise meets the Bank's credit criteria filters to the Bank as a result of wider credit spreads. Kind of the flipside of maybe the negative impact on the production sector margins that we anticipate from these wider spreads and lesser sales and associated gain on sale in the second half of the year.

Fred Cannon - Keefe, Bruyette & Woods

Thanks.

Eric Sieracki

Fred, I have a correction on that interest expense on the convertible. Based on the amount of time the convert was actually outstanding for the period, the interest expense on the convert was actually $11 million; and the approximate savings versus term debt that would have been comparable to that would have been about $14 million. So correction -- the amount of the expense was $11 million.

Fred Cannon - Keefe, Bruyette & Woods

Thanks.

Operator

Your next question comes from the line of Brad Ball. Please go ahead.

Brad Ball - Citigroup

Thanks. Trying to sort of work through how this all shakes out in the end, it seems like the problems at first were contained within subprime; we attributed it all to irrational behavior, some players that were overly aggressive during the decline in industry volumes doing things they shouldn't have done.

Now it seems that we have shifted to a discussion that surrounds prime. I am curious if you would characterize some of the behaviors in the marketplace as irrational. I wonder if you could comment, given that we have the benefit of hindsight, if we went back a year or two and you were in the marketplace at that time, what would you have done differently? Would you have put in place the actions that you have underway today, the operational responses that you discussed? Could you have put those in place a year or two ago?

Angelo Mozilo

Yes, I think your last question is a very good one. Could you have put it in place? The obvious answer is that the deterioration in house values, we knew that; that it was going to go back. We would have had to really stop doing that business, and the Company would have been a very different Company because you can't do this absent competition.

So our volumes, our whole place in the industry, would have changed dramatically because we would have arbitrarily made a decision that was contrary to what everything appeared to be. Values going up, and no delinquencies, no foreclosures, and we suddenly stop the music and say that we're not going to participate in home equity loans, in subprime, in high LTVs, no docs, and that sort of thing.

It would have been an insight that only, I think, that a superior spirit could have had at the time. So I don't mean -- because I think it is a very good question. Because I ask myself that all the time as CEO. That is the kind of question I ask. What should I have known and when should I have known it, and what should I have done about it?

As I go through that process, it is obvious that if we had stopped participating in those major areas of the business, we just couldn't stop it there. It would have affected us through the entire spectrum of our lending operations. Because you can't simply say we are out of subprime, we only want prime. Because the providers of loans provide both subprime and prime both, and will not give you the prime if you're not willing to take the subprime. That kind of thing.

So. But I think in the theoretical sense, would we do things a lot differently, knowing what we know now? Absolutely. We would have done a lot of things differently. But we didn't. The fact is we didn't know.

I think as John pointed out, the first part of your question as to prime, prime is a -- does have a lot of definitions attached to it. It is CLTV driven, it is FICO driven, it is credit, it is all downpayment driven, it is credit history driven. So it is hard to say that a prime loan is not a 70% LTV, 780 FICO. It is just not; it is a mixture of a lot of things.

So the spillover into prime, I don't think is something that should shock anybody once you understand the definition of prime. Because I think the basic issue you see today, particularly with Countrywide, is the spillover into the HELOC portfolio. At least for this quarter, it's not really a subprime story; it is a HELOC story and the deterioration in that, and the piggybacks that were originated in order to assist the mortgagor to avoid PMI and all the advantages that that avoidance provided for the borrower. And reduced costs and tax deductibility at the time, because PMI insurance was not tax-deductible at the time.

So it is a -- these are difficult things to retrace your steps on. But I think the issue now for us, speaking for Countrywide, is the HELOC portfolio. We will have to see what happens with subprime going forward. But I think that David pointed out that there is a limit to this in terms of our total. The total residual on our balance sheet is about $400 million. If everything collapsed, that would be the extent of our residual exposure.

But I do think it is a question of what happens in Alt-A. Alt-A is again a very broad definition of Alt-A, even broader than prime. We will have to see what the spillover effect is. But the bottom line is, as values decrease, the options for borrowers, homebuyers and borrowers, people -- and the combination of limiting their product available to them, is exacerbating the problem. We will just have to see how long that plays out.

In fact the Fed joint agency guidelines seriously restricted liquidity for borrowers to either refinance or for people to buy homes. I am not making a judgment whether it was right, wrong, or indifferent. It is just that that is what it did. Then combined with a volatile secondary market -- if you think about the perfect storm, that is the perfect storm.

Brad Ball - Citigroup

So it sounds like you are saying that what we are experiencing right now is not the product of irrational behavior, like the subprime debacle that came upon us late last year. But rather the product of an industry with excess capacity and pretty intense competition and a market environment that looked pretty good at the time.

Angelo Mozilo

I think that is fair to say. I mean, even in the HELOC if you look at the history of HELOCs by banks, I mean they -- very often it was 100% LTV. When they came in out of piggyback, when somebody came in for a second mortgage on their home or home equity line, it is very common to provide that type of financing by banks. And in an ordinary market, it worked fine.

It's this sudden and severe and deep deterioration in values, and particularly in certain parts of the country, has really exacerbated. Any even minor judgment area you're going to make has become a major issue. David?

David Sambol

Yes, just some additional perspective on the topic. I agree with Angelo. Certainly hindsight being 2020, there are things that would all maybe do differently. But I think it is important to emphasize some of the things we did do, and we are pleased that we did.

First, as it relates to the majority of our production, as you know, we sold all of the credit risk on the majority of the production, and that was the case in the majority of the Alt-A. Virtually all of the Alt-A production maybe with the exception of pay option ARMs, both fixed and adjustable rate, to retain no credit, retain interest.

We also sold last year a good portion of our subprime residuals, approximately 50%. We, as you now know, credit enhanced a significant portion of our held-for-investment portfolio in the bank.

So we took a lot of actions to mitigate risk, and beyond all that, we of course, as I said at the time, priced -- where we and the market liberalized guidelines, we sought to price for more risk and more liberalized guidelines. And much of that translated into fairly significant earnings over the last several years, where today only some of that is really now clawing back.

So that maybe gives some additional perspective on the topic.

Brad Ball - Citigroup

Thank you.

Operator

Your next question comes from the line of Bob Napoli.

Bob Napoli - Piper Jaffray

Thank you. We have a different moderator today. Couple of questions, I guess, first of all, on the growth of the sales force in this environment is it -- why do you feel that it is a good thing to be growing the sales force? There is overcapacity in industry. Overcapacity finally seems to be starting to roll over.

There is a lot of risk in the market, home prices are unstable. Yet, I think that Countrywide is one of the few companies that has been aggressively growing its sales force in this environment.

David Sambol

Yes, Bob. The reason is -- a lot of that is coming to us. The reason is that on the retail side, our investment and the costs associated with that sales force is very variable.

So the volume, the incremental volume, we pay them to the extent that they source incremental volume. And the incremental volume that they source is accretive to our bottom line, and it helps offset -- it will help offset some of the pressures on earnings from some of the other dynamics that we talked about.

So we are not paying big premiums applicable to the sales force expansion. We're not paying significant goodwill or big M&A premiums. It is coming to us organically, and the breakeven economics for new sales force hires is very compelling.

Several months after they're boarded, to the extent that they pass our filter, which they need to before we hire them, their contribution is accretive to us.

Angelo Mozilo

Bob, a couple things I think you should think about. One is that our market share, although rising 18%, 19%, as you saw this last quarter, in the retail sector it is very, very low. It is below 10%. And that is our most profitable, our most stable business.

We get repetitive business from them, cross-selling, refinances and insurance product. So it is very, very important that we take this opportunity, which is a rare opportunity to get high-quality salespeople into the retail sector to pick up our market share and to continue to grow the business.

The most stable part of the business we have is retail. And these people are very difficult to get when things are going well. So, as David pointed out, the economics for these people are very compelling.

The second point you raised, which is with deteriorating values and the danger of making loans today, it would seem to me that this is the time to be making loans when values are lower. Otherwise, you would only make loans when values are higher and always expose yourself to a deteriorating situation.

So that as values are going down and you stay in the business, sort of you're continuously building a better book of business, rather than just laying back, not playing in the game until values go back up and then get caught in that cycle again.

So you have to be in the game every day in a major way. I wouldn't -- the capacity in the business is being driven out quite rapidly, as you will see over the next six months, as announcements are made as to people going out of business or just merging. Again, this at least historically has been an opportune time for us to strike.

Bob Napoli - Piper Jaffray

So we shouldn't expect to see any announcements out of Countrywide you're laying off 500 people or -- I mean, is that…

Angelo Mozilo

No, I think you could -- yes. You certainly -- there's two ways, two things you have to look at. You have to bifurcate Countrywide's operations. One is its sales force, the people who make something out of nothing, who create value by bringing in business. And those who process that business, and that whole processing operation is driven by -- the number of people involved in it is driven by the volumes flowing through the operations.

As that volume decreases you, by certain measurements, begin laying off people. So it is certainly -- we are deeply involved right now in cost reduction measures. So that -- and our history is that we will, as you know, you have been involved with the Company a long time, we will take the appropriate steps in terms of reducing personnel costs when appropriate.

Bob Napoli - Piper Jaffray

Okay. Just maybe some color on the gain on sale margins that you have in your outlook. If you can give some color on the volatility in the secondary market. I mean, the agency business, where, how are the margins in the agency business? Is it -- are those margins stronger today? And the nonconforming business, how much of that would you feel comfortable holding on your balance sheet?

David Sambol

Bob, it's Dave. It is the case that the volatility that we made reference to today, and that which we anticipate, is primarily applicable to the nonagency business. Our forecast with respect to production margin -- again sale margin in the second half of the year was particularly adversely impacted by what we envision in the near term for the nonagency production.

Really, the reason for the forecast is the convergence of a number of factors that are creating what we see to be a very stressed second half, particularly the third quarter. We have seen recent trends on the application side, both refinance applications and purchase applications trend downward. We have seen, as interest rates have increased the latter part of the second quarter, and volume as new apps have fallen, increased front-end pricing pressures, competitive pressures on our margins, across the board.

Then, most notably here in the latter part of the second quarter as Kevin mentioned, and the beginning part of the third quarter, this very material disruption in liquidity for nonagency related products and widening of spread will we believe have a material impact on the back end on our gain on sale margins. Particularly for inventory that we carried into this environment from the second quarter.

We had a very healthy pipeline in inventory coming into the third quarter that is going to be subject to the spread widening stemming from what is occurring in the nonagency markets today. Really, we don't know when that will normalize and subside.

In terms of our capacity, we of course, in planning for a worst, for a stressed case, one where I think Kevin mentioned if we had to keep for all nonagency originations, all bonds below AAA, the approximate amount -- Eric, can you give us some numbers on the approximate monthly volume of additional assets we would be retaining in capital that that would consume?

Eric Sieracki

Sure. Talking about subprime, HELOC, fixed-rate, second, Alt-A, and pay option ARMs, contemplating only selling the AAA, keeping all the securities underneath that credit level, the incremental monthly equity drain would be about $100 million. The amount of debt we would have to raise would be about $600 million. That is incremental over our normal course of operations. That is a monthly level.

That is something that our contingent liquidity planning more than amply provides for. So I wouldn't call it rounding, but it is very easy for us to provide that additional financing.

David Sambol

Now the impact on second quarter or second half, I should say, production margins stems from the fact that even if we retain it we have to mark to market. So we expect that margins will be impacted by the wider spreads, certainly on the inventory coming in.

We also, as I mentioned, expect to sell less, particularly in the third quarter, and maybe divert more to the Bank given this environment, which will adversely impact production margins. And all of those factors are converging and were considered in our guidance and estimate of second-half production sector performance.

Bob Napoli - Piper Jaffray

Okay, that mark-to-market would go through the gain on sale line for the inventory?

David Sambol

Yes.

Bob Napoli - Piper Jaffray

Just on the credit side, we got -- it seemed to accelerate, obviously, in this quarter. As you said it was more of a HELOC issue this quarter. It was subprime last quarter. Fortunately, the job market has been pretty strong through this. So I wonder what it would look like if the job market starts to fall apart.

But do you see an acceleration coming out of this quarter into the back half of the year on the negativity of the credit statistics? I don't know if you can give any color on the rate of trend of change in credit statistics that you see.

John McMurray

Well, I think you make a good point, Bob; the unemployment situation has been favorable. So I do think if that was more adverse, we would see higher delinquencies and defaults.

Angelo Mozilo

I think that, however, again, these numbers are averages. If you look at three states were there's very serious problems -- in Michigan, in Ohio, and Indiana, where the unemployment rate is very high -- we are experiencing high delinquencies today in those three states.

Bob Napoli - Piper Jaffray

That was one of my questions as well, the geography. How regional is the problem on the credit side?

Angelo Mozilo

I'm sorry, just generally speaking and I will let John do this, but where you have had high acceleration in prices -- for example, in California where people stretched themselves to get into the home, as John pointed out in Sacramento, Fresno, Modesto, Stockton, you see high delinquencies for that reason.

In the states that I just mentioned, you see high delinquencies as a result of unemployment. In Dade and Broward County, in San Diego, as John pointed out, you see high delinquencies because of oversupply of condos and speculation. So it is regional, but each of these regions have their specific reason as to why we see exacerbated delinquencies. John, would you agree with that?

John McMurray

I would not only agree with that and let me just make two quick points. So geography matters a great deal. The first region -- the first reason it matters is with respect to home prices. So we see big differences from MSA to MSA, even MSAs that are close to one another.

The second reason that geography matters is, even after we adjust the effect of home prices out of the equation, we still see many geographies underperforming. So the Midwest as Angelo mentioned is an example which doesn't show up on a lot of the charts as having a serious risk of a home price depreciation going forward; still has very bad performance even after adjusting both for home prices and all loan attributes.

David Sambol

I would also point out that in the scenario where the economy worsened and we begin to see what has happened in housing spillover, the broader economy, and if unemployment picks up, the business as I imagine you know is subject to somewhat of an intrinsic hedge. It is an inversely correlated risk we have would suggest that the Fed in that kind of a scenario would be motivated to maybe ease. With an easing we would expect a bond market rally. And with a reduction in long-term interest rates, the nature of our model is such that we believe that the increased origination volumes and profits will likely more than offset the increased credit cost in that kind of an environment. So, that is important to understand as well.

Angelo Mozilo

It just doesn't go one way, Bob. Dave is absolutely right, historically, that if in fact this does have an impact on the economy, unemployment begins to increase, the Fed would have to lower rates. That is the overall cure for all of this, lower rates.

Bob Napoli - Piper Jaffray

Last question is any easy one. What was the ending fully diluted share count?

Angelo Mozilo

It's obviously not so easy…

Bob Napoli - Piper Jaffray

Thank you. You can get back to me with that.

Eric Sieracki

We will announce that shortly, Bob.

Bob Napoli - Piper Jaffray

Thank you.

Eric Sieracki

595, 540.

Bob Napoli - Piper Jaffray

Thank you.

Operator

Thank you sir. Our next question comes from the line of Ken Bruce. Please go ahead.

Ken Bruce - Merrill Lynch

Good morning. Thank you for the question. There is not a lot of encouraging news that we are hearing today. I was hoping maybe to go back and get you to answer a question that you danced around with Bob but I think is important.

Could you reflect back to how much of your business is agency eligible, and what the pricing dynamic is within that market, please? Obviously there is quite a bit of uncertainty as it relates to the nonprime or the nonagency business. But if you could just maybe discuss what the implications are for your margins with respect to prime and how that business may reshape or begin to reassert itself in terms of the prime component of your business model going forward, please?

Angelo Mozilo

We understand the question. The question is how much of our business that we're doing is agency eligible?

Ken Bruce - Merrill Lynch

Yes. You said in the past about 75% is GSE eligible. I had hoped to maybe update that number and just give us some sense as to what the pricing dynamic is there, please.

David Sambol

I would -- these are just rough estimates, but at the height of the housing and refi boom, the percentage of agency production in total production fell into the 30s where the non GSE eligible production was in the 60s, maybe even approaching 70.

The agencies have since, in the last several quarters, picked up dramatic market share. Our mix of business, the weighting has increased materially towards the agency side, where today -- between Fannie, Freddie, and Ginny eligible production -- it exceeds 50%.

In terms of the margins on that business, historically the margins on the agency business were lesser than the nonagency business; although in a stressed environment where spreads have widened, mostly in nonagency, I would say the agency margins are superior to the nonagency margins over the last month or so, as a result again of widening on the nonagency side.

But the agency business is affected as is the overall volume that we are originating by competitive pressures that are increasing and impacting that business as well. So margin trends are on the decline or under pressure, I would say.

Ken Bruce - Merrill Lynch

Do you see that generally speaking having any impact in terms of your prospective business, as it relates to just positioning in terms of the different products that you're focusing on? I mean, if you look back over the course of the last few years, there has been such a profit-driven increase in the origination of some of these nonprime assets or these alternative assets, that in many ways that is ultimately kind of creating its own demand.

As that market goes away because of the dislocation in the capitol markets, are you seeing an increasing interest in that GSE product that -- different maybe in terms of margins, but just in terms of stability is going to be a very different marketplace going forward?

David Sambol

Very much. One notable area where we are seeing a material shift in mix emerging is in the government side. We have seen over the last several months and quarters the FHA business growing. In fact it has doubled. While it's been very stagnant and really very immaterial, it is still not hugely material to us; it is 5% of our overall volume; but it was as low as 2%.

I think the FHA programs are growing in popularity with the decline in the declining offerings in the Alt-A and subprime market. So that is a trend that is notable to mention.

In terms of our posture, I think somebody mentioned that the best way to look at Countrywide is somewhat of a supermarket. We offer all products that are otherwise appropriately or legitimately offered in the marketplace; and we don't try to direct or divert people, necessarily, into one product versus the other. We just provide for a complete menu, except in educating our customers as to the pros and cons of the alternatives available to them.

Just market forces right now are such that business mix is shifting towards the agency products, government and conventional.

Angelo Mozilo

Let me just intercede here. Two things. One, Bob Napoli's question was not answered correctly. He asked for the ending shares at the end of the period outstanding; and we gave him the average number. We will get back to him on the ending number of shares.

In terms of the agencies, Fannie and Freddie would love to take this opportunity that they have, which is a unique opportunity, and play a much more compelling role in taking in more product. But they do have, despite the tremendous improvement in performance in both of those entities, they still are laboring under restrictions from their regulator. Therefore, even though they are willing and able to participate more in this market, take advantage of the situation, at least at this juncture, unable to do as much as they would like to do.

David Sambol

You know, Kevin Bartlett, our Chief Investment Officer, just handed me a note to convey that while I mentioned that slightly over 50% of our volume today or in Q2 was agency eligible, if you include -- that does not include agency eligible Alt-A. If you include agency eligible Alt-A, the number that we're delivering to the agencies is closer to 70%. It's 68%.

Ken Bruce - Merrill Lynch

Okay, thanks, that is closer to the number you had mentioned earlier this year. Maybe one other piece of color from Kevin. Is there a time frame that you think is going to be required to have the secondary market begin to calm down or otherwise distill what the S&P ratings changes are, and ultimately might suggest what this dislocation -- or at least in tenure what the duration of this dislocation in the capital markets may be?

Kevin Bartlett

Yes. Sure. I think the first thing that we need to have happen is that the rating agencies need to clear up the uncertainty that relates to how they are going to rate different loan programs going forward.

The one area where we have a little bit of certainty is on the subprime area. But the other products, I think they are still open. So we need them to clear up their uncertainty.

Then the next thing is I think once the new deals are rated by the agencies and they come into the market, I think the market is going to see that the greater level of protection afforded by their new levels will make the new securities more attractive. I think the market will come back for those securities, and probably become somewhat bifurcated from kind of post-rating agency changes and pre-rating agency changes.

So I think the market will probably separate and the new stuff will eventually find either the traditional buyers of credit, which I mentioned in my remarks earlier -- the banks, the insurance companies, money managers, et cetera. Then perhaps, we will see some reformed CDO machine that rehabilitates itself for the future. But at this point, the press on the CDO machine is not real positive.

Ken Bruce - Merrill Lynch

Thank you for all of your comments.

Operator

Your next question comes from the line of Samuel Crawford. Please go ahead.

Samuel Crawford - Citigroup

Thank you for taking the question. I wonder if I could get just a brief idea of how important fraud has been in the course of this quarter as a cause of loss relative to Q4 or Q1. Is it just as relevant? Is it dropping off precipitously? Is it increasing? What is happening?

Angelo Mozilo

Again, I think like all of these definitions, they are very broad. I think the primary issue has been an issue of speculation, rather than fraud. I am not saying there hasn't been some fraud in the traditional -- where people, just crooks, got involved in totally fraudulent transactions, straw buyers and that kind of thing. I think that appears to be de minimis.

It is really where people have given us information in terms of a variety of issues where really their intention was always to speculate on the properties. I think it is more of this speculation issue, which the builders try to address, the homebuilders in their contracts, which was deemed to be unconstitutional and they couldn't do that. But in these condos, particularly in the condo area there is a lot of speculation. I think that is where the -- I think history -- we will only know a couple years from now, but that seems to us where the problems are, more than the speculative area. Would you agree with that, David?

David Sambol

Yes.

Samuel Crawford - Citigroup

Okay, and you have…

Angelo Mozilo

Now obviously, in terms of your credit question, about has it slowed down, I would think it's probably nonexistent today because everything has tightened up so much, that everybody's antenna has been so sensitized to all the possibilities here, that it's pretty hard to get through the system now if you're not telling us the truth.

Samuel Crawford - Citigroup

All right. You all have tried to speak, I think, very clearly about the situation in regards to resets. But I am afraid I am still a little lost as to where Countrywide finds itself in the process of absorption.

If you think back to the portfolio as it stood at the beginning of 2006, and you look forward from that date, the historical portfolio; and you look forward to the resets this summer and the resets again in 2008, the two bunches, of those resets is it possible to give us an idea what portion of that original early 2006 portfolio has by now refinanced itself out of the way of danger?

Angelo Mozilo

Yes, we can, I think. Do you have that number, John?

John McMurray

Are you speaking of subprime or --?

Angelo Mozilo

No, no. Total.

Samuel Crawford - Citigroup

I was talking about total, but you know the rubber hits the road on subprime.

John McMurray

Typically, typically the behavior that we have observed over many years is that most borrowers do refinance prior to the reset. One of the worries that we have going into this environment is there may not be as many programs for the borrowers to refi into it as had been the case in the past.

But so far, going up to the point now, we still see a substantial number of the borrowers refi-ing prior to that point.

Angelo Mozilo

Do you have the percentage? We had a percentage at one time that I was using in terms of the 2006 book that had refinance. Do we have that number?

John McMurray

Not handy. You know, when we take a look at the remaining loans, in any given vintage, after the reset and within the six to 12 months after the reset, it's a very small percentage of the original balance. So the vast majority, the overwhelming majority of the original book will have refinanced out or otherwise paid off.

The exact number I don't have. Do we have factors, though?

David Sambol

We're looking.

Angelo Mozilo

Okay, we will give you that as soon as we find it.

Samuel Crawford - Citigroup

All right, I will come back around for that. One very last question if I may, please. When S&P took its most recent ratings action, some of the press releases that they put out in connection with that action, struck me that while they were not saying explicitly this, they were substantially saying this -- that they felt they gotten so far outside of the realm of normal experience with certain types of assets that their models simply were irrelevant. They had to go back and redo the whole thing.

I am just wondering if you felt like they did encounter -- was that your impression as well? -- they encountered that degree of model risk, and it shocked them so severely that they basically just started tearing up method and going back to rewrite?

Did you all encounter anything in your scoring or your advanced modeling which surprised you with equal drama in terms of unexpected model risk that appeared?

Angelo Mozilo

I think your S&P question, and not to skirt -- I don't really know the answer to that. Because I don't know what emotional trauma they went through or what their conclusions were and why they came to those conclusions that caused the actions they have taken.

I think the obvious -- it seems to me, the obvious conclusion you come to is that the models were not working the way they thought they were going to work; and they began the process adjusting those models so that they performed in accordance with what their expectations were. But that seems to me to be the case, but I don't have the insights into that, do you? Into what caused them to do that?

Kevin Bartlett

This is Kevin. I think that the rating agencies were looking at recent performance in the context of the housing decline. They're concerned about the -- I think you have seen this in the press and also directly in some their publications. They're concerned about the '06 to early '07 vintages of loans. I think when they saw the performance of those loans versus their models, they are trying to make appropriate adjustments to those models to reflect their new views.

David Sambol

As we have, by the way, and I think it is safe to say -- this is Dave Sambol -- that with credit having liberalized in the market the last several years, with higher LTVs and limited documentation and the combination of those things, they're really had not been for our models or for the rating agencies' models very much in the way of historical empiricals for that kind of product going through such a significant HPI decline. Home price decline, with which to accurately develop models.

As empirical data has now presented itself, I know in our case and now certainly based on what the rating agencies are saying in their case, everybody's models have or are being adjusted accordingly.

Angelo Mozilo

Including corporate debt. I mean as I think about it and reflect upon it, it's been common knowledge at least among observers that it appeared that the market in all debt categories, whether it be corporate debt or home finance debt, they were not pricing for risk. You can see all those adjustments are taking place now throughout the entire debt universe.

John McMurray

We can dimension the resets for you. So in the prime sector, 24% of the loans to be reset in 2007 and 35% scheduled to reset in 2008 are still currently active. That would compare to 30% and 38% at the end of Q1.

In subprime, 36% of subprime loans scheduled to reset in 2007 remain active. 60% scheduled to reset in 2008 remain active, and that compares with 37% and 66% at the end of the first quarter.

Samuel Crawford - Citigroup

That's very helpful. Thank you very much, John. Thank you all for your time.

Operator

Your next question comes from the line of Gary Gordon. Please go ahead.

Gary Gordon - PaineWebber

Okay, thank you. In light of your outlook for credit, I thought about your captive insurance, reinsurance business, and how you are treating that. Any particular changes in the business? Specifically, MGIC had about 45% of its loans insured in the second quarter, were 100% LTV. Does your reinsurance in Q2 have anything like that kind of 100% LTV proportion?

Angelo Mozilo

I think first of all, it is important to understand the level of risk we're taking. We're taking the mezzanine risk, so we don't have the first loss nor do we have the cat piece. We have the middle piece of it.

Now, in terms of your question about the type of loans that we have reinsured, do you know if you have that?

John McMurray

The type of loans that we reinsure will cover the whole spectrum, so it would include some of the 100% that you're talking about. Though I don't know whether it is as high as the figures that you quote for MGIC.

This high LTV business, along with the high CLTV business that we talked about earlier, is going to be subject to -- and in some cases has already been subject to -- the cutbacks that we described. I would anticipate the mortgage insurers tightening up their guidelines as well.

David Sambol

I would also add, by the way, that we only reinsure in our reinsurance entity primary mortgage insurance and not pool insurance of the type that MGIC and others write. While I am not certain, I would guess that if there was that high percentage of 100% LTV financing that they insured, it must have emanated in material part from pool insurance policies they wrote. Because 100% financing in the primary MI business would represent, I think, John, a small percentage, would it not?

John McMurray

I believe that is the case. The other important point, just to piggyback on Dave's pool insurance comment, many of the pool insurance policies are written -- are called modified pools. So they will be subject to a loan level stop loss. So they're still pool-style coverage, but oftentimes some of the insurers include that or report that, at least, as primary business. Or they have in the past.

Gary Gordon - PaineWebber

Okay, thanks. On subprime, just to check some numbers, the residual, the subprime residual, your residual balance grew by $31 million so far this year. I believe your write-downs are $256 million, so the math would say $287 million of new retained residuals. Does that number sound about right?

John McMurray

It sounds close. All of the increase between Q2 and Q1 relates to new deals. In fact, it is probably more than all of the increase between Q2 and Q1.

Gary Gordon - PaineWebber

Okay, and one last question for Angelo. You assume that the industry should see some consolidation or shrinkage in light of shrinking demand, tough competition. The industry is fairly concentrated at this point. The top 10 players I think are 70% of the business, which would argue that to really get a capacity shrink you would need to see some of these big players pull back in a material way. Is that your assumption?

Angelo Mozilo

Let me just break it down this way. I think that about 50% of the originations are from mortgage brokers or small mortgage bankers. I think clearly most of those, many of those, are going out of business today. They are fed into what you just talk about, the 10 top players.

But even if you look at the 10 top players, some of those there's been public announcements about problems among the top 10. That I believe what you're going to see over the next few years is a concentration down to five. They are the five major players today. At least that I -- again, I could be wrong on all of these things.

But I am just seeing continuous consolidation. Several years ago I said it would be about 10. It is about 10 now, and I think we will get to five for two reasons. One, it is increasingly becoming a more complicated business. It requires a lot of capital scale to make this thing work. So I think you can see further consolidation in the mortgage banking portion of the 10 players.

You have seen it in terms of Wachovia and World and Fleet and Bank of America and -- all big players at one time. I think nothing is out of the realm of possibility in this environment. So I think you're going to see further capacity shrink, even among the top 10 players.

Gary Gordon - PaineWebber

Okay, thanks.

David Sambol

Also, I think you will be seeing more of this from us in future disclosure when we talk about market share; but what we are going to seek to do is increasingly clarify the components of our market share and what is happening, I think, also in the market relative to consolidation. The way we look at market share internally is separately, when we look at our originated share, which is our wholesale and retail operations -- and that share, by the way, today for us is a number less than 10%, so a little bit above 9%. Earlier Angelo was talk about the retail-only share. That number is slightly above 5%.

So we look at our share in the market separately in what we described as originated share, which is wholesale and retail; and then purchased share, which represents our correspondent business and our capital markets conduit purchases and the bulk purchases that our Bank does. That is what gets our share up to the 18% that we quote in the aggregate.

But if you looked at the market and the share of the top 10 just on the basis of originated share, you would see that that number is still below 50% and not 70%, suggesting that there is a lot more consolidation in the origination market than what would be implied with a 70% number for the top 10 players.

And for us, by extension, a lot more opportunity than what might be apparent if you looked at the 20% number, considering that where we expect share growth for Countrywide is to come from the higher margin originated, or origination channels, excluding our purchased and lower margin business there.

Gary Gordon - PaineWebber

Okay, thanks a lot for that. Take care.

Angelo Mozilo

Thank you Gary.

Operator

Your next question comes from the line of Adam Weinrich.

Angelo Mozilo

Adam is out dancing. Can we go to the…

Operator

We'll move on to the line of Howard Shapiro. Please go ahead.

Howard Shapiro - Keefe, Bruyette & Woods

Hi. Thank you very much. I just wanted to ask two questions. The first is on the decision to keep all of the residual securities or all the bonds below the AAA. Given your share price today of -- it just passed below $30 a share, you have got to think that the IRR on retaining these assets is significantly higher than repurchase. So I'm wondering if you can share with us even approximately what you think the IRR on those investments would be? Then I do have one follow-up.

Kevin Bartlett

This is Kevin. I think first of all, I think, Howard, what we plan to do is we are reacting to the secondary market to the extent that there is still uncertainty posed by the rating agencies in their approach to new transactions.

We anticipate holding the below AAAs to the extent that they can't be distributed at attractive levels. I think if we look at the levels that we think that we would retain them at, we are going to see ROEs in the North -- probably North of the high teens to very high returns, depending on which particular bond class we are retaining.

So part of it is just in reaction to the secondary market. What we will do is we are going to adapt as the market returns in terms of the liquidity of those particular bond classes. We will distribute them when it makes sense to do so.

Howard Shapiro - Keefe, Bruyette & Woods

Okay, and just the other question, just to clarify. When you value your residuals, are you essentially using a mark-to-market methodology or a mark-to-model methodology? Could you just explain?

Kevin Bartlett

Well, it is a bit of a combination of both. The assets are carried at market value. The way we do that is we approximate it by using models that many in the industry would use as well. We use discount rate speeds and loss assumptions that we believe the market would employ.

Howard Shapiro - Keefe, Bruyette & Woods

Okay, so then given the lack of liquidity in the market today, how comfortable are you with your model?

Kevin Bartlett

I think, with the most recent loss in prepay estimates and discount rates, I think we are comfortable.

Howard Shapiro - Keefe, Bruyette & Woods

Okay, thank you very much.

Operator

Thank you. And next we'll go to the line of Jim Fowler. Please go ahead.

Jim Fowler - JMP Securities

Thank you. Two questions, please. Kevin, could you give me the subordination levels of the AAA across the broad mortgage classes that you will be retaining?

Kevin Bartlett

Yes, sure. It kind of depends on the product. You know, for Alt-A, depending on the -- like on the hybrid side, you are probably in the 7% area. In the fixed Alt-A, you are probably at, I would say, let's call it for the Alt-A stuff that is hybrids, probably in the mid 7s to low 7s; low to mid 7s.

For the fixed-rate stuff, you are probably around 7%. For the subprime, I believe we are probably in the 17% range.

In the HELOCs, it is -- we will probably hold that stuff in the HFI portfolio; so it's going to be the ca